CIGRE Green Books

International Council on Large

Electric Systems (CIGRE)
Study Committee B2: Overhead Lines

Overhead
Lines
CIGRE Green Books

Series editor
CIGRE, International Council on Large Electric Systems (CIGRE), Paris, France
CIGRE presents their expertise in unique reference books on electricalpower net-
works. These books are of a self-contained handbook character coveringthe entire
knowledge of the subject within power engineering. The books arecreated by
CIGRE experts within their study committees and are recognized bythe engineering
community as the top reference books in their fields.

More information about this series at http://www.springer.com/series/15209

Konstantin O. Papailiou
Editor

Overhead Lines

With 868 Figures and 175 Tables

Editor
Konstantin O. Papailiou
Malters, Switzerland

ISSN 2367-2625 ISSN 2367-2633 (electronic)

CIGRE Green Books

ISBN 978-3-319-31746-5 ISBN 978-3-319-31747-2 (eBook)

ISBN 978-3-319-31748-9 (print and electronic bundle)
DOI 10.1007/978-3-319-31747-2

Library of Congress Control Number: 2016946971

© Springer International Publishing Switzerland 2017

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology
now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the
editors give a warranty, express or implied, with respect to the material contained herein or for any errors
or omissions that may have been made.

Printed on acid-free paper

This Springer imprint is published by Springer Nature

The registered company is Springer International Publishing Switzerland Pte Ltd.
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Message from the President

Dear Members of Cigré, dear reader,

Since its creation in 1921, Cigré follows the mission to be a platform for exch-
ange and elaboration of knowledge in the electric power system. Being a global and
nonprofit organization throughout the decades Cigré has been a key factor for gai-
ning unbiased knowledge and understanding of the principles of the electric net-
work in all aspects e.g. for material, equipment, control- and system questions.
Most important are the biannual sessions in Paris and many smaller events on nati-
onal and international level every year. Also numerous Cigré Working Groups gene-
rated and synthesized knowledge throughout the decades. At this time, 230 groups
are active involving more than 3500 experts spread over the globe. Their product is
mainly the so-called Technical Brochures which are invaluable publications as they
contain the collective experience and expertise of most prominent experts in the
field on an international basis. The information made available is used in many
ways, such as textbooks for students and/or information for specialists in the various
topics. Quite often Cigré Technical Brochures are used as a reference when stan-
dards are not available. Frequently they are the basis for IEC standardization. Major
strategic directions of Cigré’s activities are; best practice issues, as well as future
aspects of the power system and ecological topics.
Today Cigré’s publications are available in hard copy and in electronic form back
to the year 1968 to be searched in “e-cigre”. An invaluable amount of information
has been published in conference reports and more comprehensive working group
documents. However, it turns out that there is still a gap as a regular review of the
State of the Art in the various fields is missing, that would be useful for education
and/or as reference books for experts.
For that reason Cigré made the strategic decision to develop a series of reference
books for the various fields, the so-called CigréGreenBooks. The goal would be to
compile state of the art knowledge in the field in a comprehensive and well-rounded
manner. The books should be updated periodically by the Study Committee respon-
sible for the topic.
For the time being two Cigré Green Books are available. The first one elaborated
by SC B2 with the title “Overhead Lines” (in hand) and a second one titled “Acces-
sories for HV Underground Cables” by SC B1. Both books compile a unique state
of the art review. Basic principles as well as important data for experts can be found
easily as the books are well structured. History, material, technology and system

v
vi Message from the President

aspects are well covered. The books are an essential enrichment of the Cigré publi-
cation portfolio. More books in other fields will follow.
The authors and contributors of the two books (“Overhead Lines” in hand) have
to be congratulated for this first successful edition and their extreme efforts are sin-
cerely appreciated. The Technical Committee also has to be acknowledged as it took
patronage.
In particular I like to express my sincerest thanks to Konstantin Papailiou, Chair-
man of Study Committee B2 (Overhead Lines) and to Pierre Argaut, Chairman of
Study Committee B1 (Insulated Cables) for their unflagging efforts to make both
books a success and having finished them in time for the Paris Session 2014. Kons-
tantin deserves particular credits as he was the one who offered the idea for a Refe-
rence Book in the first place to the Technical Committee.
Klaus Fröhlich
President of Cigré

Klaus Fröhlich received a Ph.D. in Technical Scien-

ces from the University of Technology in Vienna, Aus-
tria. For more than 11 years he worked for ABB
Switzerland and USA in development of high voltage
equipment. From 1990 till 1997 he was employed as a
full Professor for Switchgear and High Voltage Tech-
nology at the University of Technology in Vienna,
Austria followed by a full Professorship for High Vol-
tage Technology at the Swiss Federal Institute of Tech-
nology (ETH) in Zurich, Switzerland. Klaus Fröhlich
has a Fellow Membership in Electrosuisse and IEEE. He also is a member of the
Swiss Academy of Engineering Sciences. In Cigré his latest positions were Chair-
man of Study Committee A3 and chairman of the Cigré Technical Committee. Cur-
rently Klaus Fröhlich is the president of Cigré.
Message from the Chairman of the Technical
Committee

Efficient use of electric energy is at the very heart of a sustainable future for us all
and for almost 100 years now, Cigré has provided a worldwide platform for achie-
ving such an ambitious target.
Initially, as integrated, high voltage, electric power networks were developed and
became established in various parts of the world, Cigré was very much focussed on
the technical aspects of transmission of electric energy. As the electric power indus-
try evolved it was vital that Cigré also evolved. Over time, greater focus was placed
on aspects such as markets, regulation, system planning, sustainability and informa-
tion systems but this was certainly not at the expense of the more fundamental
technical aspects.
Today, as the distinctions between transmission and distribution and between
end user and electricity provider are eroded and as the entire electric power system
becomes more interactive and reliant upon intelligent systems, Cigré’s focus has, of
course, widened to address the entire electric power system. Generation, transmis-
sion, distribution and end-use of electric energy are all addressed across the entire
spectrum from 1200 kV transmission grids to local micro-grids, employing AC or
DC.
The present day activities of Cigré can be divided into three key themes namely:
“Developing the power system of the future”, “Making best use of existing power
systems” and “Environment and sustainability”. Within this framework Cigré stri-
ves to bring together the widest possible range of experts from across the world to
share and exchange knowledge and to use this combined knowledge and experience
to develop and publish pre-eminent technical information and state of the art gui-
dance.
Our aim is to prepare documents and communications that are clear, readily
accessible, unambiguous and appropriate to the intended audience, and which also
promote the value and importance of electrical engineering and the electric power
industry within technical, political, business and academic arenas. This has been
achieved very successfully over many years and Cigré’s ever growing library of
Technical Brochures, conference papers, tutorials and articles is a unique and unpa-
ralleled resource in the electric power industry. Nevertheless, recognising that dis-
semination of high quality, unbiased information is Cigré’s singular focus, finding
new ways to make our work visible is always a priority, which brings us to the Cigré
Green Book initiative.

vii
viii Message from the Chairman of the Technical Committee

Cigré Green Books are a way of consolidating, enhancing and disseminating

Cigré’s accumulated knowledge in specific fields. Addressing all aspects of Cigré’s
key themes, prepared and edited by world recognised experts, and building upon
Cigré’s established library of world class publications, the Green Books provide a
single, invaluable reference source within their specific field of application. They
also provide a unique resource for those wishing to develop themselves, for those
wanting to make their contribution to the power system of the future, and to the
vision of access to reliable, affordable and sustainable electric energy.
The Technical Committee is committed to the continuing development of Cigré’s
technical leadership in the electric power industry and the future expansion of the
Green Book series is a key part of this commitment.
Mark Waldron
Chairman of Technical
Committee of Cigré

Mark Waldron graduated in Electrical Engineering in

1988, joined the Research Division of the Central Elec-
tricity Generating Board and then, following privatisa-
tion, National Grid in the UK by whom he is still
employed. He has been involved in all aspects of
lifetime management of switchgear & substation
equipment from research & development, specifica-
tion, assessment, maintenance & monitoring, condition
assessment & end of life management. He presently
holds the position of Switchgear Technical Leader in
addition to his role as the Technical Committee Chair-
man of Cigré. His involvement in Cigré spans in excess
of 20 years during which he has been a participant in several Working Groups,
Working Group Convenor and Study Committee Chairman of Study Committee A3
and has led the Technical Committee project on Ultra High Voltage Transmission.
Message from the Secretary General

While these lines are being written, Cigré counts more than 7700 individual mem-
bers and 1100 collective members from 90 countries. All the members have access
to the publications produced by the Cigré Working Groups in the form of “Techni-
cal Brochures” when their work is completed.
Between 40 and 50 new Technical Brochures are published yearly. The brochu-
res are announced in Electra, Cigré’s magazine, and are available for downloading
from e-cigre, the online library of Cigré (www.e-cigre.org).
Over 6800 references of publications, from 1968, can be accessed from this
library, one of the most comprehensive accessible databases of relevant technical
work on power engineering.
From 1931 to 1974 the technical reports of Working Groups were published only
in Electra. As some brochures were becoming voluminous it was decided to publish
them separately, as Technical Brochures.
The first Technical Brochures were published in around 1974, and until 2000
Electra or separate Technical Brochures could be used to deliver the work of the
Working Groups, depending on the size of the document, 6 pages being the limit for
a publication in Electra.
In 2000, Electra was redesigned, and as a result, no longer published the final
reports of Working Groups. Today only summaries of Technical Brochures are pro-
vided in Electra, in both English and French.
From 2002 to 2014 some Study Committees have produced many Technical Bro-
chures: up to 75 for one of them, the average being 30 per Study Committee.
Therefore it is a good idea to organize over twenty years of accumulated
knowledge into comprehensive books.
Cigré Green Books are a new collection of publications, in a new format, a good
method to compile a large amount of knowledge, and the additional efforts of the
experts of the Study Committees involved in such projects should be recognized.
I am sure that the work involved will be appreciated by all the Cigré community.
Welcome to this new collection which I wish every success!
Philippe Adam
Secretary General of Cigré

ix
x Message from the Secretary General

Graduate of the Ecole Centrale de Paris, Philippe

Adam started his career in EDF in 1980 as a research
engineer in the field of HVDC and was involved in the
studies and tests of outstanding projects like the Cross
Channel 2000 MW link and the first multiterminal DC
link between Sardinia, Corsica and Italy. After this
pioneering period he managed the team of engineers in
charge of HVDC and FACTS studies of the R&D divi-
sion of EDF. In this period, his Cigré membership as a
working group expert and then as a working group con-
vener in Study Committee 14, was a genuine support to
his professional activities. Then he held several
management positions in EDF Generation and Transmission division in the fields of
substation engineering, network planning, transmission asset management and
international consulting until 2000. When RTE, the French TSO was created in
2000, he was appointed manager of the financial and management control depart-
ment, in order to install this corporate function and the necessary tools. In 2004 he
contributed to the creation of RTE international activities as project director first and
then deputy Head of the International relations department. From 2011 to 2014 he
has been the Strategy Director of Infrastructures and Technologies of the Medgrid
industrial initiative. In the meantime, between 2002 and 2012 he have served Cigré
as the Technical Committee Secretary and as the Secretary and Treasurer of the
French National Committee from 2009 till 2014. He was appointed Secretary Gene-
ral of Cigré in March 2014.
Preface

Exactly at the date this book is being presented for the first time during the 45th
Cigré Session, i.e. on August 24th 2014, in Paris, the first high voltage AC transmis-
sion in the world took place. It was actually on August 24th 1891 when one of the
main players of this memorable event shouted from the top of a wooden pole in the
line: “The current is now in Frankfurt!”. And indeed, for the first time in world his-
tory, electric power from the picturesque small town of Lauffen in Southern Ger-
many travelled more than 185 km to provide light to 3000 incandescent bulbs and
an artificial waterfall at the then electro-technical exhibition in Frankfurt/Main. This
was the beginning of -also in the physical sense- “a long” success story. From the
single 15 kV, and later 25 kV-AC line supported on wooden poles and with 4 mm
diameter Copper conductors (the losses were in the range of 25%!), there are
nowadays in Europe alone more than 100,000 km of 380 kV AC lines, fully inter-
connected! Electric power lines are thus most probably the longest and most com-
plex artifact mankind has ever conceived and created. Cigré, the Council on Large
Electric Systems (Conseil International des Grands Réseaux Electriques), was foun-
ded 30 years later in 1921 in Paris, and by its unique structure has been and is the
supreme reference for electric power networks. The Cigré Study Committee on
Overhead Lines, one of the oldest in Cigré, combines through its presently 25 Wor-
king Groups (WG) and some 300 experts from more than 40 countries, a massive
–and impressive- expertise in this field. This guarantees valuable exchange and dis-
semination of unbiased information for technical and increasingly for nontechnical
audiences, like the public, journalists and politicians.
To be frank, overhead lines, despite a practically worldwide constant demand for
new lines and the refurbishment of old lines, and despite all the interesting enginee-
ring but also management tasks related to them, have been considered by some as
becoming old-fashioned. What a misconception! Through the “Energiewende”, i.e.
the massive integration of renewables in the network, on one side and the huge
demand for electric power in the developing countries, China, India and Brazil are
good examples, on the other, overhead lines have recently become focus items for
utilities, investors, researchers, media and the public, for the last not always with a
positive attitude.
In this sense the idea of writing a reference book on Overhead Lines was born in
order to present to all interested parties the bundled knowledge and experience of
Cigré SC B2 in this exciting field. It was also considered as a unique opportunity to

xi
xii Preface

pass on valuable information and experience to the next generation of transmission

line engineers, as many of the contributors to this volume have been active for innu-
merable years in Cigré in this field and have accumulated a wealth of knowledge
which has to be preserved.
Preparing a book, and in particular such a big and complex one with many coau-
thors, is like coaching a world class soccer team for the editor, a parallel to the
World Cup 2014 that has just finished, i.e. it is not an easy task. The basic book
structure has to be defined, the content but also the focus agreed upon, overlaps to
be avoided. And many technical issues to be solved: the size of the book, the picture
on the cover, the fonts to be used, one- or two-columns, the style of the tables, the
numbering of the figures, etc., etc.
And it is like building a house: everything looks half-finished until the very last
minute. And you never believe that it will be completed on time. But it is extremely
rewarding. In particular when the book is nearing completion, the “happy-end”, this
is evident by the fact that the Editor starts preparing his Preface, as is the case now.
Because at that stage the hard work of so many people involved becomes visible.
I know it is a daring comparison, but when the first trial print is ready, it feels like
holding your baby in your hands for the first time.
But to continue in this tune, a book has its own life. And it is never finished.
Fortunately with the advances in electronic printing, oversights, omissions, sugge-
stions can be relatively easily incorporated in future editions, which I am confident,
will come. So please feel free to send your comments to [email protected]
and to [email protected] and they will be given thoughtful consideration.
But enough said, or better written: It is now time dear reader that you start
enjoying!

Athens/Greece, July 2014 Konstantin O. Papailiou

Konstantin O. Papailiou studied electrical engineering

at the Braunschweig University of Technology and civil
engineering at the University of Stuttgart. He received
his doctorate degree from the Swiss Federal Institute of
Technology (ETH) Zurich and his post doctoral qualifi-
cation as lecturer (Dr.-Ing. habil.) from the Technical
University of Dresden. Until his retirement at the end of
2011 he was CEO of the Pfisterer Group in Winterbach
(Germany), a company he has served for more than
25 years. He has held various honorary positions in
Technical Bodies and Standard Associations, being pre-
sently Chairman of the Cigré Study Committee “Over-
head Lines” (SC B2). He has published numerous papers in professional journals as
well as co-authored two reference books, the EPRI Transmission Line Reference
Book – “Wind -Induced Conductor Motion” and “Silicone Composite Insulators”. He
is also active in power engineering education, teaching Master’s level courses on
“High Voltage Transmission Lines” at the University of Stuttgart and the Technical
University of Dresden.
Preface of the Republished Edition

What a great coincidence! The very days the new CIGRE Green Book “Overhead
Lines” is to be presented during the bi-annual CIGRE session in the last week of
August, our industry – and I mean not only lines but the whole power system – will
celebrate the 125th anniversary of the first high voltage (15 kV to be precise) AC
transmission, which took place in the southern part of Germany in August 1891.
In 2014 I had the pleasure to write a few lines (lines again!) on the new series of
Reference Books CIGRE had started, the Overhead Lines Green Book and the HV
Cable Accessories Green Book. This has been an important step for CIGRE to make
available to the interested public valuable information collected over many years of
hard work in CIGRE working groups by a huge number of internationally-renowned
experts. The success of these books – hundreds have been sold since their introduc-
tion – proved that there is a demand for such information, which, because of its
plurality, is accurate and unbiased.
What better reason could there be to continue the series with subjects covered by
other CIGRE Study Committees, such as High Voltage Equipment, Insulated Cab-
les, Substations, HVDC and Power Electronics, Protection and Automation and
Distribution Systems and Dispersed Generation? And what better concept than for
CIGRE to liaise with a renowned international publisher like Springer for the
GREEN Book Series, which will be published by Springer as part of their so-called
Major Reference Works, a brand recognized worldwide for high quality content and
layout and enjoying high visibility in a large number of libraries, book fairs, and
internet platforms?
Another big advantage of this concept is that the Green Books will be available
not only in print but also as#2003(eBook)s, with all their advantages, such as avai-
lability on mobile devices, easily searchable, fully linked content. At the same time
Springer will establish a so-called living book platform for each Green Book, where
updates, corrections, digital media material, etc. can be readily uploaded and will
become immediately available to the public.
The first book which will be published under this collaboration is this Overhead
Lines Green Book. Within my editorials duties I had the opportunity during proof
reading to go through it again two years later and make some corrections, additions
and updates here and there. And I have to confess that, although I thought I knew a
few things about Overhead Lines (a wonder after 40 years – since 1976 – of being
active in CIGRE, another anniversary!), I have been more or less struck by the

xiii
xiv Preface of the Republished Edition

wealth of information this book contains and also the way this information is pre-
sented: clear and to the point; I dare say educational, so that also non-experts on the
subject will profit by reading it.
This gives me before closing the opportunity to thank once more all authors for
their valuable contributions, the CIGRE officials for their continuing support and
the Springer team for its great expertise and happily cheer:
Happy Birthday Overhead Lines, welcome new Green Book!

Malters/Switzerland Konstantin O. Papailiou

July 2016 Editor
Acknowledgements

A reference book like this one is always a big collective effort; and a hard piece of
work. Many individuals have been involved to accomplish it. First of all the chapter
lead authors and the chapter authors as indicated in the individual chapters, but also
the respective reviewers, all of them internationally recognized experts in their
fields. I would like to thank them all.
Luckily enough I could rely upon a competent team of book advisors, who made
valuable suggestions and have been always at my disposal every time need arose.
They are Bernard Dalle, former SC B2 Chairman, Normand Bell, former SC B2
Secretary and David Havard, former SC B2 WG11 (predecessor to TAG B2 06)
Convenor. In addition the last two offered their invaluable services by reviewing the
content of the whole book for consistency and overlaps and by performing a langu-
age check. All three of them deserve my gratitude and thanks.
This is also the time and the place to offer my thanks and gratitude to the Presi-
dent of Cigré, my good friend Prof. Klaus Fröhlich, who was immediately enthused
by the idea of a Green Book series (green evidently because of the Cigré-logo green
color) when first presented to him. Thanks and cheers also for my two dear col-
leagues, Mark Waldron, the Chairman of the Technical Committee of Cigré, and
Philippe Adam, the Secretary General of Cigré, as well as the other TC Chairmen.
All of them supported the idea and were always helpful in making it happen.
Because of extraordinary circumstances, this book (and its companion book on
“Accessories for HV Extruded Cables”) would not have been published on time for
the Cigré session 2014, if my beloved wife Margarita has not put all her energy and
skills to support me in the layout and production process. She and her publishing
team have done a great job which is herewith thankfully acknowledged.
Konstantin O. Papailiou
Editor

xv
xvi Acknowledgements

Normand Bell received his B.Sc.A. Degree in civil

engineering from Sherbrooke University (Québec) in
1974. He joined Hydro-Québec in 1974. He has
40 years experiences in the design, construction and
managing of overhead transmission lines and underg-
round lines. For the last 5 years, he has been providing
consultancy services to electrical utilities, consultancy
firms and developer for OHL projects. In Study Com-
mittee B2 (overhead lines), he has been a member and
secretary of working group, special reporter, advisory
group convener and was secretary of SCB2 from
2004–2010. He has published over 30 papers and been involved in the preparation
of Cigré TB 147, 265, 274 and 320.

Bernard Dalle is a Consultant on Overhead Lines and

a witness Expert at the Paris Court of Appeal. He has
been Chairman of Cigré Study Committee B2 Over-
head Lines for 6 years (2004–2010). He has worked as
a Senior Executive Consultant within RTE – Power
Transmission Infrastructures and as Director of Infra-
structure Grid R&D Programme within EDF/R&D. He
has also been Chairman of UF11, the Standardization
Committe on Overhead lines within UTE, the French
Organization for Standardization for electric and elec-
tronic products. He is a honorary member of Cigré and a member of SEE.

Dr. David (Dave) Havard, Ph.D., P.Eng. President of

Havard Engineering Inc., has over 50 years experience
solving the mechanical and civil engineering problems
of power delivery systems. His work involves analysis
of problems and finding solutions, particularly those
problems involving vibration, wear and fatigue. As a
Senior Research Engineer in the Mechanical Research
Department of Ontario Hydro, he coordinated Ontario
Hydro’s assessment of older transmission lines for the
province-wide refurbishment and upgrading, and has
worked closely with design and maintenance staff to
solve problems on vibration and galloping of overhead lines. Since establishing his
own company in 1992, Dr. Havard continues to provide engineering services to
utilities on control of vibration and galloping and testing and analysis of compo-
nents of transmission systems. David has been conducting open and in-house trai-
ning courses in these topics and degradation and upgrading of transmission lines for
utility staff. David is a long time active member of IEEE, CEA and particularly
Cigré. He served from 1987 to 1999 first as Secretary of the Task Force on Gallo-
ping, then Secretary and later Convenor of Cigré Study Committee B2, “Overhead
Acknowledgements xvii

Lines”, Working Group 11 “Mechanical Behaviour of Conductors and Fittings”,

and continues to be an active contributor. Under his leadership the working group
produced a number of technologically significant reports, ELECTRA papers and
technical brochures on overhead conductor vibration topics. Dr. Havard has autho-
red over 200 published papers and reports and is a Registered Professional Engineer
in the Province of Ontario.
Contents

Volume 1
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Konstantin O. Papailiou
2 History of Overhead Lines in Cigré . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Bernard Dalle
2.1 OHL Major Item of Discussion: 1880–1920 . . . . . . . . . . . . . . . . . . 19
2.2 The Creation of Cigré and its Development from 1921 to 1940
and the Role of OHL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.3 Reactivation of Cigré in 1948 and the Place of OHL in the
Evolution of Cigré Organisation: 1948–1966 . . . . . . . . . . . . . . . . . 22
2.4 OHL and Preferential Subjects from 1966 to the Present . . . . . . . . 23
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3 Planning and Management Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Rob Stephen
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.2 Management Concepts up to Commissioning . . . . . . . . . . . . . . . . . 28
3.2.1 Management Concepts for Preliminary Design
and Optimisation Studies . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.2.2 Management Concepts for Route Selection and
Property Acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.2.3 Management Concepts for Construction . . . . . . . . . . . . . . 30
3.3 Responsibilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.4 Life Cycle Process up to Commissioning . . . . . . . . . . . . . . . . . . . . 31
3.4.1 Planning Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.4.2 Route Selection and Property Acquisition . . . . . . . . . . . . . 33
3.4.3 Management Process for Preliminary Design and
Optisation Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3.4.4 Management Process for the Detailed Design Phase . . . . . 35
3.4.5 Project Execution (Construction) . . . . . . . . . . . . . . . . . . . . 38
3.5 Forms and Records (Including Accreditation) . . . . . . . . . . . . . . . . . 41
3.6 Summary of Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.7 Management of Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.7.1 Involvement at Design Stage. . . . . . . . . . . . . . . . . . . . . . . . 43

xix
xx Contents

3.7.2 Information Required and Handover (Submission) . . . . . . 43

3.7.3 Information for Maintenance during Operation . . . . . . . . . 44
3.8 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
3.9 Highlights. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
3.10 Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4 Electrical Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Joao Felix Nolasco, José Antonio Jardini, and Elilson Ribeiro
4.1 Electrical Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4.1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4.1.2 Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.1.3 Inductance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.1.4 Capacitance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
4.1.5 Negative and Zero Sequence Parameters. . . . . . . . . . . . . . . 50
4.1.6 Representation of Lines. . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.1.7 General Overhead Transmission Line Models . . . . . . . . . . 55
4.2 Surge Impedance and Surge Impedance Loading
(Natural Power) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
4.2.1 Methods for Increasing SIL of Overhead Lines . . . . . . . . . 69
4.2.2 Compact Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
4.2.3 Bundle Expansion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
4.3 Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
4.4 Thermal Limit and Voltage Drop . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
4.5 Capability of a Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
4.6 Reactive Power Compensation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
4.7 Electromagnetic Unbalance - Transposition . . . . . . . . . . . . . . . . . . 75
4.8 Losses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
4.8.1 Losses by Joule Heating Effect (RI2)
in the Conductors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
4.8.2 Dielectric Losses: Corona Losses, Insulator and
Hardware Losses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
4.8.3 Losses by Induced Currents . . . . . . . . . . . . . . . . . . . . . . . . 76
4.9 Reliability and Availability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
4.10 Overvoltages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
4.10.1 Fast-front Overvoltages (Lightning Overvoltages) . . . . . . . 77
4.10.2 Temporary (Sustained) Overvoltage . . . . . . . . . . . . . . . . . . 114
4.10.3 Slow-Front Overvoltages (Switching Surges). . . . . . . . . . . 116
4.11 Insulation Coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
4.11.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
4.11.2 Statistical Behavior of the Insulation . . . . . . . . . . . . . . . . 120
4.11.3 Insulation Coordination Procedure . . . . . . . . . . . . . . . . . . . 122
4.11.4 Withstand Capability of Self Restoring Insulation . . . . . . . 127
4.12 Electric and Magnetic Fields, Corona Effect . . . . . . . . . . . . . . . . . . 128
4.12.1 Corona Effects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
4.12.2 Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
Contents xxi

4.13 Overvoltages and Insulation Coordination . . . . . . . . . . . . . . . . . . . 143

4.13.1 Overvoltages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
4.13.2 Insulation Coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
4.14 Pole Spacing Determination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
4.14.1 Case of I-Strings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
4.14.2 Case of V-Strings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
4.15 Conductor Current Carrying Capability and Sags . . . . . . . . . . . . . . 159
4.16 Tower Height . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
4.17 Lightning Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
4.18 Right-of-Way Requirements for Insulation . . . . . . . . . . . . . . . . . . . 163
4.18.1 Line with I-Strings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
4.18.2 Line with V-Strings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
4.19 Corona effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
4.19.1 Conductor Surface Gradient and onset Gradient . . . . . . . . 165
4.19.2 Corona Loss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
4.19.3 Radio Interference and Audible Noise . . . . . . . . . . . . . . . . 171
4.20 Electric and Magnetic Field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
4.20.1 Ground-Level Electric Field and Ion Current . . . . . . . . . . . 174
4.20.2 Magnetic Field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
4.21 Hybrid Corridor or Tower . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
4.21.1 Conductor Surface Gradient . . . . . . . . . . . . . . . . . . . . . . . . 183
4.21.2 Radio Interference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
4.21.3 Audible Noise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
4.21.4 Corona Losses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
4.21.5 Electric and Magnetic Fields . . . . . . . . . . . . . . . . . . . . . . . 186
References 4.1–4.12 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
References 4.13–4.21 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
5 Structural and Mechanical Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
Elias Ghannoum
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
5.2 Deterministic and Reliability Based Design (RBD)
Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
5.2.1 Historical Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
5.2.2 The Need for Reliability Based Design in
Overhead Line Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
5.2.3 How RBD Methods Address the Deficiencies of
Deterministic Design Procedures . . . . . . . . . . . . . . . . . . . . 196
5.2.4 How to Apply IEC 60826 . . . . . . . . . . . . . . . . . . . . . . . . . . 197
5.2.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
5.2.6 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
5.3 Comparison of RBD Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
5.3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
5.3.2 Documents Compared and References . . . . . . . . . . . . . . . . 205
5.3.3 Basis of Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
5.3.4 Basic Design Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
xxii Contents

5.3.5 Combination of Loads. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207

5.3.6 Load Factors for Permanent and Variable Loads . . . . . . . . 207
5.3.7 Wind Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
5.3.8 Drag Coefficient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
5.3.9 Span Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
5.3.10 Ice Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
5.3.11 Combined Wind and Ice Loads . . . . . . . . . . . . . . . . . . . . . . 215
5.3.12 Failure and Containment Loads (Security Loads) . . . . . . . 216
5.3.13 Construction and Maintenance Loads (Safety Loads) . . . . 216
5.3.14 Other Loads. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
5.3.15 General Comparative Overview . . . . . . . . . . . . . . . . . . . . . 217
5.3.16 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
5.4 Tower Top Geometry and Mid-span Clearances . . . . . . . . . . . . . . . 225
5.4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
5.4.2 Part 1: Existing National Practices . . . . . . . . . . . . . . . . . . . 226
5.4.3 Part 2: Swing Angles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
5.4.4 Part 3: Required Clearances . . . . . . . . . . . . . . . . . . . . . . . . 229
5.4.5 Part 4: Coordination of Conductor Positions and
Electrical Stresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230
5.4.6 Part 5: Application Example . . . . . . . . . . . . . . . . . . . . . . . . 231
5.4.7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
5.4.8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
5.5 Load Control Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
5.5.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
5.5.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
5.5.3 Classification of LCD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
5.5.4 Technical Data Related to Available LCD . . . . . . . . . . . . . 237
5.5.5 Specification for an Ideal LCD . . . . . . . . . . . . . . . . . . . . . . 248
5.6 Mechanical Security of Overhead Lines Containing
Cascading Failures and Mitigating their Effects . . . . . . . . . . . . . . . 254
5.6.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254
5.6.2 Exceptional Loads, Accidental Loads and
Mechanical Security of Overhead Lines . . . . . . . . . . . . . . . 255
5.6.3 Line Cascade or Multiple Support Failures? . . . . . . . . . . . 255
5.6.4 Learning from Recent Major Tower
Cascading Failures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258
5.6.5 Current Understanding of Dynamic Line Cascading . . . . . 258
5.6.6 Recent Developments in OHL Cascading Mitigation. . . . . 259
5.6.7 Security Design Criteria to Prevent OHL Cascades . . . . . . 261
5.6.8 Framework for Successful Design to
Limit Overhead Line Cascades . . . . . . . . . . . . . . . . . . . . . . 263
5.6.9 Conclusions and Recommendations for Future Action . . . 263
5.6.10 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
5.7 Influence of Design Parameters on Line Security . . . . . . . . . . . . . . 265
5.7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
5.7.2 The Need for Unbalanced Longitudinal Loads . . . . . . . . . . 266
Contents xxiii

5.7.3 Requirements of Standards, Design Codes, and

Utility Practices. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
5.7.4 Comparison of BCL and UIL with Weather
Load Cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
5.7.5 Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
5.7.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
6 Environmental Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
Cathal Ó Luain, Lionel Figueroa, and Paul Penserini
6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278
6.2 Environmental Procedures and Assessment - Guidelines . . . . . . . . 280
6.2.1 Strategic Environmental Assessment (SEA). . . . . . . . . . . . 280
6.2.2 Permit Procedures and Environmental Impact
Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
6.3 Environmental Impacts and Mitigations - Guidelines . . . . . . . . . . . 286
6.3.1 Visual Impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
6.3.2 Impact on Land Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290
6.3.3 Impact on Ecological Systems . . . . . . . . . . . . . . . . . . . . . . 291
6.3.4 Impact of Construction and Maintenance . . . . . . . . . . . . . . 293
6.3.5 Environmental Management Plans . . . . . . . . . . . . . . . . . . . 295
6.4 Fields, Corona and other Phenomena, Impacts and Mitigations . . . 295
6.4.1 Electric and Magnetic Fields at Extremely Low
Frequency (ELF-EMFs) . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
6.4.2 Electric Field at Extremely Low Frequency ELF- EF . . . . 296
6.4.3 Magnetic Field at Extremely Low Frequency ELF-MF . . . 298
6.4.4 Assessment of the Exposure to Magnetic Field
for Epidemiological Studies . . . . . . . . . . . . . . . . . . . . . . . . 301
6.4.5 DC-EF and Ion Current Phenomena . . . . . . . . . . . . . . . . . . 304
6.4.6 Corona . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 306
6.4.7 Radio and Television Interferences . . . . . . . . . . . . . . . . . . . 309
6.4.8 Atmospheric Chemistry (Ions and Ozone) . . . . . . . . . . . . . 312
6.4.9 Aeolian Noise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313
6.4.10 Conclusions/Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . 314
6.5 Concerns and Issues, Consultation Models for OHL Projects
and Stakeholder Engagement Strategies . . . . . . . . . . . . . . . . . . . . . 315
6.5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315
6.5.2 Concerns and Issues Facing Utilities -Guidelines . . . . . . . 315
6.5.3 Consultation Models for OHL Projects . . . . . . . . . . . . . . . 316
6.5.4 Stakeholder Engagement Strategies . . . . . . . . . . . . . . . . . . 321
6.6 Life Cycle Assessment (LCA) for OHLs. . . . . . . . . . . . . . . . . . . . . 326
6.6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326
6.6.2 LCA Development and Early Applications . . . . . . . . . . . . 327
6.6.3 Power System and Overhead Line LCA in Scandinavia . . 328
6.6.4 Comparison of LCA Software . . . . . . . . . . . . . . . . . . . . . . 329
6.6.5 LCA, Overview for OHL Components, Construction
and Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
xxiv Contents

6.6.6 LCA, Studies on OHL. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 330

6.6.7 Conclusions/Recommendations . . . . . . . . . . . . . . . . . . . . . 331
6.7 OHL and Sustainable Development . . . . . . . . . . . . . . . . . . . . . . . . . 335
6.8 Highlights. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 336
6.9 Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
7 Overhead Lines and Weather . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341
Svein Fikke
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342
7.2 Wind . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343
7.2.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343
7.2.2 Extratropical Cyclones (after Cigré TB 256) . . . . . . . . . . . 344
7.2.3 (Sub-)Tropical Wind Systems . . . . . . . . . . . . . . . . . . . . . . . 346
7.2.4 High Intensity Winds Connected to Thunderstorms . . . . . . 348
7.2.5 Special Wind Systems (after Cigré TB 256) . . . . . . . . . . . . 350
7.2.6 Topographical Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351
7.3 Atmospheric Icing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354
7.3.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354
7.3.2 Icing Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 357
7.3.3 Measuring Ice Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 360
7.3.4 Icing Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362
7.3.5 Identification of Wet Snow . . . . . . . . . . . . . . . . . . . . . . . . . 366
7.3.6 Application of Numerical Weather Prediction Models . . . . 366
7.4 Other Topics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 368
7.4.1 Combined Icing and Pollution . . . . . . . . . . . . . . . . . . . . . . 368
7.4.2 Effects from Changes in Global Climate . . . . . . . . . . . . . . 368
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372
8 Conductors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375
Dale Douglass, Mark Lancaster, and Koichi Yonezawa
8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378
8.2 Conductor Materials & Manufacturing . . . . . . . . . . . . . . . . . . . . . . 380
8.2.1 Wire Material Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . 381
8.3 Electrical & Mechanical Characteristics . . . . . . . . . . . . . . . . . . . . . 385
8.3.1 DC Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385
8.3.2 AC Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387
8.3.3 Proximity Effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 388
8.3.4 Inductance and Inductive Reactance . . . . . . . . . . . . . . . . . . 388
8.4 Limits on High Temperature Operation . . . . . . . . . . . . . . . . . . . . . . 391
8.4.1 Thermal Rating and High Temperature Limits
(Cigré TB 601) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391
8.4.2 Maintaining Electrical Clearances (Cigré TB 244) . . . . . . 392
8.4.3 Limiting Loss of Tensile Strength (Cigré TB 244) . . . . . . . 392
8.4.4 Avoiding Connector Failures . . . . . . . . . . . . . . . . . . . . . . . 394
8.5 Sag-Tension & Stress-strain Models . . . . . . . . . . . . . . . . . . . . . . . . 394
Contents xxv

8.5.1 The Catenary Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 395

8.5.2 Mechanical Coupling of Spans . . . . . . . . . . . . . . . . . . . . . . 397
8.5.3 Conductor Tension Limits. . . . . . . . . . . . . . . . . . . . . . . . . . 398
8.5.4 Conductor Elongation – Elastic, Plastic, and Thermal . . . . 398
8.5.5 Sag-tension Calculation Methods . . . . . . . . . . . . . . . . . . . . 398
8.5.6 Parameter Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399
8.5.7 Sag-Tension Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . 400
8.6 Special Purpose Conductors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 401
8.6.1 Conductors for Use with Maximum Temperature
<100 °C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402
8.6.2 Conductors for Operation at High Temperature
(>100 °C) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404
8.7 Selecting the “Right” Conductor . . . . . . . . . . . . . . . . . . . . . . . . . . . 411
8.7.1 Factors in Conductor Selection for New Lines . . . . . . . . . . 411
8.7.2 Replacement Conductor Selection for Existing Lines . . . . 412
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 414
9 Fittings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417
Pierre Van Dyke, Umberto Cosmai, and Christian Freismuth
9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 421
9.2 Production Processes and Technologies . . . . . . . . . . . . . . . . . . . . . 421
9.3 Surface Finishing and Corrosion Protection . . . . . . . . . . . . . . . . . . 422
9.3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 422
9.3.2 Grinding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423
9.3.3 Tumbling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423
9.3.4 Sand Blasting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423
9.3.5 Brush Finishing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 424
9.3.6 Corrosion Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 424
9.3.7 Surface Protection of Ferrous Materials – Galvanizing . . . 426
9.3.8 Stainless Steel Surface Finishing . . . . . . . . . . . . . . . . . . . . 427
9.3.9 Aluminium Surface Finishing . . . . . . . . . . . . . . . . . . . . . . . 428
9.3.10 Copper Surface Finishing . . . . . . . . . . . . . . . . . . . . . . . . . . 428
9.3.11 Rubber Surface Conditions . . . . . . . . . . . . . . . . . . . . . . . . . 429
9.4 Electrical Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 430
9.4.1 Corona and RIV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 430
9.4.2 Short Circuit Loading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 432
9.4.3 Electrical Contacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 435
9.5 Test on New Fittings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437
9.5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437
9.5.2 Type Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 439
9.5.3 Sample Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441
9.5.4 Routine Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441
9.6 Tests on Aged Fittings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 442
9.6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 442
9.6.2 String Hardware Evaluation Guidelines . . . . . . . . . . . . . . . 444
xxvi Contents

9.6.3 Conductor Fittings Guidelines . . . . . . . . . . . . . . . . . . . . . . 448

9.6.4 Guidelines for Sample Removal,
Packing and Labeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 452
9.7 Safe Handling of Fittings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453
9.7.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453
9.7.2 Installation of Spacers and Spacer Dampers . . . . . . . . . . . . 455
9.7.3 Installation of Vibration Dampers . . . . . . . . . . . . . . . . . . . . 457
9.7.4 Installation of Compression Fittings . . . . . . . . . . . . . . . . . . 458
9.7.5 Installation of Preformed Fittings . . . . . . . . . . . . . . . . . . . . 459
9.7.6 Installation of Other Fittings . . . . . . . . . . . . . . . . . . . . . . . . 460
9.7.7 Live Line Installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 460
9.8 Damages on Fittings in Service . . . . . . . . . . . . . . . . . . . . . . . . . . . . 461
9.9 Influence of Fittings Design on Other Components . . . . . . . . . . . . 462
9.10 Connection Types. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 464
9.10.1 Clevis-Eye Connection . . . . . . . . . . . . . . . . . . . . . . . . . . . . 465
9.10.2 Ball-Socket Connection . . . . . . . . . . . . . . . . . . . . . . . . . . . 466
9.10.3 Y-Connection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 467
9.10.4 Oval Connection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 467
9.10.5 Bolted Connection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 468
9.11 Clamping Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 468
9.11.1 General Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 468
9.11.2 Suspension Clamps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 469
9.11.3 Spacer and Spacer-Damper Clamps . . . . . . . . . . . . . . . . . . 477
9.11.4 Vibration Damper Clamps . . . . . . . . . . . . . . . . . . . . . . . . . 482
9.11.5 Other Fittings Clamps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 483
9.11.6 Termination (dead-end) Clamps . . . . . . . . . . . . . . . . . . . . . 484
9.11.7 Fatigue Failure at Suspension/Clamping Point . . . . . . . . . . 489
9.11.8 Wear and Abrasion at Clamping Point . . . . . . . . . . . . . . . . 489
9.11.9 Corrosion Damage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 492
9.12 Aeolian Vibration Dampers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 493
9.12.1 Type of Aeolian Vibration Dampers . . . . . . . . . . . . . . . . . . 493
9.12.2 Conductor Damage due to Failure of Damping
Mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 499
9.13 Spacers and Spacer Dampers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 500
9.13.1 Type of Spacers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 500
9.13.2 Materials Used in Spacers . . . . . . . . . . . . . . . . . . . . . . . . . . 502
9.13.3 Conductor Damage due to Failure of Damping
Mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 505
9.14 Aircraft Warning Markers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 505
9.14.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 505
9.14.2 Current Practices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 506
9.14.3 Visibility of AWMs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 506
9.14.4 Types of AWMs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 507
9.14.5 Installation, Inspection and Maintenance of AWMs . . . . . . 509
Contents xxvii

9.14.6 Problems with AWMs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 510

9.15 Joints and Fittings for Conductor Repair . . . . . . . . . . . . . . . . . . . . . 511
9.15.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 511
9.15.2 Failure of Joints (Cigré WG22.12 2002) . . . . . . . . . . . . . . 512
9.15.3 Replacement of Joints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 526
9.15.4 Types of Conductor Damage. . . . . . . . . . . . . . . . . . . . . . . . 529
9.15.5 Classification of Conductor Condition . . . . . . . . . . . . . . . . 531
9.15.6 Remedial Actions Used by Utilities . . . . . . . . . . . . . . . . . . 534
9.15.7 Conductor Repair using FFH Fittings . . . . . . . . . . . . . . . . . 536
9.15.8 Conductor Repair Using HCI Fittings . . . . . . . . . . . . . . . . 543
9.15.9 Other Types of Repair . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 553
9.16 Highlights. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 555
9.17 Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 556
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 556
10 Conductor Motions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 559
Umberto Cosmai, Pierre Van Dyke, Laura Mazzola,
and Jean-Louis Lillien
10.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 560
10.2 Symbols and Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 561
10.3 Wind-Induced Conductor Motions . . . . . . . . . . . . . . . . . . . . . . . . 562
10.3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 562
10.3.2 Aeolian vibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 563
10.3.3 Wake – Induced Oscillations: Subspan
Oscillations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 626
10.3.4 Conductor Galloping . . . . . . . . . . . . . . . . . . . . . . . . . . . . 642
10.3.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 668
10.4 Non Sustained Conductor Motions . . . . . . . . . . . . . . . . . . . . . . . . 669
10.4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 669
10.4.2 Short-circuit Forces in Power Lines . . . . . . . . . . . . . . . . 670
10.4.3 Corona Vibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 690
10.4.4 Bundled Conductor Rolling . . . . . . . . . . . . . . . . . . . . . . . 691
10.4.5 Ice and Snow Shedding . . . . . . . . . . . . . . . . . . . . . . . . . . 693
10.4.6 Wind Gust Response (Tunstall 1997) . . . . . . . . . . . . . . . 695
10.4.7 Earthquake . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 696
10.5 Highlights. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 696
10.5.1 Aeolian Vibration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 696
10.5.2 Wake Induced Oscillations . . . . . . . . . . . . . . . . . . . . . . . 697
10.5.3 Energy Balance Principle. . . . . . . . . . . . . . . . . . . . . . . . . 697
10.5.4 Galloping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 698
10.5.5 Fatigue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 698
10.5.6 Assessment of Vibration Severity . . . . . . . . . . . . . . . . . . 698
10.5.7 Other Conductor Motions . . . . . . . . . . . . . . . . . . . . . . . . 699
10.6 Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 700
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 701
xxviii Contents

Volume 2
11 Insulators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 713
Frank Schmuck, and Konstantin O. Papailiou
11.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 714
11.2 Composite Insulator Product Generations, the Demand
for and Status of Standarized Tests . . . . . . . . . . . . . . . . . . . . . . . . 716
11.3 Selected Contributions by Cigré for Insulators and
Insulator Sets, and in particular Composite Insulators . . . . . . . . . 721
11.3.1 Insulator Groups 22.03, 22.10, B2.03, B2.21 . . . . . . . . . 721
11.3.2 Material Groups D1.14, D1.27 . . . . . . . . . . . . . . . . . . . . 724
11.3.3 Task Force Groups of SC 33 (Power System
Co-ordination) dealing with Insulator Pollution
Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 725
11.3.4 Contributions by WG’s of SC C4 - System
Technical Performance - in Terms of Insulator
Selection and Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . 726
11.4 Cigré Publications Reflecting the Status Quo of Insulators,
in Particular Composite Insulators . . . . . . . . . . . . . . . . . . . . . . . . . 726
11.4.1 Information on Surveys, Reliability, Failures . . . . . . . . . 726
11.4.2 Field Evaluation and in-service Diagnostic Methods . . . 739
11.4.3 Material Components of Composite Insulators . . . . . . . . 750
11.4.4 A Selection of Topics on the Dimensioning
of polymeric Insulators and Insulator Sets . . . . . . . . . . . 775
11.5 Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 814
11.5.1 Technology of Manufacture . . . . . . . . . . . . . . . . . . . . . . . 814
11.5.2 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 814
11.5.3 Tests for Material and Insulator Selection . . . . . . . . . . . . 815
11.5.4 Material Development . . . . . . . . . . . . . . . . . . . . . . . . . . . 815
11.5.5 Insulator Diagnostic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 815
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 815
12 Supports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 825
João B.G.F. Silva, Andreas Fuchs, Georgel Gheorghita,
Jan P.M.van Tilburg, and Ruy C.R. Menezes
12.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 826
12.2 Types of Supports. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 827
12.2.1 Regarding Function in the Line . . . . . . . . . . . . . . . . . . . . 827
12.2.2 Number of Circuits/Phase Arrangements/Tower Top
Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 828
12.2.3 Structural Types, Structural Modeling . . . . . . . . . . . . . . . 828
12.2.4 Formats, Aspects, Shapes . . . . . . . . . . . . . . . . . . . . . . . . 829
12.2.5 Material Used. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 830
12.3 Design Loadings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 831
12.3.1 Design Philosophy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 831
Contents xxix

12.3.2 Loadings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 836

12.3.3 Static and Dynamic Loads . . . . . . . . . . . . . . . . . . . . . . . . 838
12.4 Structural Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 839
12.4.1 Structural systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 839
12.4.2 Structural Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 842
12.4.3 Structural Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 844
12.4.4 Advanced Tools and Techniques . . . . . . . . . . . . . . . . . . . 846
12.5 Calculation and Dimensioning. . . . . . . . . . . . . . . . . . . . . . . . . . . . 850
12.5.1 Materials and Standards . . . . . . . . . . . . . . . . . . . . . . . . . . 850
12.5.2 Lattice Towers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 853
12.5.3 Metallic Poles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 867
12.5.4 Concrete Poles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 868
12.5.5 Wooden Poles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 871
12.6 Detailing Drawings and Fabrication Process . . . . . . . . . . . . . . . . . 871
12.6.1 Lattice Supports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 871
12.6.2 Metallic Poles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 879
12.6.3 Concrete Poles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 880
12.6.4 Wooden Poles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 881
12.6.5 Fabrication Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 881
12.7 Prototype Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 883
12.7.1 Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 883
12.7.2 Normal Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 884
12.7.3 Destructive Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 886
12.7.4 Acceptance Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 886
12.8 Special Structures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 890
12.8.1 Guyed Supports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 890
12.8.2 Guyed Structure Types. . . . . . . . . . . . . . . . . . . . . . . . . . . 890
12.8.3 Supports for Direct Current Lines . . . . . . . . . . . . . . . . . . 895
12.8.4 Supports for Large Crossings . . . . . . . . . . . . . . . . . . . . . 895
12.9 Environmental Concerns & Aesthetic Supports . . . . . . . . . . . . . . 899
12.9.1 Environmental Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . 899
12.9.2 Innovative Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 900
12.9.3 Landscape Towers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 901
12.9.4 Overhead Line Supports into Artworks . . . . . . . . . . . . . . 910
12.9.5 Experiences around the World: Conclusions . . . . . . . . . . 912
12.10 Existing Lines & Tower Aging . . . . . . . . . . . . . . . . . . . . . . . . . . . 917
12.10.1 Asset Management/Grid service . . . . . . . . . . . . . . . . . . . 917
12.10.2 Assessment of Existing Supports. . . . . . . . . . . . . . . . . . . 917
12.10.3 Inspection Philosophies . . . . . . . . . . . . . . . . . . . . . . . . . . 925
12.10.4 Types and Causes of Defects/Industry
Repair Practices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 926
12.11 Highlights. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 927
12.12 Future of Overhead Line Supports . . . . . . . . . . . . . . . . . . . . . . . . . 928
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 929
xxx Contents

13 Foundations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 933
Neil R. Cuer
13.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 938
13.1.1 Reasons for the Failure . . . . . . . . . . . . . . . . . . . . . . . . . . 941
13.2 Health, Safety, Environmental Impacts and Quality
Assurance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 941
13.2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 941
13.2.2 Health and Safety: General . . . . . . . . . . . . . . . . . . . . . . . 942
13.2.3 Risk Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 943
13.2.4 Environmental Impact . . . . . . . . . . . . . . . . . . . . . . . . . . . 946
13.2.5 Quality Assurance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 949
13.2.6 Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 954
13.3 Foundation Design (Part 1): Design Concepts and
Applied Loadings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 954
13.3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 954
13.3.2 Basis of design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 954
13.3.3 Interdependency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 956
13.3.4 Static Loading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 957
13.3.5 Dynamic Loading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 959
13.3.6 Foundation Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 959
13.3.7 Ground Conditions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 968
13.4 Foundation Design (Part 2): Site Investigation . . . . . . . . . . . . . . . 969
13.4.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 969
13.4.2 Initial Appraisal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 971
13.4.3 In-depth Desk Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . 975
13.4.4 Ground Investigation Methods . . . . . . . . . . . . . . . . . . . . 977
13.4.5 Factual Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 981
13.4.6 Interpretive Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 982
13.4.7 Ongoing Geotechnical Assessment . . . . . . . . . . . . . . . . . 984
13.4.8 Geotechnical Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . 985
13.5 Foundation Design (Part 3): Geotechnical and Structural . . . . . . . 985
13.5.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 985
13.5.2 System Design Considerations . . . . . . . . . . . . . . . . . . . . 986
13.5.3 Foundation Design – Geotechnical
and Structural . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 986
13.5.4 Interaction with Installation Process . . . . . . . . . . . . . . . . 997
13.5.5 Calibration of Theoretical Foundation
Design Model. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1002
13.5.6 Foundation Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1005
13.5.7 New Developments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1006
13.5.8 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1008
13.6 Foundation Testing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1009
13.6.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1009
13.6.2 Full-Scale Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1010
13.6.3 Model Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1015
13.6.4 Testing Benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1016
Contents xxxi

13.7 Foundation Installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1017

13.7.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1017
13.7.2 Pre-site Activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1018
13.7.3 Foundation Installation Method Statement . . . . . . . . . . . 1018
13.7.4 Temporary Works. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1019
13.7.5 Foundation Excavation . . . . . . . . . . . . . . . . . . . . . . . . . . 1019
13.7.6 Drilled Shaft, Pile and Ground Anchor Installation . . . . 1021
13.7.7 Formwork. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1024
13.7.8 Stub and Bolt Setting Assemblies . . . . . . . . . . . . . . . . . . 1025
13.7.9 Concrete . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1026
13.7.10 Backfilling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1031
13.7.11 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1032
13.8 Foundation Refurbishment and Upgrading . . . . . . . . . . . . . . . . . . 1033
13.8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1033
13.8.2 Foundation Deterioration . . . . . . . . . . . . . . . . . . . . . . . . . 1034
13.8.3 Foundation Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . 1034
13.8.4 Foundation Refurbishment . . . . . . . . . . . . . . . . . . . . . . . 1038
13.8.5 Foundation Upgrading . . . . . . . . . . . . . . . . . . . . . . . . . . . 1039
13.9 Outlook for the Future . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1041
13.10 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1042
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1044
14 Overall Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1047
Rob Stephen
14.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1048
14.2 AC Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1049
14.2.1 Line Impedance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1049
14.2.2 AC Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1050
14.2.3 AC Inductance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1050
14.2.4 Determination Of C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1051
14.2.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1051
14.2.6 Corona Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1051
14.2.7 Lightning Performances. . . . . . . . . . . . . . . . . . . . . . . . . . 1053
14.2.8 Thermal Rating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1055
14.2.9 Environmental Constraints. . . . . . . . . . . . . . . . . . . . . . . . 1057
14.2.10 Mechanical Design Configurations . . . . . . . . . . . . . . . . . 1060
14.2.11 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1062
14.3 Optimisation of AC Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1062
14.3.1 Factors Relating to Conductor Choice . . . . . . . . . . . . . . . 1064
14.3.2 Steps Required in Optimisation . . . . . . . . . . . . . . . . . . . . 1067
14.4 Need for an Objective Measure . . . . . . . . . . . . . . . . . . . . . . . . . . . 1069
14.4.1 Combining Line Parameters . . . . . . . . . . . . . . . . . . . . . . 1069
14.4.2 Different Indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1070
14.4.3 Application of the Indicator . . . . . . . . . . . . . . . . . . . . . . . 1072
14.4.4 Analysis of Designs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1075
14.4.5 Inclusion of the Constructibilty and Reliability
Factors in the Indicator . . . . . . . . . . . . . . . . . . . . . . . . . . 1075
xxxii Contents

14.5 HVDC Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1078

14.5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1078
14.5.2 Load Flow Characteristics . . . . . . . . . . . . . . . . . . . . . . . . 1079
14.5.3 Calculation of DC Resistance . . . . . . . . . . . . . . . . . . . . . 1079
14.5.4 Construction of the Conductor. . . . . . . . . . . . . . . . . . . . . 1080
14.5.5 Corona Inception Gradient . . . . . . . . . . . . . . . . . . . . . . . 1080
14.5.6 Corona Power Loss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1080
14.5.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1081
14.5.8 Mechanical Considerations . . . . . . . . . . . . . . . . . . . . . . . 1081
14.5.9 Thermal Rating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1081
14.5.10 Other Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1082
14.5.11 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1082
14.6 Optimising HVDC Line Design. . . . . . . . . . . . . . . . . . . . . . . . . . . 1083
14.6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1083
14.6.2 Suggested Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1083
14.6.3 Proposed Optimisation of HVDC
Lines – Voltage Assumed . . . . . . . . . . . . . . . . . . . . . . . . 1084
14.6.4 Optimisation Process Voltage Variable . . . . . . . . . . . . . . 1085
14.6.5 Summary of Optimisation Process . . . . . . . . . . . . . . . . . 1085
14.6.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1086
14.7 HVDC Indicator for Objective Determination of
Line Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1086
14.8 Application of Indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1087
14.8.1 Application of Indicators for AC Lines . . . . . . . . . . . . . . 1087
14.8.2 Application of Indicator on HVDC Lines . . . . . . . . . . . . 1091
14.9 Component Cost of Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1093
14.9.1 Method Applied to General Costing of Lines . . . . . . . . . 1093
14.9.2 Questionnaire . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1095
14.9.3 Comparison with Previous Work . . . . . . . . . . . . . . . . . . . 1097
14.10 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1101
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1101
15 Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1103
Zibby Kieloch, João B.G.F. Silva, Mauro Gomes Baleeiro,
Mark Lancaster, Marcin Tuzim and Piotr Wojciechowski
15.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1104
15.2 Construction Surveys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1104
15.3 Right-Of-Way Clearing and Site Access . . . . . . . . . . . . . . . . . . . . 1105
15.4 Foundations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1108
15.4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1108
15.4.2 Excavation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1108
15.4.3 Concrete and Reinforcement Works . . . . . . . . . . . . . . . . 1110
15.4.4 Drilling and Blasting . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1112
15.4.5 Assembly and Setting of Foundations . . . . . . . . . . . . . . . 1112
15.4.6 Backfilling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1113
15.4.7 Foundation Installation Challenges . . . . . . . . . . . . . . . . . 1113
Contents xxxiii

15.4.8 Foundation Failures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1115

15.5 Structure Assembly and Erection. . . . . . . . . . . . . . . . . . . . . . . . . . 1117
15.5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1117
15.5.2 Installation Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . 1119
15.5.3 Bolt Tightening and Finishing . . . . . . . . . . . . . . . . . . . . . 1129
15.5.4 Erection Method Selection Criteria . . . . . . . . . . . . . . . . . 1131
15.6 Conductor Stringing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1132
15.6.1 Preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1132
15.6.2 Stringing Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1132
15.6.3 Tension Stringing Equipment and Setup . . . . . . . . . . . . . 1134
15.6.4 Conductor Sagging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1136
15.6.5 Offset Clipping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1138
15.6.6 Conductor Creep and Pre-stressing . . . . . . . . . . . . . . . . . 1139
15.6.7 Crossings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1140
15.6.8 Grounding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1140
15.7 Insulators, Hardware and Fittings . . . . . . . . . . . . . . . . . . . . . . . . . 1141
15.7.1 Insulators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1141
15.7.2 Conductor Hardware . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1142
15.7.3 Vibration Control Devices . . . . . . . . . . . . . . . . . . . . . . . . 1143
15.7.4 Warning Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1143
15.7.5 Conductor Fittings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1143
15.8 As-Built Inspection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1144
15.8.1 Needs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1144
15.8.2 Documentation Review . . . . . . . . . . . . . . . . . . . . . . . . . . 1145
15.8.3 Field Inspection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1145
15.9 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1146
15.10 Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1146
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1147
16 Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1151
André Leblond, and Keith E. Papailiou
16.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1152
16.2 Maintenance Strategy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1153
16.2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1153
16.2.2 Steps in Developing a Maintenance Strategy . . . . . . . . . 1154
16.2.3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1156
16.3 Condition Assessment of OHTL . . . . . . . . . . . . . . . . . . . . . . . . . . 1156
16.3.1 Conductor System Including Joints and Fittings . . . . . . . 1157
16.3.2 Insulators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1171
16.3.3 Supports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1174
16.3.4 Foundations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1180
16.4 Use of Carts for In-Span Maintenance Work . . . . . . . . . . . . . . . . . 1182
16.4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1182
16.4.2 Technical Considerations . . . . . . . . . . . . . . . . . . . . . . . . . 1182
16.4.3 Sources of Tensile Strength Loss with Time . . . . . . . . . . 1183
16.4.4 Alternate Methods and Criteria for Cart Use . . . . . . . . . . 1185
xxxiv Contents

16.5 Live Work Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1185

16.5.1 Why Consider . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1186
16.5.2 What Can Be Done . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1186
16.5.3 General Cost Comparisons . . . . . . . . . . . . . . . . . . . . . . . 1188
16.6 The Use of Robotics and New Maintenance Techniques . . . . . . . 1194
16.6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1194
16.6.2 Transmission Line Robotics . . . . . . . . . . . . . . . . . . . . . . 1194
16.7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1203
16.8 Highlights. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1204
16.8.1 Maintenance Strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . 1204
16.8.2 Condition Assessment of OHTL . . . . . . . . . . . . . . . . . . . 1204
16.8.3 Use of Carts for In-Span Maintenance Work. . . . . . . . . . 1205
16.8.4 Live Work Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . 1205
16.9 Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1206
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1206
17 Asset Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1209
Jarlath Doyle
17.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1209
17.2 Asset Management Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1210
17.3 Guideline for Overhead Line Asset Management . . . . . . . . . . . . . 1210
17.3.1 Net Present Value (NPV) of Annul Expenditure . . . . . . . 1211
17.3.2 Annual Expenditure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1213
17.3.3 OHTL Asset Management Process . . . . . . . . . . . . . . . . . 1216
17.4 Data Collection for Overhead Line Asset Management . . . . . . . . 1218
17.4.1 Consequences of Failures . . . . . . . . . . . . . . . . . . . . . . . . 1219
17.4.2 Failure Analysis Data Collection . . . . . . . . . . . . . . . . . . . 1219
17.5 Database Management for Overhead Line Asset Management . . . 1221
17.5.1 The Different Kinds of Data . . . . . . . . . . . . . . . . . . . . . . 1222
17.5.2 Storing and Extracting Data. . . . . . . . . . . . . . . . . . . . . . . 1222
17.5.3 Storing Reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1223
17.5.4 The Link with other Databases . . . . . . . . . . . . . . . . . . . . 1223
17.5.5 The Quality of the Data . . . . . . . . . . . . . . . . . . . . . . . . . . 1224
17.5.6 Conditions for Success . . . . . . . . . . . . . . . . . . . . . . . . . . 1224
17.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1224
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1225
18 Uprating and Upgrading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1227
Gary Brennan, Zibby Kieloch, and Jan Lundquist
18.1 Introduction and Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1228
18.2 Purpose. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1228
18.3 General Economic and Technical Considerations . . . . . . . . . . . . . 1229
18.3.1 Increasing System Capacity . . . . . . . . . . . . . . . . . . . . . . . 1229
18.3.2 Optimum Time for Renewal
(Cigré TB 294 2006; Cigré TB 353 2008). . . . . . . . . . . . 1233
18.3.3 Planning Horizon and Net Present Value. . . . . . . . . . . . . 1236
Contents xxxv

18.3.4 Cost-Benefit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1237

18.3.5 Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1237
18.3.6 Constraints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1238
18.3.7 Terminal Equipment Considerations . . . . . . . . . . . . . . . . 1239
18.3.8 Electric and Magnetic Fields . . . . . . . . . . . . . . . . . . . . . . 1240
18.4 Overhead Line Uprating. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1240
18.4.1 Increasing Thermal Rating (Cigré TB 353 2008) . . . . . . 1241
18.4.2 Increasing Voltage Rating (Cigré TB 353 2008) . . . . . . . 1260
18.4.3 AC to DC Overhead Line Conversion
(Cigré TB 583 2014) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1267
18.5 Overhead Line Upgrading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1275
18.5.1 Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1276
18.5.2 Foundations (Cigré TB 141 1999;
Cigré TB 308 2006) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1279
18.5.3 Insulator Strings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1283
18.5.4 Upgrading or Improving Electrical Characteristics . . . . . 1286
18.6 Highlights. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1295
18.7 Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1295
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1295

19 Overhead Lines and Underground Cables . . . . . . . . . . . . . . . . . . . . . . 1299

Herbert Lugschitz
19.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1300
19.1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1300
19.1.2 Technical Basics and Differences
Between OHL and UGC . . . . . . . . . . . . . . . . . . . . . . . . . 1301
19.2 Advantages and Disadvantages of both Techniques . . . . . . . . . . . 1302
19.2.1 Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1303
19.2.2 Reliability and Repair Time . . . . . . . . . . . . . . . . . . . . . . . 1304
19.2.3 Lifetime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1305
19.3 Operational Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1305
19.4 New Techniques (Superconducting Cables, Gas Insulated
Lines “GIL”, High Temperature Conductors for OHL,
AC to DC, New Tower Design, DC with VCS) . . . . . . . . . . . . . . . 1306
19.4.1 UGC: Superconducting Cables . . . . . . . . . . . . . . . . . . . . 1306
19.4.2 Gas Insulated Line “GIL” . . . . . . . . . . . . . . . . . . . . . . . . 1307
19.4.3 OHL: High Temperature Conductors . . . . . . . . . . . . . . . 1307
19.4.4 OHL: Conversion AC to DC . . . . . . . . . . . . . . . . . . . . . . 1307
19.4.5 OHL: New Tower Design . . . . . . . . . . . . . . . . . . . . . . . . 1308
19.4.6 UGC: DC with Voltage Source Converters . . . . . . . . . . . 1309
19.5 Mitigation Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1309
19.5.1 Visual Impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1309
19.5.2 Electric and Magnetic Fields (EMF) . . . . . . . . . . . . . . . . 1311
19.5.3 Audible Noise, Induced Voltages, Impact on
Other Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1311
xxxvi Contents

19.6 Public Debate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1312

19.7 Main Applications of UGC, Technical Challenges . . . . . . . . . . . . 1313
19.8 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1316
19.9 Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1316
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1317
Introduction
1
Konstantin O. Papailiou

Overhead Lines look back to a long history from the first AC transmission in 1891
with 15 kV, Figure 1.1, to the 1000 and 1200 kV AC lines of today, Figure 1.2. This
unique and this exciting development has been tracked and supported by Cigré in
general and the Study Committee for Overhead Lines in particular. Chapter 2 “History
of Overhead Lines in Cigré” describes in detail this successful “partnership”. From its
very beginning in autumn 1921 Cigré aimed to provide a forum for technical studies
on the generation, transmission and distribution of electric energy. In this sense Cigré
brought together, on one hand, equipment manufacturers and operators of power
plants and transmission lines and electric energy producers, consultant engineers and
engineers of major public administration bodies on the other. In addition 1931, coin-
cidentally the year the Study Committee of Overhead Lines was established, Cigré put
in place a monthly journal named “Electra”, which is still the printed (today also the
electronic) voice of the association. It is at the least remarkable, that in 1946, Cigré
was the first technical organization in the world which organized an international con-
ference and brought together worldwide experts who had the enormous task to rebuild
electrical infrastructure after the perils of the second world war. In 2000 as another
landmark and as a key element in the new approach to communication, the “Conference
des Grands Réseaux Électriques à haute tension” became the “Conseil International
des Grands Réseaux Électriques”, that it changed from a “Conference” to a “Council”
and its scope was extended also to lower voltages. At the same time the areas covered
by Cigré’s field of action were redefined. Cigré now covers not only conventional
technical expertise but also economic and environmental aspects.

Originally published by Cigré, 2014, under the ISBN 978-2-85873-284-5. Republished by Springer
International Publishing Switzerland with kind permission.

K.O. Papailiou (*)

Malters, Switzerland
e-mail: [email protected]

© Springer International Publishing Switzerland 2017 1

K.O. Papailiou (ed.), Overhead Lines, CIGRE Green Books,
DOI 10.1007/978-3-319-31747-2_1
2 K.O. Papailiou

Figure 1.1 Wood pole of

the first 15 kV AC line
between Lauffen/Neckar to
Frankfurt/Main (1891).

An Overhead Transmission Line is a very complex structure often spanning

thousands of kilometers, crossing different weather zones and being subject to
huge electrical, mechanical and environmental stresses. Because of this it is very
important that in the management of a transmission line from conception to
decommissioning to realize the nature of the line as a device (or system) and to
ensure management structures do not compromise any aspect of the life cycle.
Chapter 3 “Planning and Management Processes” covers the various management
concepts that should be employed as well as the processes for line design, const-
ruction and maintenance.
The following Chapters of the book cover many of these processes in detail. For
instance Chapter 4 on “Electrical Design” contains all the basic information needed for
the electrical design of a transmission line. Subjects covered include such diverse
topics such as Surge Impedance and Surge Impedance Loading, Insulation Coordination,
Electric and Magnetic Fields and also Electric Parameters of DC Lines which show the
breadth of knowledge required for the proper design of a line. Specific emphasis is
given to the importance of the natural power as a key design factor. Especially nowadays
with the advent of HTLS (High Temperature Low Sag) conductors it is worth to
remember that their higher thermal capacity cannot be utilized except for relative short
lines, Figure 1.3. Also the aspects of proper grounding of transmission line towers,
eminently important for personal safety, as well as the issues of insulation coordination
and Corona are given the place they deserve. Finally the increasing use of DC-lines for
long distance transmission is adequately and in detail addressed.
1 Introduction 3

Figure 1.2 1200 kV AC

line at Bina test station in
India (Photo: Alberto
Pigini).

1000

800 MVA Thermal Limit on Power Flow is independent of length

800
Line Maximum Power Flow - MVA

AC Power Flow limited by

voltage drop or stability
600 concerns (electrical)
uncompensated, long lines
Power Flow
limited by
400 thermal
short lines limited to 25% of
Thermal Rating
for 1000 km line
200

0
0 200 400 600 800 1000 1200
Line Length in Kilometers

Figure 1.3 300-400 kV Transmisison Line max Power Flow dependence on length.
4 K.O. Papailiou

Overhead Transmission Lines are insofar unique, as they encompass mechanical

design issues of equal importance, and research interest, as the electrical issues
above. “Structural and Mechanical Design” is thus the focus of Chapter 5. This
Chapter is an excellent example on the pursuing of new venues by Cigré and also on
the cooperation between Cigré and other leading international organizations, in this
case IEC, a strategy successfully pursued up to these days. The subject is reliability
based design (RBD), an issue started in Cigré more than 40 years ago under the aus-
pices of Study Committee 22, the predecessor of SC B2. Further technical develop-
ment was pioneered by a joint effort between Cigré and IEC leading to a milestone
in 1991 when IEC 8261 entitled “Loading and Strength of Overhead Lines” was
published, introducing reliability and probabilistic concepts for calculation of loa-
ding and strength requirements for overhead line components Figure 1.4, a major
improvement compared to deterministic methods and nowadays widely used world-
wide. A good part of Chapter 5 deals extensively with the calculation of meteorolo-
gical loads on line structures notably wind and ice loads, followed by an explicit
calculation or tower top and mid-span clearances, demonstrating nicely the already
mentioned multidisciplinary nature of a line, as the clearances are significant for the
proper electrical functioning of the line. The Chapter concludes with quite a specific
subject, i.e. the design and use of load control devices for containing possible cascade
failures and explains the influence of line parameters on line security.
Chapter 6 “Environmental Issues” is, in addition to providing very valuable infor-
mation on this nowadays very central subject, a good example of the interdisciplinary
way Cigré works, as it encompasses important input from other Study Committees,
notably SC C3 (System Environmental Performance). Overhead Lines and environ-
mental issues and their interaction have been under consideration within Cigré for
many years. In this sense the issues covered in this chapter range from permit

QT = ΦRRC Rmed

Qmean

RC

QT

R10%

Figure 1.4 IEC 60826 reliability based design philosophy.

1 Introduction 5

Figure 1.5 A double

circuit 110 kV line corridor
used as a car park in an
industrial area (Ireland).

procedures, environmental impact assessments and consultation methodologies for

Overhead Line projects to mitigation of environmental impacts be they visual, ecolo-
gical, on land use or of construction and maintenance. Extensive space is given to the
development of reduced visual impact designs and aesthetic designs, a subject further
expanded in Chapter 12 “Supports” and to multiple use of Overhead Line corridors,
Figure 1.5. Last but not least all issues associated with field effects inclusive of the
debates on EMF and mitigation measures are presented including Corona, Radio and
Television Interferences and ion current phenomena associated with DC.
For a better assessment of the meteorological loads acting on a transmission
line and presented in Chapter 4, it is important to understand the weather and its
interaction with a line. This is exactly the subject of Chapter 7 “Overhead Lines
and Weather”. Therein not only the basic mechanisms of the different weather
phenomena, as for example the creation of a cyclone, a thunderstorm or the diffe-
rent icing procedures, which can have detrimental effects on a line, Figure 1.6, are
well explained, but also the ever increasing, and worrying, influences to the
weather due to changes in global climate are adequately covered. The chapter
further includes information on the application of numerical weather prediction
models, a subject of increasing interest thanks to the recent advances in informa-
tion and data managing techniques.
6 K.O. Papailiou

Figure 1.6 Rime icing on

a 420 kV line in Norway,
1400 m above sea level.

Figure 1.7 ACCR conduc-

tor showing the stranded
metal composite core.

Aluminum- Composite
Zirconium Alloy core

Conductors are the only active component of a line, i.e. the one component invol-
ved in the transmission of electric power and their costs, including the associated
fittings, can reach up to 50% of the total line investment costs. Because inadver-
tently they generate Ohmic losses, their properties influence significantly also the
operating costs of a line. It is thus no wonder that many Cigré Working Groups have
over the years investigated them. The summary of their findings is given in Chapter 8
“Conductors”. This chapter focusses on the following three areas:

• Calculation of AC resistance
• Sag-tension calculations
• Conductors for operation at high temperature

Regarding the latter, it has been astonishing to follow, how many innovative con-
ductor concepts have been created in the last years, Figures 1.7 and 1.8, following
the need to increase the power transfer capacity of a line with minimal changes of
the structure of the line.
1 Introduction 7

Figure 1.8 ACCC

conductor showing its
carbon fiber thermoset
resin core.

Figure 1.9 Short circuit

test of spacer damper.

Line components to anchor, clamp, connect (join), damp and repair the conduc-
tor are collectively known as “Fittings”. They are extensively covered in Chapter 9.
The chapter starts with a detailed section on the various manufacturing and finis-
hing processes of such fittings to continue with the electrical stresses they have
often to endure, such as Corona, short circuits, Figure 1.9, and contact deteriora-
tion. The latter are of increasing interest due to the elevated conductor temperatu-
res applicable nowadays. Another important issue is the test methodology
developed within Study Committee B2 for testing “old” fittings, i.e. fittings which
have been for long time in service, in order to decide on their usability. This goes
hand in hand with the extensive coverage of repair methods and the related hard-
ware, also covered in this chapter.
Conductor vibrations have plagued transmission line engineers since the early
days of overhead line transmission. In particular when the copper conductors ori-
ginally used started being replaced in the twenties, because of economic conside-
rations, by ACSR (Aluminium Conductors Steel Reinforced), vibration damages
have been so heavy, that the use of the latter has been seriously questioned. The
8 K.O. Papailiou

Figure 1.10 Wear and failure of conductor strands due to spacer clamp loosening.

understanding of the vibration phenomenon and its detrimental effects on line con-
ductors, Figure 1.10, has been thus for innumerable years a focus area of studies
within Cigré and it is not exaggerated to state, that quite a number of seminal
papers and state of the art Technical Brochures have been produced by Cigré
experts and published for the first time by Cigré. It is thus no wonder that Chapter 10
“Conductor Motions” turned out to be the lengthiest of this book. The following
examples might illustrate Cigré’s legacy in this field: a) the concept of the lifetime
estimation of conductors undergoing Aeolian vibrations, Figure 1.11, b) the EDS
principle followed by the safe design tension method, Figure 1.12, and c) the
energy balance principle (EBP) for the calculation of the vibration activity. This
chapter evidently includes also details on other types of wind-induced motions
such as sub-span oscillations and galloping, valuable information on conductor
self-damping and external damping devices (e.g. Stockbridge dampers and spacer
dampers) as well as a section dedicated to mechanical effects of short circuit loads
on conductors and line hardware, a phenomenon occurring mainly at the physical
interface of lines and outdoor substations.
The following Chapter 11 “Insulators” is equivalent to Chapter 10 as far as ori-
ginality and value of published information by Cigré is concerned. As for conductor
vibrations, also in the field of insulators, Cigré has played a pioneering role over the
years. And doubtlessly, the fact that Composite Insulators, a relative new, for
Overhead Line time frames, technology has advanced tremendously in the last
decades is due largely to the work of Cigré. This Chapter covers surveys (a very
powerful and frequently used Cigré tool which takes advantage of the unique inter-
national character of the organization) on many topics, on screening tests on aged
porcelain and glass insulations, service experience of composite insulators), stan-
dardization issues (where Cigré Working Groups have prepared the basis for many
IEC Standards) and design recommendations (for instance for the proper Corona
and power arc protection of insulator strings), Figures 1.13 and 1.14.
For many years “Supports”, the subject of Chapter 12, were more or less equiva-
lent to steel lattice towers and Cigré Working Groups have delivered important con-
tributions on their proper design philosophy, static and dynamic loads, structural
modeling and analysis, calculations and dimensioning, including advanced tools
and techniques, materials and standards but also detailing, fabrication and prototype
testin, g not to mention types and causes of defects and industry repair practices.
The last years though, increased awareness to transmission line structures and
1 Introduction 9

Figure 1.11 S-N curves 180

(Woehler curves) for
N/mm2
individual wires and for
stranded conductors. 140
1. Safe border line
2. Aluminium based 120
conductors 3. Pure
aluminium individual 100
wires 4. Aluminium alloy
individual wires
80 4

60

3
40
σa

30

1
20

10
104 105 106 107 108
N

20

18 Special Application #1:

Zone
16
#2:
without any obstruction, summer time.
14
#3:
12 open grass of farmland with few trees, hedgerows
and other barriers; pririe, tundra.
Safe Design
10 Zone #4: Built-up with some trees and buildings, e.g.
No Damping residential suburbs; small towns; woodlands and
8

6 Safe
Design
4 Zone
Span-
2 End
Damping
0
0 500 1000 1500 2000 2500 3000
Η/w, (m)

Figure 1.12 Recommended safe design tension for single aluminium based conductors.
10 K.O. Papailiou

Composite insulator
Power arc protection ring
made of steel

Corona ring made of

aluminium

Figure 1.13 Coordination between Corona and power arc protection: correct (left), incorrect
(right).

∅ 105

∅ 45

Figure 1.14 Scale comparison between a porcelain longrod, a cap and pin and a composite
insulator.

environmental concerns by the public, have propelled the development of so called

landscape or aesthetic towers, some of them very unique, Figure 1.15. For these
innovative solutions Cigré has collected worldwide examples and presented them in
a very informative Technical Brochure (TB 416, “Innovative Solutions for Overhead
Line Supports”) in 2010. The chapter concludes with information on steel monopo-
les, often used for compact lines, Figure 1.16, concrete poles and wood poles, while
work on supports with high-tech materials like composites is just starting.
Another “classic” component of a line is “Foundations”, presented comprehensi-
vely in Chapter 13. Foundations are evidently –and in every sense of the word- very
“basic” for the reliability and the structural integrity of a line and thus they are
addressed with great care. Quite unique is the fact that for foundations, because of
the variable soil conditions, from rocks to swamps and to undisturbed mother soil,
over the length of a line, site investigations during the design phase of a line project
are absolutely necessary. This is also one of the three pillars of this chapter, the other
two being foundation design (basic theory, static and dynamic loadings, foundation
types, Figure 1.17, and their interaction with the surrounding ground) and the
post-mortem examination of failures with related risk assessment, Figure 1.18.
1 Introduction 11

Figure 1.15 Pylone

Raquette in west
Switzerland.

Figure 1.16 Visual comparison of 420 kV conventional and compact lines side by side carrying
the same power in Dubai/UAE.
12 K.O. Papailiou

a b c d e
G.L. G.L. G.L. G.L. G.L. G.L.

Upper External
Excavation Included Grillage Shear
pad if Pad Included
line for angle 45 Bearers key
required -70 deg angle
undercut
25 deg
Braced Tretrapod Single leg
Pad and Chimney Stepped block Pyramid and Chimney Shallow Pyramid and
Steel grillage
Chimney

Figure 1.17 Spread Footings Foundations.

Figure 1.18 Failure of a 500 kV suspension tower drilled shaft foundation.

Probably for the first time in transmission line literature the concept of
“Overall Line Design”, part of ongoing Cigré work, is given such a prominent
place as in Chapter 14. Therein one can find the complex, sometimes unexpected
interactions and interdependencies of the decisions taken during the design phase
of a line. For instance while phase spacing increase (as for example in compact
lines) is beneficial for the SIL (surge impedance loading) of a line and the mecha-
nical loads of the line supports and foundations, it has negative effects on Corona
and AN/RIV, as both increase, Figure 1.19. While this is understandable form
basic physical principles, other interdependencies are not so obvious. For
example by lowering the knee-point temperature of an ACSR conductor (this is
the temperature at which the slope of the sag-temperature curve changes), the
longer the sag relationship will follow the steel core, resulting in a higher tempe-
rature and lower sag condition. This is possible for conductors with annealed
aluminium wires and also with so-called Gap-conductors, both costing more than
1 Introduction 13

SIL Corona Mechanical loading Thermal rating

Phase spacing
decrease

Large Al area/cond
(less conductors)

Diameter bundle
increase

High steel content

Figure 1.19 Relationship between actions taken in line design and effect on SIL, Corona,
Mechanical loading and thermal rating.

Optimal Voltage
Yearly cost: line investment and losses, and station cost
300
3,000 km
1,500 km
250

200
750 km
Million U$/yr

150

±800 kV
100
±600 kV
±500 kV
50
Optimal Voltage
±300 kV
0
0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500 5,000 5,500 6,000
MW
Legend: Red → ±800 kV; green → ±600 kV; pink → ±500 kV; blue → ±300 kV

Figure 1.20 Optimal voltage as a function of converter station power and line length.

conventional conductors. This leads to the second part of this chapter, where a
methodology for line optimization is given by introducing a suitable composite
indicator for comparison of the different design options including cost aspects.
The methodology can be also applied to HVDC lines in continuation of previous
Cigré work, Figure 1.20).
Traditionally Cigré has dealt more with design and field experience issues
than with the construction of a line. This short-coming is thankfully rectified
with Chapter 15 “Construction”. Therein practical aspects of line construction
are addressed providing an concise picture of the most common construction
activities and installation techniques used in installation of an overhead trans-
mission line. Special attention is being paid to construction activities requiring
14 K.O. Papailiou

Figure 1.21 Trailer mounted payoffs and a truck mounted bullwheel for tension stringing.

either high degree of accuracy or those posing high safety risks to construction
crews. Items covered include line survey, Right-Of-Way (ROW) clearing and
site access, assembly and setting of foundations, structure assembly and erec-
tion including insulators, hardware and fittings, conductor stringing and as-built
inspection, Figure 1.21.
On the other hand “Maintenance”, as presented in Chapter 16, has attracted
over the years the interest for quite a few Cigré Working Groups, such as the joint
(with SC B3, “Substations”) Working Group on “Live work - a Management
Perspective” and the recently created Working Group on “The Use of Robotics in
Assessment and Maintenance of OHL”, an exciting new option in overhead line
maintenance. Proper and efficient maintenance starts with the formulation of a
maintenance strategy. This should include the prioritizing of the transmission
lines in the network, carrying out periodic inspections, setting-up a data base sto-
rage and retrieval systems (otherwise sooner or later the results of the inspections
would be lost) and the -management- decision, whether maintenance work should
be done in-house or outsourced and done on live or de-energized lines. The core
of maintenance work is the condition assessment of line components. The Chapter
continues with valuable information on such assessment for the line conductors
including joints and fittings, the insulators, the supports and the foundations.
Special attention is then given to the use of carts for in-span work, Figure 1.22,
such as the repair or replacement of spacers, dampers, aircraft warning markers,
followed by a detailed description of the costs and benefits of live line work ver-
sus de-energized methods with interesting examples of typical live work opera-
tions. The chapter closes with the use of Robotics based on input from an ongoing
Cigré working group, Figure 1.23.
With the advent of liberalization of the electricity markets in the last years, a
holistic view of overhead lines as an asset has become a reality and a necessity.
1 Introduction 15

Figure 1.22 Cart used on

a horizontal twin bundle.

Figure 1.23 Photo of the

LineROVer robot
inspecting a transmission
line.

The importance of “Asset Management” for electric utilities and network opera-
tors is well demonstrated in Chapter 17. In the past decisions on the manage-
ment of existing overhead transmission lines were frequently based on the
qualitative judgment of experienced individuals. This chapter quantifies such
decisions using risk management techniques and presents methodologies for
estimating costs and risks associated with various actions required for proper
management of an overhead transmission line asset such as adequate inspec-
tions, analysis of a database of the conditions of the transmission line compo-
nents, cost factors, safety and regulatory and environmental considerations.
Management actions to be considered include risk reduction, risk acceptance
and risk increase also well described in Cigré TB 175. The chapter concludes
with important information for establishing and updating of databases, as this
16 K.O. Papailiou

will lead to an improvement of transmission line availability and reliability and

give the owner a better insight as to the remaining life expectancy and future
operating costs of their transmission assets.
Liberalization and the opening of the markets has also led, together with the
tremendous increase in renewables in the last years, and in combination with serious
objections by the public to new transmission corridors, to the development of a new
field in overhead line techniques, the “Uprating and Upgrading” of Chapter 18. The
purpose of this chapter is to provide a general overview of the economic and techni-
cal considerations in order to facilitate decisions for uprating and upgrading of over-
head lines. Upgrading will increase the original structural strength and thus decrease
the probability of failure of a line. Uprating on the other hand will improve the
electrical characteristics and thus also increase the power transmitted over the line.
The latter is nowadays of particular importance and is basically accomplished by
increasing the thermal rating and/or the voltage rating of the line in question. Of
specific interest is the information provided for the world’s first AC/DC hybrid line,
which is going to be presented during this Cigré session (Paris 2014), Figure 1.24.
For the same reasons, i.e. increased demand for power transfer and at the same
time public resentments against new line corridors, has favored in recent years the
combination of “Overhead Lines and Underground Cables” in particular for higher
voltage levels. This is the theme of the concluding Chapter of this book. Its scope is
to give an overview and comparison between Overhead Lines (OHL) and

AC/DC pylon
or
AC pylon
Hybrid pylon

380-kV- 380-kV- 380-kV- ±380-kV

AC-System AC-System AC-System DC-System

R R R +

Conversion

S T S T S T k –

R S T R S T R S T R S T

110-kV- 110-kV- 110-kV- 110-kV-

AC-System AC-System AC-System AC-System

Figure 1.24 Conversion of a double circuit of a 380 kV AC line to a hybrid 380 kV AC/±400 kV
DC line.
1 Introduction 17

Pogliano Rho
Milanese

Figure 1.25 Example of a partial undergrounded line route (Italy).

Underground Cables (UGC) regarding techniques, costs, failure rates, operating

issues and life expectancy with the objective to provide a sound technical base for
discussions (which are unfortunately often controversial) and to present the newest
developments and outlooks for both technologies, Figure 1.25. It is a very happy
coincidence, that the other book of the Cigré Green Book Series, is the book prepa-
red by Study Committee B1 “Underground Cables” on “Accessories for HV
Extruded Cables”. In this way the coexistence of both important transmission tech-
nologies for bulk power transfer is very visibly documented, as well as the excellent
cooperation between the two related Study Committees.

Konstantin O. Papailiou studied electrical engineering at the

Braunschweig University of Technology and civil engineering
at the University of Stuttgart. He received his doctorate degree
from the Swiss Federal Institute of Technology (ETH) Zurich
and his post doctoral qualification as lecturer (Dr.-Ing. habil.)
from the Technical University of Dresden. Until his retirement
at the end of 2011 he was CEO of the Pfisterer Group in
Winterbach (Germany), a company he has served for more
than 25 years. He has held various honorary positions in
Technical Bodies and Standard Associations, being presently
Chairman of the Cigré Study Committee “Overhead Lines”
(SC B2). He has published numerous papers in professional
journals as well as coauthored two reference books, the EPRI
Transmission Line Reference Book - “Wind -Induced
Conductor Motion” and “Silicone Composite Insulators”. He is also active in power engineering
education, teaching Master’s level courses on “High Voltage Transmission Lines” at the University
of Stuttgart and the Technical University of Dresden.
History of Overhead Lines in Cigré
2
Bernard Dalle

Contents
2.1 OHL Major Item of Discussion: 1880–1920 ....................................................................... 19
2.2 The Creation of Cigré and its Development from 1921 to 1940
and the Role of OHL ........................................................................................................... 21
2.3 Reactivation of Cigré in 1948 and the Place of OHL in the Evolution of
Cigré Organisation: 1948–1966 .......................................................................................... 22
2.4 OHL and Preferential Subjects from 1966 to the Present ................................................... 23
References ................................................................................................................................... 24

2.1 OHL Major Item of Discussion: 1880–1920

After the electric-telegraph and the first arc lamp lighting systems, electric techno-
logy entered a new historical phase of outstanding acceleration in the last quarter of
the 19th century.
At the Vienna International Electric Exhibition in 1873, Hyppolyte Fontaine
(France) realized the possibilities offered by long distance transmission of electri-
city. First industrial-scale transmission of electrical power was developed on a long
distance transmission between Vizille and Grenoble in 1883 by Marcel Deprez for
railways. However, until 1883, despite all the efforts of Marcel Deprez, the efficien-
cies remained too low for commercial purposes. The use of transformers gave alter-
nating current an essential advantage.

Originally published by Cigré, 2014, under the ISBN 978-2-85873-284-5. Republished by Springer
International Publishing Switzerland with kind permission.

B. Dalle (*)
Paris, France
e-mail: [email protected]

© Springer International Publishing Switzerland 2017 19

K.O. Papailiou (ed.), Overhead Lines, CIGRE Green Books,
DOI 10.1007/978-3-319-31747-2_2
20 B. Dalle

The first experimental three phase AC transmission between Lauffen (near

Heilbronn) and Frankfurt in 1891 was made by two Swiss companies AEG and
Oerlikon Maschinen Fabrik Oerlikon (MFO). This project had a crucial impact on
the following history of electricity (a good result of this project was that Lord
Kelvin, as a chairman of Commission of the Niagara Power station 2 project, sug-
gested AC as solution for transmission of power from the Niagara Power Generation
Station– at the time the largest project in the world – 5000 HP).
The father of the Transmission project for the Frankfurt Fair was Oskar van Miller
from Munich, who later established the excellent Deutsches Museum in there. The
designer of the first AC 3-phase generator and oil transformer (550 V to 15 kV) was
Charles L. Brown, who in the same year, 1891, co-established with Walter Boveri the
Brown-Bovery Company (BBC), which merged in 1987 with Asea (to become
ABB). The designer of the overhead line as well as of the electrical pump (100 HP)
was Mihael Dolivo – Dobrowolski, from AEG. The line powered 1000 incandescent
bulbs and an artificial water fall at the Frankfurt fair site. OHL was a 15 kV, later
25 kV, 40 Hz system, on 3200 wooden poles with copper conductors 3 × 12,6 mm2,
for a distance of 175 km. The operating voltage was 15 kV, later 25 kV, and the effi-
ciency of transmission was 72.5 %, which is equivalent to 22.5 % losses! The prota-
gonists of this historical event can be seen in the nostalgic picture of Figure 2.1.
The voltage problem was very important in the history of OHL’s. Growth and of
voltages depends of the history of insulators. The first suspension insulators were
invented in 1907 enabled, in 1908, the first OHL with voltage over 100 kV. The
invention of cap-and-pin insulator in 1910 opened the possibility for higher voltage
overhead lines.
Figure 2.1 shows the development of voltage level in the world.

U/kV

1150 kV
750 kV
500 kV
1985
380 kV 1965
287 kV 1960
220 kV 1950
150 kV 1938
110 kV
1923
60 kV 1913
25 kV
1908
1896

1891 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990

Figure 2.1 Development of the OHL voltages.

2 History of Overhead Lines in Cigré 21

In this spectacular process of the expansion of electric energy in industrial civi-

lization, one must stress the decisive role of the International Exhibitions. In 1881,
the International Electric Exhibition had an enormous success and the Scientific
Congress organized in parallel brought together the greatest scientists of the time,
who adopted the first international system of electrical units. The impacts of the
Paris events were the impetus for the constitution of a new professional community,
and marked the foundations of the Societé Internationale des Electriciens.
Hence these large scale events opened the way for creation of specific International
bodies, and particularly with the creation of the Internatioanl Electrotechnical
Commission (IEC) in 1906 and later the foundations of Cigré (Conférence Inter-
nationale des Grands Réseaux Electriques in 1921, the World Energy Conference
(WEC) in 1924 and the International Union of Producers and Distributors of
Electricity (UNIPEDE) in 1925.

2.2 The Creation of Cigré and its Development from 1921

to 1940 and the Role of OHL

From 1911 to 1920, C.O. Mailloux, President of IEC and Charles Le Maistre, its
Secretary General recommended the formation of a specialized body of a technical,
scientific and applied technology character.
The contacts established in various countries confirmed these aims and led to the
organization of the first International Conference on Large Electric Systems (Cigré)
in PARIS in the autumn 1921. This meeting was considered as the first inaugural
meeting of the new international engineering organization.
Cigré aimed to provide an international setting for the discussion and the study
of technical questions concerning the generation, transmission and distribution of
electric energy.
Therefore, when it was founded, Cigré brought together, on one hand, manufac-
turers of electrical machines and equipment and operators of power plants and trans-
mission lines and on the other hand, electric energy producers, consultant engineers
and engineers of major public administration bodies.
In 1931 Cigré put in place an ambitious journal “Electra”, a monthly journal devo-
ted to the study of the generation, transmission and transformation of electric energy.
The rapid organization and rapid growth of Cigré’s Study Committees started in
1925: the first Study Committee was the Statistics Study Committee which lasted only
a short while having been transferred to UNIPEDE, founded in 1925. The Insulators
Committee was created in 1929, the Overhead Lines Committee in 1931 and the
Towers and Foundations Committee in 1935.
The main subjects dealt within the reports during the inter-war period were:

• Parallel operations of power plants and oscillation between machines,

• Problems in construction of large generators and transformers,
• Laws and electrical calculation of energy transmission, voltage adjustment and
reactive power,
22 B. Dalle

• Reliability of cables for high voltage,

• Insulation of lines, the nature and properties of insulators, and dielectric strength
of insulation,
• Earth connection of the neutral and extinction coils, and interference caused in
telecommunication circuits.

2.3 Reactivation of Cigré in 1948 and the Place of OHL

in the Evolution of Cigré Organisation: 1948–1966

During the Second World war, Cigré activities stopped naturally just after the pub-
lication of the main conclusions of the 1939 Session.
In 1946, Cigré was the first technical organization in the world which organized
an International Conference and brought together worldwide electrical experts to
rebuild electrical infrastructure after World War 2, and start with intensive
electrification.
In 1956, the fields corresponding to the preferential subjects devoted to over-
head lines were divided among no less than 4 Study Committees for the general
design dimensioning of lines: SC 22 for towers and foundations blocks, SC 23–24
for conductors, SC 25 for insulators, SC 40–42 for very high voltage lines (above
220 kV) and 3 “electrical” study committees: SC 33 for overvoltage and light-
ning, SC 35 for telephone and radio interference and SC 41 for “Insulation
Coordination”.
Regarding these Study Committees scopes, a key technical achievement was rea-
lized in 1962 by the Volgograd-Donbass (USSR) DC Transmission line +/−
400 kV. Around the same time, AC transmission lines were put in operation in the late
60s – 735 kV in Canada, 750 kV in the USSR and 765 kV in the USA. Until 1970,
groups 22, 23, 24, 25 and 33 alternated with groups 35, 41 and 40–42 and, as a result,
the different fields related with overhead lines were discussed only every four years.
The preferential subjects were then very specific to each of the components: insula-
tors, foundations, conductor bundles,… At the 1960 Session, this very specific nature
was very clearly visible through the subjects developed: insulators’ thermal shock
tests, withstand tests in a polluted atmosphere, heaving and corrosion of founda-
tions,… Yet, in 1964, the subjects became more general. People studied the lifetime
of structures in relation to the effects of vibrations, weathering and the associated
safety coefficient, conductor creep,… Lastly, at the 1966 Session, the reduction in the
costs of towers was one of the priority subjects.
In the late 1950s and early 1960s, Cigré’s main priorities for study were the
increase in the transmission capacities, mainly by a voltage increase, and the stabi-
lity of power systems which were becoming more and more interconnected. A recur-
rent topic was Corona radio interference, which made it necessary to oversize the
conductors in relation to the size required for an economic utilization of their ther-
mal capacities.
2 History of Overhead Lines in Cigré 23

2.4 OHL and Preferential Subjects from 1966 to the Present

The reorganization of Study Committees in 1966 led to a new configuration of the

field of overhead lines, with Study Committee 22, officially named “Overhead
Lines”, Study Committee 33 for insulation coordination, and for 36 for interference.
This new configuration made it possible to combine within a single Study Committee,
all the issues concerning design dimensioning of structures: foundations, towers,
multiple conductors in bundles and insulators.
From the early 1960s to the early 1980s, overhead transmission lines at 735, 750
and 765 kV were mastered in Canada, USSR, USA andBrazil. Thermal and electro-
dynamic problems also arose after the increase in levels of transmitted power
(effects of nominal, exceptional current intensities and of short-circuit currents).
For the first time, in 1974, consideration of the environment in the design of over-
head lines, and therefore their acceptability, was chosen as a preferential subject. In
the 1990s, the aesthetic of towers design was a major area of the Cigré community’s
work. Tubular towers, architectural towers and compact towers were the subject of
very interesting presentations and discussions.
In the late 1970s, the development of information technology resources became
a vital tool in the development of power systems: programs for optimization of
structures (optimization of towers and optimization of conductors) also appeared,
and they gave rise to regular discussions. The development of personal computers
led to the development of very efficient software applications, which are now new
standards in their respective fields.
In 1981 Cigré organize the first Symposium in Stockholm which was “Overhead
lines impacts on environment and vice versa”. From that time began a discussion on
how to improve the aesthetic view of OHL in landscapes, impacts of lines on EMC,
how to made new standards to mitigate impacts of nature on lines, etc.
Probabilistic approaches were discussed in 1988: probabilities of maximum cur-
rents, of maximum ambient temperatures, and low wind speeds,…
At the same time, optic fibers also appeared as one of the preferential subjects.
This was somewhat the starting point of the rapid development in the incorporation
of these new telecommunication links in the networks, in spite of the additional
constraints they induce on the design and the operation of the lines.
In 1990, since the oldest EHV lines had been in operation for 60 years, the lifetime
of structures started to be a concern for operators. It remained one of the recurring
subjects of the Sessions, with significant feedback on operation and maintenance
methods, considering the strategic stakes attached to these topics. In 1994, for the
first time, live line work aimed to improve the availability of structures, was tackled.
In 2000, as a key element in the new approach to communication, the “Conference
des grands Réseaux électriques à haute tension” became the “Conseil International
des Grands Réseaux Electriques”, that it changed from a “Conference” to a “Council”.
The area covered by Cigré’s field of action was redefined in 2000. It now covered not
only conventional technical expertise but also economic, environmental aspects and
the impact of aspects related to organization and to regulations.
24 B. Dalle

In 2002, a new organization was set up and SC 22 became SC B2. In 2003, a new
reference model for Study Committees was ratified by the Cigré’s governing bodies
and gradually applied. This reference model stressed the need for the proactive cha-
racter of the Study Committees, which had to absolutely avoid gradually becoming
artificially expanded and self-sustaining techno structures with organization and run-
ning costs far exceeding their technical output.
In the 2000s, studies and discussions in the area of overhead lines focused on the
increase in transmission capacities and on the ageing of structures: increases in
voltage, taking into account in real-time weather conditions, development of new
types of conductors (composite conductors with metal or composite fiber core). The
storms observed around year 2000 throughout the world led to renewed discussions
on the dimensioning of structures based on methods combining static and dynamic
loads in a probabilistic approach.
The considerable progression, in terms of general innovation, and of Geographic
Information Systems (GIS) was acknowledge in 2006. It was confirmed that they were
applied to all phases of the life cycle of a line: planning, design, construction, opera-
tion, management and control of vegetation, forest fires, inspection, maintenance,
rehabilitation and dismantling.
To increase the transmission capacities, different solutions were discussed: real-
time management of certain structures became possible as a result of the develop-
ment of sensors indication the real sag at the most sensitive point and notable gains
in transit capacities could then be envisaged, but this required specific organization
of the control actions. In some configurations, the transformation of alternating cur-
rent overhead lines into direct current overhead lines could increase transit capaci-
ties by more than 100 %, but with costly conversion stations.
Lifetime assessment and lifetime extension, precise knowledge on the equipment
condition, diagnostic methods, environmental and societal acceptability of over-
head lines were other topics in recent group meeting.
In 2016, Study Committee B2 “Overhead Lines”, counted 22 working groups with
371 experts from 43 countries.

References
Ishkin, V. Kh.: International Council on Large Electric Systems -Cigré History activity
1921 –2006 – Moscow 2006
Subic, S.: On transmission of active power by electricity. Home World (12) (1900). Catholic Press
Association
The History of Cigré – A Key Player in the Development of Electric Power Systems Since 1921.
Cigré (2011)
2 History of Overhead Lines in Cigré 25

Bernard Dalle is a Consultant on Overhead Lines and a witness

expert at the Paris Court of Appeal. He has been Chairman of
Cigré Study Committee B2 Overhead Lines for 6 years (2004–
2010). He has worked as a Senior Executive Consultant within
RTE – Power Transmission Infrastructures and as Director of
Infrastructure Grid R&D Programme within EDF /R&D. He has
also been Chairman of UF11, the Standardization Committe on
Overhead lines within UTE, the French Organization for Stan-
dardization for electric and electronic products. He is a honorary
member of Cigré and a member of SEE.
Planning and Management Concepts
3
Rob Stephen

Contents
3.1 Introduction .......................................................................................................................... 28
3.2 Management Concepts up to Commissioning .................................................................... 28
3.2.1 Management Concepts for Preliminary Design and
Optimisation Studies ................................................................................................ 28
3.2.2 Management Concepts for Route Selection and Property
Acquisition ............................................................................................................... 29
3.2.3 Management Concepts for Construction ................................................................. 30
3.3 Responsibilities ................................................................................................................... 30
3.4 Life Cycle Process up to Commissioning ........................................................................... 31
3.4.1 Planning Requirements ........................................................................................... 31
3.4.2 Route Selection and Property Acquisition .............................................................. 33
3.4.3 Management Process for Preliminary Design and
Optimisation Studies ................................................................................................ 34
3.4.4 Management Process for the Detailed Design Phase .............................................. 35
3.4.5 Project Execution (Construction) ............................................................................ 38
3.5 Forms and Records (Including Accreditation) .................................................................... 41
3.6 Summary of Process ........................................................................................................... 41
3.7 Management of Maintenance .............................................................................................. 43
3.7.1 Involvement at Design Stage ................................................................................... 43
3.7.2 Information Required and Handover (Submission) ................................................ 43
3.7.3 Information for Maintenance during Operation ...................................................... 44
3.8 Conclusion .......................................................................................................................... 44
3.9 Highlights ............................................................................................................................ 44
3.10 Outlook ............................................................................................................................... 45
References ................................................................................................................................... 45

Originally published by Cigré, 2014, under the ISBN 978-2-85873-284-5. Republished by

Springer International Publishing Switzerland with kind permission.
R. Stephen (*)
Durban, South Africa
e-mail: [email protected]

© Springer International Publishing Switzerland 2017 27

K.O. Papailiou (ed.), Overhead Lines, CIGRE Green Books,
DOI 10.1007/978-3-319-31747-2_3
28 R. Stephen

3.1 Introduction

In the management of a Transmission line from conception to decommissioning it

is important to realise the nature of the line as a device (or system) and to ensure
management structures do not compromise any aspect of the life cycle.
This chapter covers the various management concepts that should be employed
as well as the process for line design, construction and maintenance with role clarity
provided.
The organograms or management structures (hierarchy) have deliberately been
excluded as the concepts may be met in many different ways depending on the uti-
lity structure and the insourcing or outsourcing of resources.

3.2 Management Concepts up to Commissioning

A Transmission line, as defined in Chapter 14, is a device that transmits power over
long distances. It should be seen as a single device (or system) with electrical properties
that contribute to power transmission within the supply grid. Chapter 14 outlined how
a line can be “tailor made” to meet the system planner’s requirements within the grid.
A line can also be regarded as a large mechanical structure (system) spanning
many kilometres over many terrain types that must endure varying adverse weather
conditions. It is therefore a device requiring mechanical, civil, environmental, geo-
technical and electrical consideratons in the design of the line.
This section deals with the different possible management concepts for the line
from planning to commissioning.

3.2.1 Management Concepts for Preliminary Design

and Optimisation Studies

3.2.1.1 Series Model

It is often the case that the tower designs are done separately from the conductor
selection and the electrical parameters which are normally determined up front
often from standard configurations. This is either outsourced or performed in house
by a separate division. The tower design is determined from the phase configuration
and bundle configuration supplied and is considered fixed. Likewise the foundation
design is then determined based on the loadings provided by the tower designers
and is also considered fixed.
In this model the tower designers do not consider or question the reasoning
behind the conductor bundle and phase configuration but convert the requirements
into loads for a tower to withstand. Likewise the electrical engineers who determine
the conductor bundle and phase spacing maybe ignorant as to the effect their requi-
rements have on the mechanical aspects of the line.
The series model and distance between the disciplines is often further exacerba-
ted with outsourcing of tower and foundation designs. This often requires a specific
scope and specification which has little flexibility.
3 Planning and Management Concepts 29

The series model will result in sub optimal designs in most cases. It is a more
simple model to manage than the iterative model described in the next section.

3.2.1.2 Iterative Model

In order to optimise the design, either for a specific line or to develop new or revise
standard designs, the electrical, mechanical and civil engineers need to work as a
team to ensure iteration of the design components. This is because each aspect
affects the other. For example an electrical engineer may specify a 6 bundle Zebra
conductor with a certain phase spacing for a 765 kV line. The tower designer will
design accordingly without comment in the series model. However, if the iterative
model is applied, it may be found that a quad bundle with a lower steel content
conductor with slightly wider phase spacing could be used at a reduced cost due to
the lower mechanical forces (wind load and guy wire for example).
If the line is to be optimised as proposed in Chapter 14, then the design of the line
cannot be performed in series with the planner determining the conductor type and the
tower being designed based on the conductor specified followed by foundation design.
The iterative model is more difficult to implement from a management viewpoint
as it involves engineers from different disciplines working together and understan-
ding each other’s field with regard to transmission lines. This can be achieved in a
number of ways:

• In house design (insource): a line design department is established with different

disciplines under one manager. The aim is to create a group of line design
engineers with in depth knowledge in one of the disciplines. Tower detailing and
detailed foundation design can still be outsourced after the tower outline is opti-
mised and conductor bundle and type chosen.
• In house management (outsource): in this case the company employs one or two
experienced line designers who manage an outsourced team of engineers of dif-
ferent disciplines. The contracting of the engineers may be done on a time basis
and not on an output based using a defined scope.

3.2.2 Management Concepts for Route Selection and Property

Acquisition

The route selection and acquisition may form a separate project as the time duration
is normally far longer than the construction period. However, a project management
approach with designer involvement is recommended. This team may include a
variety of environmental, legal, negotiators and technical experts some of which
may be outsourced.
It is important that the negotiator understands the line design as well as the impli-
cations of certain concessions made with the land owner. An additional strain tower
to avoid a certain point in the land or to follow the border of the property can
increase the line cost overall.
It is important that the line designers are part of the team to advise on technical
matters as well as to be aware of the agreements reached. For example in certain areas
30 R. Stephen

of the line route, guyed structures maybe prohibited or existing servitudes (right of
ways) used requiring narrow servitude multicircuit multi voltage pole structures.
In addition to all the permits required for a line to be built, it is also important that
the team include wildlife experts for flora and fauna as well as bird experts to ensure
flight paths and nesting grounds are catered for.
The project team for the route selection and property acquisition is often the
largest and most diverse of teams required to realise a final constructed line. It is
also the longest serving. For this reason the handover to different Project Managers
needs to be well managed with decisions clearly documented.

3.2.3 Management Concepts for Construction

The construction phase consists of construction activities and handover for operation.
This phase is best managed by a team under a project manager using matrix
management whereby the team members are seconded to the team from manage-
ment. The Project Manager decides on the timelines and outputs and informs the
team members. There may be a sub structure in the team for environmental, design,
construction elements.
The team can be outsourced or insourced to varying degrees depending on the
company. If the team is totally outsourced from various suppliers it is important not
to duplicate the structure with in house staff. This could result in an in house project
manager managing the outsourced project manager with duplication of work, con-
flicting instructions to contractors etc. If the project manager is outsourced the in
house resources need to manage the outputs and milestones by exception and not
interfere with the team below the Project Manager.
It is also important that the project manager be appointed as early in the project
life cycle as possible so that he understands the reasons for the design chosen, issues
with line route and stakeholders. It is also desirable that one project manager be
accountable from the start to the end of the project to avoid handover points with
possible misinformation arising. It is also necessary to have one person accountable
for time, cost and quality to avoid blaming later in the project.

3.3 Responsibilities

In any management structure it is important that roles and accountabilities are cle-
arly defined. The three main role players in the establishment of a line are the
Network Planner, Project Engineer and Project Manager. These may work for diffe-
rent companies and be contracted separately. They do need to fulfil the responsibi-
lities as listed below.
The following roles are responsible for the design and construction of overhead lines-

• Network planner
– Responsible to ensure the line requirements (R,X,B, loading for AC lines) are
correct.
3 Planning and Management Concepts 31

– Responsible to ensure the project is released for preliminary design timely.

– To ensure correct targeted in service dates are communicated.
• Project Engineer
– To assist in the route selection process
– To ensure the consultative process is followed relating to the design options
and selection of the optimum design.
– To ensure full understanding of all aspects of the line design as well as how to
implement this on site.
– To assist the Project Manager in the construction phase of the project.
• Project Manager
– To ensure milestones are set and resources are arranged so that target dates are
met.
– The project manager is to understand the nature of lines and the types of pro-
blems and issues that may result from the line design and construction
process.
– To take charge of the programme (target dates, milestones), from the pre-
engineering stage to final commissioning.
– The project manager is to authorise all costs to the line. This is to include
over-heads which are normally accrued without the project manager being
aware of them.

3.4 Life Cycle Process up to Commissioning

The life cycle of a transmission line commences with the system planner and is
completed when the line is decommissioned. This section deals with the planning,
design and construction aspects.

3.4.1 Planning Requirements

As mentioned in Chapter 14 the planner is to provide information on the electrical

requirements of the line. This can be submitted in the form of a table as indicated
below. Note that even if standard designs are used, it is useful to have the require-
ment specification from the planner to check whether the proposed standard design
will meet the requirements.
The planner should provide the information to the line designers as indicated in
Table 3.1. A few fields may need explanation.

• The profile of the proposed line load on a yearly and daily basis is essential for
the designer to determine the templating temperature of the line. If the load pro-
file shows a morning peak and a winter peak it may be possible to use a smaller
conductor as the temperature reached under peak load may be relatively low. If
the daily load profile is very peaky, it may be possible to use small conductors
with a higher templating temperature. If the load profile is flat it may be neces-
sary to use larger conductors with a lower templating temperature.
 32

Table 3.1 Information required to be supplied by Network Planner.

Project Project name and number of line to be built.
Name
Start Sub Substation at which the line starts.
Finish Sub Substation at which the line terminates
NDP ref. (Name and reference of Network development plan)
Planner (enter name of planner responsible)
Year 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Load MVA
Hour 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Profile
Month 1 2 3 4 5 6 7 8 9 10 11 12
Profile
Imped. R X B R X B
(Z) (Z) (Z)
Max
Min
Cost/km
Environ (Include possible environmental constraints)
Time (Include cost to utility for every Month delayed from preferred due date)
Voltage Preferred Second option
kV
R. Stephen
3 Planning and Management Concepts 33

• The load growth is necessary to determine the optimum aluminium area of

conductors.
• The impedance is necessary to determine the optimum bundle and phase confi-
guration (AC lines). The designer needs two values, a minimum and maximum
value to ensure the optimum impedance to the network.
• The voltage of the line is often fixed and cannot be altered, however, in certain
cases the costs can be such that the lower voltage option provides a major cost
saving. For example a 10MVA load required to be transferred at 132 kV may
prove far more effective at 66 kV (sub transmission)
• The time the line is required is often exaggerated by the planner who expects a
certain load growth, however, there are times when the utility can lose substan-
tial amounts of revenue should the line be delayed. In these cases it is often cost
effective to utilise more than one contractor and fast track the project.

In addition to the above the maximum voltage level and fault level needs to be spe-
cified. The reliability level also needs to be specified by the system planner. Line
ratings (normal, emergency) also need to be known (these may be a minimum and
the line designer must ensure these are met).

3.4.2 Route Selection and Property Acquisition

The selection of a line route is often a very time consuming task that could take
many years (in some cases over 20). In the management of the line route acquisition
it is important that the route of “least objection” is not always obtained. It is often
required that the route chosen is along the border between properties or along a
road. In this case there is normally a large amount of bends and angle structures
required which results in an increase of the overall line cost. Management needs to
be cognisant of these issues when appointing consultants to obtain route or when
using in house resources.
In contracts for negotiators, if they are outsourced and in house resources are not
utilised, the payment method should include the cost of the line as a result of the
route chosen. This could include for example a ratio of tangent (suspension) to
angle or strain towers. If the contract is time based the negotiation process may be
extended, if the contract is a turnkey for the servitude/easement to be signed up it
may result in a large number of bends.
A project management approach is also recommended for this process as it
covers many years, different resources and involves many public forums which
require knowledgeable presenters. The project team may include environmentalists,
negotiators, surveyors, marketers, design engineers and even health professionals if
the electromagnetic field issues are raised.
As the process is a long one, the decisions and concessions taken need to be
clearly documented and stored in an organised manner to ensure that no concession
is overlooked in the design and construction phase. Concessions may be made to
land owners in the early stages of the route acquisition. However, the nature of the
34 R. Stephen

concession may be overlooked in the design and construction stage resulting in

rework or lengthy delays.
Chapter 6 and Cigré TB 147 cover the requirements for property route selection,
consultation models and property acquisition.

3.4.3 Management Process for Preliminary Design

and Optisation Studies

As proposed in Chapter 14, the purpose of this stage is to determine a group of

design options (ten options provide a wide range that will ensure possible optimisa-
tion) and, via the use of optimisation tools, to select the final group options for
further analysis. The number in the final group may be around 3 or 4 depending on
whether an option is obviously superior.
The project engineering team is to be formed under a knowledgeable line design
expert. The team should include members knowledgeable in conductor, tower and
foundation issues. This encompasses electrical, mechanical and civil disciplines. As
mentioned in section 3.2.1, these team members can be from the same or different
companies.
Possible routes can be determined from the digital terrain map or from the
Environmental Impact Analysis. If this is not possible the design team should select
a profile similar in nature to that expected in the line that they are designing.
The line route can seriously affect the overall cost of the line. It is imperative that
the line design leader approves the line design route prior to final negotiations and
servitude acquisition.
The possible conductor and bundle options are then considered.
The tower family can then be determined from the selected terrain (either the
actual terrain or a sample selected from terrain that is similar in nature) using the
conductor families selected and the available tower families.
The planners are then to be consulted again to ensure that the proposed set of line
design options is in line with the planner’s requirements.
The Appropriate Technology Indicator (ATI) (covered in Chapter 14) is then to
be determined for the options. The best (approximately four) options are to be taken
to the next stage. This can be achieved by using the programme (spreadsheet) that
determines the scores for each option.
The ATI takes into account the initial cost, cost of losses, MVA thermal rating
and MVA surge impedance loading.
The motivation for project finance (may be made in the preliminary design
phase) is to be completed. This is to include the following:-

• A design document which should contain the following information:

– Reason for the line including extract from the planning proposal documenta-
tion. This normally includes the Net present value cost benefit analysis.
– Information from planners on line requirements including time and cost
constraints.
3 Planning and Management Concepts 35

– Possible routes with cost options if possible.

– Environmental Impact Assessment (EIA) details (extract from the EIA pro-
duced if possible. This may vary from country to country).
– Options selected for analysis.
– Appropriate Technology Indicator (ATI) analysis and results.
– Design options to be analysed further.
• A cost estimate of the following
– Labour cost to complete the detailed design stage.
– Cost of the geo tech survey
– Cost of tower development (if required).

The cost estimate is to be added to the project costs for approval via the relevant
governance procedures.

Note that the geo tech survey is often overlooked. It is a critical aspect of the
detailed design stage. It determines the suite of foundation options that
should be used on the line. It assists in determining the numbers to specify
in the enquiry document or for construction. It also highlights the type of
foundations that need to be designed or developed.

3.4.4 Management Process for the Detailed Design Phase

The detailed design phase precedes the execution phase. The output of this phase is
a detailed design and costing which is then presented to management for execution
approval.
In some cases the utility may decide to go to market to obtain actual tendered
prices for submission to management. In other cases estimates may be made from
previous projects. In the former case the risk of error in costing is lower.
Approval is normally required from management for the detailed design phase to
commence. The output of the design phase is a detailed cost of the line (90 % or
higher) accurate. This can only be achieved if the following are known:

• The line route and line profile.

• The tower positions and types per position.
• The detailed bill of materials including conductor, towers, insulators and
fittings.
• The contract prices for the erection phase as well as material costs.

It should be noted that the construction and material costs can be very volatile and
depends on the exchange rate, the availability of materials and labour. It can vary
from line to line even if they are in the same geographical area.
The line route and actual line profile (as accurate as possible) needs to be used to
determine the optimum tower, conductor and foundation combination.
36 R. Stephen

The options short listed from the ATI analysis are to be used to determine the opti-
mum combination. These are to be used using the actual line profile and line route
chosen. Tower spotting packages should be used to determine the best option in rela-
tion to the ATI. This will mean that network planners have to be consulted once again
to determine the best option including the R, X, B values as well as the cost of losses.

Note that the best design combination is to be determined together with

Maintenance staff. Preferably there should be a technical governance
structure which will assess the proposed design option. This committee
should consist of maintenance and operation staff as well as line designers
and system planners. These may be from different companies. If new towers
are to be designed, it is necessary that the maintenance and construction
staff are to be involved. This could mean engaging with contractors who are
likely to be involved in the construction of the line.

Once the best combination is determined, it is necessary to optimise the tower selec-
tion for each tower position using a tower spotting programme. This is best perfor-
med by a “peg walk” of the route. A “peg walk” or “tower staking” is a walk down
the line by the line designers to determine if the proposed tower selection in the
proposed tower sites are optimal. This also includes accessibility, constructability
and proximity to roads, drains or even unmarked graves that may not have been
known about when the line was profiled.
In order to determine the type of foundation to use at each site, it is necessary to
perform a detailed geotechnical survey at each tower site. This can only be perfor-
med after the tower, foundation and conductor combination has been chosen and the
first selection of tower types per site is completed.
The peg walk can be conducted at the same time as the geo tech survey. The peg
walk will determine if any tower position changes are required or if a different tower
needs to be placed at a specific location. In addition access roads can be planned as
well as farm gate positions and types of gate (e.g. game gates).
The information gained from the peg walk and geo tech survey can be used in the
finalisation of the Environmental Management programme which is essential for
each project.
The information available at this point is the full tower schedule, the detailed bill
of materials, the line profile with tower types, the access road locations and gate
locations and types.
This information may then be compiled into an enquiry document. The docu-
ment is then sent to prospective contractors for tender.
Once the tenders are received it is necessary to analyse the tender in detail. The
services of a quantity surveyor should be used at this stage.
Once the appropriate tender has been selected the price and motivation needs to
be submitted for approval (this will depend on the utility governance requirements),
it may be permitted to proceed if the tendered prices are within a certain percentage
of the estimate.
3 Planning and Management Concepts 37

The design document should now be updated to include:

• Planning information
– Reason for the line including extract from the planning proposal documentation.
– Information from planners on line requirements including time and cost
constraints.
• Survey and Environmental
– Possible routes with cost options if possible.
– EIA details (extract from the EIA produced if possible).
– Route and profile
• Initial tower, conductor, foundation combinations
– Options selected for analysis.
– ATI analysis and results.
– Design options to be analysed further
• Towers
– Tower design chosen with reasons
– Tower schedule summary
• Conductor and earthwire
– Final conductor or conductor bundle chosen with reasons.
– Final earthwire chosen with reasons.

Note the earthwire selection is dependent on the fault level as well as the
fault dissipation in the towers and ground. This analysis is to be performed
as part of the earthwire selection. In addition the interference criteria for
telecommunication lines are to be taken into account and described here.

• Foundations
– Geotechnical survey results
– Foundation designs for each soil category and tower type
– Schedule summary of foundation types
• Hardware
– Outline of suspension and strain assemblies
– Damping system used
– Clamps and fittings
• Insulators
– Analysis of pollution and other requirements
– Insulator options that could be used
– Final insulator selection and reasons
• Performance assessment
– Performance analysis of other lines in the vicinity
– Lightning performance studies
– Environmental impact studies such as bird pollution/interaction with the line,
veld or cane fires
38 R. Stephen

• Contracts (tenders received)

– Summary of tenders received with technical analysis as to their suitability.

Once the overall project is approved by management, the contractor responsible for
construction may be appointed.

3.4.5 Project Execution (Construction)

An overhead power line is unlike a substation in that there are many factors that
cannot be taken into account before construction begins. Items such as access to
site, soil types and tower erection methods may necessitate that the tower type,
foundation type, equipment used or tower location be changed on site subsequent to
the design being approved. This includes right of way clearing requirements and
stringing specifications (location of equipment, drums).

3.4.5.1 Pre Construction Planning

The Project Manager responsible for the project is to set up a pre-construction mee-
ting to plan the construction activities with the contractor, clerk of works and Project
Engineer (who may include design staff). The following items should be
discussed:

• Material arrival and storage

Note that the control of material on site is critical to the success of the pro-
ject. Nuts and bolts as well as spacer dampers, and insulators should be kept
in a clean environment preferably off the ground. Composite insulators
should be handled in accordance with the Cigré composite insulator hand-
ling guide (Cigré TB184 2001) (TB 184). There must also be a system whe-
reby the material issue is controlled and the stock levels are known. A
person should be placed in charge of the store which should be fenced off.

• Project plan for construction including possible dates for line and road
crossings.

Note that line and road crossings need detailed up front planning. In the
case of line crossings the permission to take the line out needs to be obtained
from the Operations authority. The detailed bill of material, tools, and pro-
cedure needs to be drawn up and agreed to well in advance. Special items
such as cranes, helicopters etc may also be required. This activity occurs
after the permission has been obtained from the relevant authorities rela-
ting to the crossing.
3 Planning and Management Concepts 39

• Environmental Management Plan (EMP) issues and plans as to how the EMP
will be met. The EMP is a plan to meet the environmental requirements of the
line. It includes the rehabilitation of the environment.

Note that this will include the formation of access roads, the clearing of the
servitude and whether tyre or track vehicles are to be used.

• Foundation design issues and tools to construct the foundations.

Note that it may be possible that certain foundation designs will need to be
done or modified. There may, for example be a large rock area and rock
piles may not have been designed. Or the contractor may not have the drill
bits for the type of rock foundation required. This needs to be taken into
account and resolved up front.

• Safety

Note that this is a standard item on all site meetings. It includes the safety
procedures and equipment required and available to staff.

3.4.5.2 Foundation Nominations

The foundation types are to be nominated by an experienced person other than the
contractor. This is due to the fact that the contractor is likely to err on the conserva-
tive side or on the side of higher financial returns.
The foundation types need to be documented per tower installation.

3.4.5.3 Backup Technical Support

Issues may arise on a daily basis that require urgent solutions. An example could be
a drain or grave that is found to be in the foundation location. In these cases where
a tower needs to be moved or a plan made on other matters, it is essential that com-
petent backup available for assistance. The Project Engineer is to ensure that this
back up is available.

3.4.5.4 Change Control Process

The resulting solution from the back up support may result in a tower move or other
modification from the original design. These changes must be controlled in the fol-
lowing manner:

• The Project Engineer is to examine the proposal and agree to the change if
applicable.
40 R. Stephen

• The Project Engineer is to ensure the design document as well as line profiles or
other drawings as appropriate are updated (once approved and executed).
• The cost of the change is to be determined by the Project Engineer.
• The reason for the change as well as the cost is to be submitted to the project
manager for approval.
• The Project Manager needs to sign off the proposal and update the project costs
and, where applicable, the projections (both cost and time).

Note that in order for the Project Engineer to understand the issues relating
to problems on site he is to visit site regularly and have a very good under-
standing of the requirements of the project and the final design option
chosen.

3.4.5.5 Clerk of Works

Clerk of Works need to have the skills and knowledge to oversee the following
activities:

Note that Clerk of Works (COWS) are key to the success of the construction
stage. The COW are the “eyes and ears” of the customer on site. They need
to have knowledge of what is required to be done on site to ensure that the
contractor is executing the work correctly.

• Excavations and foundation confirmation/selection

• Foundation cementing including testing (slump and cube test) as well as the
correct method to vibrate the concrete.
• Assembly and erection of structures.
• Stringing of conductors including running out, regulating and clipping in.

In some cases it may be necessary to have more than one COW on site at one time
and certain COW’s may be specialised different areas such as foundations, tower
erection and stringing.

3.4.5.6 Tests Required During Construction

There are a number of tests that are required during the construction process on site.
Examples of these include:

Foundations: Cube and slump tests for concrete

Pull out tests for guy anchours

Tests may also be required in the laboratory such as tensile tests for compression
fittings and Guy anchor assembly tensile tests.
Tower footing resistances need to be measured per tower and documented.
3 Planning and Management Concepts 41

Impedance measurements should be conducted on the completed line to compare

them with the designed values as well as to update the planning and fault level parame-
ters with exact values thereby enabling a more robust planning database to be created.

3.4.5.7 Inspection Prior to Commissioning

A walkdown to each structure should be performed by the COW, Site Manager, the
Project Manager and the asset owner representative. A list of defects need to be
created which needs to resolved within a specified time.

3.4.5.8 Site Access Control

The design staff should be permitted to visit site at any time. However, they must
inform the Project Manager who needs to make arrangements. The design staff must
not communicate directly with the Contractor unless the Project Engineer or a
representative of the Project Manager is present.

3.5 Forms and Records (Including Accreditation)

The following staff that are involved in the following activities need to be accredited
to perform these functions:

• Design document – It is assumed that the design team would have sufficient
knowledge to conduct the design process. The Design document needs to have
each section signed by the designer and a person who is checking the section.
The latter needs to be a registered professional (Professional or certificated
engineer depending on the country).
• Tender evaluation – The technical evaluation of tenders needs to be performed
with the presence of at least one expert in the field of line design. This needs to
be a Professional engineer or registered professional.
• Foundation nomination – A technician approved by the foundation designer or
member of the design team assigned to foundations (must be a registered profes-
sional) should carry out nominations.
• Clerk of Works – the clerk of works needs to have had training and experience in
each of the fields mentioned before he may oversee the work in these areas.
• Records of site visits by the design engineer as well as the findings must be
recorded on site together with the required actions.

3.6 Summary of Process

The process comprises the following steps:

1. Planning proposal (concept) release of project and pre-engineering funding is

obtained to conduct the concept design. The output is the conductor, tower and
possible foundation combinations.
42 R. Stephen

2. Obtaining of servitude and Environmental approval. In some cases this may

take up to 20 years. In cases where it is likely to take longer than 3-5 years a
separate project may be commenced. Often, however, the negotiations may
require a final tower design to be presented. In this case the concept design will
need to be conducted.
3. Perform pre-engineering design and determine a short list of options that could
be investigated further.
4. Submit the project for detailed design stage approval and funds for detailed
design.
5. Obtain route and line profile.
6. Determine the optimum conductor, tower, foundation combination.
7. Determine tower positions and types per position.
8. Conduct geotechnical survey, peg walk to determine gate positions, foundation
types and access road.
9. Complete Environmental Management Plan.
10. Complete and issue enquiry documents based on results of optimisation and
geotech survey.
11. Evaluate tenders (using line design experts and quantity surveyor)
12. Compile and submit request for permission to construct line to Investment
committee or management structure
13. Conduct pre-construction meeting
14. Construct line.
15. Process for inspection and commissioning.

Note that in some cases the steps 10 and 11 take place after the step 12. This
is due to the time required for step 12. This is not ideal, however, as the cost
of construction submitted by contractors is difficult to determine upfront
and depends on workload, competition and cash requirements of
contractors.
It is critical, that the step 5 is obtained before steps 6–12. The temptation
is to begin construction or issue the tender before the route or line profile
has been completed. As the line is useless until completed, by commencing
the line too early may result in a weaker case for obtaining the route (right
of way or servitude) or standing time claims from contractors. In cases of
extremely long lines it may be prudent to commence construction on a sec-
tion if the line route on the remaining sections can be altered if need be.
It is a fallacy that the quicker the tender is issued the quicker the line is
completed. It is preferable to have all permissions approved prior to com-
mencement of construction.
3 Planning and Management Concepts 43

3.7 Management of Maintenance

3.7.1 Involvement at Design Stage

It is critical that the maintenance staff are involved in the design of the line irrespec-
tive of whether the towers and conductors are standard or not. This is to ensure
access to towers is mutually agreed prior to construction as well as maintenance
methods.
In the case of Live line maintenance the maintenance staff must be involved in
determining the live line maintenance methods prior to design finalisation. Special
tools may need to be developed or insulated cranes or personnel lifts purchased. In
cases of new tower designs it is necessary to determine the adequate spacing by
using dummy objects with full scale electrical impulse tests. This will determine
whether the spacing proposed is adequate.

3.7.2 Information Required and Handover (Submission)

It is critical that the asset owner and maintenance staff be involved as early as pos-
sible in the construction of the line. his will allow for all issues to be resolved prior
to handover from the construction company. A detailed handover check sheet needs
to be established and agreed to prior to the commencement of construction. This
should include checks per tower.
Maintenance of the constructed asset may be contracted out to a number of exter-
nal suppliers. It is important for the asset owner to ensure that, prior to taken over
the line, all information is available on the line that has been constructed or refur-
bished. This includes:

• Line as built profiles

• Map of line route with landowner details
• Tower type per tower position
• Conductor details (supplier, type, contract, numbers etc)
• Earthwire and OPGW details (supplier, type, contract numbers etc)
• Hardware details (supplier, type, contract numbers etc)
• Drawings of towers and assemblies.
• Foundation types per tower leg (this is particularly important with large tower
footing areas such as the cross rope suspension)
• Results of concrete slump and test cube (compression) tests.
• Tower footing resistance per tower.
• Earthing used especially if additional earthing has been applied.
• Accessories such as aerial warning spheres, bird guards.
• Insulator types per tower if different (composite and glass may be used on one
line for example), information to include creepage, material, insulator profile.
• Location of conductor joints as well as the compression tool number to perform
the joint (if such joints were installed).
44 R. Stephen

3.7.3 Information for Maintenance during Operation

In addition to the above it is important that the maintenance required is also provi-
ded. This includes the type of inspections to be conducted and the time interval.
Items such as retensioning guy wires should also be included.
Regular line inspections should be conducted with focus on the following:

• Tension in guy wires

• Evaluation of the earthing system at the towers
• Structure condition
• Vibration and spacer damper – condition and orientation
• Aircraft warning spheres, condition
• Condition of guy anchors
• Insulator damage
• Servitude condition and access to towers – poor maintenance of servitudes could
inhibit effective restoration of lines as well as affect line performance depending
on the vegetation below the line.
• Thermal imaging of the line for potential electrical hot spots
• Periodic assessment of the conductor – especially in corrosive environments –
this may entail taking physical conductor samples and testing them in a
laboratory
• Records should be maintained with the fault information for each line so that
analysis and performance improvement projects can be undertaken if necessary.

3.8 Conclusion

The current utility structure often differs from vertically integrated to fully outsour-
ced with many different companies involved in the life of the asset. This chapter
highlights the concepts for management of the line in the different stages. The
organogram or management structure can vary between utilities and companies and
still comply with the proposed concepts. As a result the management structures
have not been discussed.
The aim is to make the utility aware of the management concepts to be applied
and to apply these in the best manner that befits the company structure and insour-
cing or outsourcing policy.

3.9 Highlights

The management of the line life cycle is consistent irrespective of the structure of
the company or companies involved with different stages of the life cycle.
It is important for the asset owner and operator to be aware of the management
concepts and process to ensure all stages are well catered for even if they fall outside
the domain of the company.
3 Planning and Management Concepts 45

The concepts highlighted in this chapter have been proven over many years and,
although they may vary from country to country in detail (for example it may not be
practice for the planners to provide the information as suggested and only state a
conductor type), each stage must be covered to some extent.

3.10 Outlook

It is likely that the planning, design, construction, maintenance and operation of

overhead power lines will be conducted by different companies with different goals,
make up and skills compared to the past. In this case it is important that the process
through the life cycle of the line is not compromised. This can be achieved by the
role players being aware of the process as described in this chapter.
The risk exists that the more independent the role players are the less the overall
planning and management concepts will be understood and executed. it is possible
that each stage is seen in isolation to the detriment of the overall line design, opera-
tion and maintenance. This risk is likely to increase in future as the role players
become more segregated.

References
Cigré TB147: SC 22 WG 22.14. High voltage overhead lines. Environmental concerns, procedu-
res, impacts and mitigations (1999)
Cigré TB184: SC 22 WG 22.03. Composite insulator handling guide (2001)
Cigré TB265: SC B2 WG B2.15. Life Cycle Assessment (LCA) for overhead lines (2005)

Rob Stephen has MSc, MBA degrees and recently received

his PhD degree in line optimisation from the University of Cape
Town South Africa. He is employed in Eskom, the South African
utility where he holds the position of Master Specialist. He has
been involved in all aspects of line design as well as network
planning, electrification and project management. In study com-
mittee B2 (overhead lines) he has held position of working group
convener, special reporter, advisory group convener and was
chairman of SC B2 from 2000–2004. He has published over 100
papers and been involved in technical brochures on aspects of
thermal rating, real time monitoring and overall line design since
1988. He is an horonary member of Cigré and a fellow of the
South African Institute of Electrical engineers.
Electrical Design
4
Joao Felix Nolasco, José Antonio Jardini,
and Elilson Ribeiro

Contents
Section 1: Electric Parameters of Overhead AC Transmission Lines
4.1 Electrical Characteristics..................................................................................................... 48
4.1.1 Introduction............................................................................................................. 48
4.1.2 Resistance................................................................................................................ 49
4.1.3 Inductance............................................................................................................... 49
4.1.4 Capacitance............................................................................................................. 50
4.1.5 Negative and Zero Sequence Parameters................................................................ 50
4.1.6 Representation of Lines........................................................................................... 51
4.1.7 General Overhead Transmission Line Models........................................................ 55
4.2 Surge Impedance and Surge Impedance Loading (Natural Power).................................... 68
4.2.1 Methods for Increasing SIL of Overhead Lines...................................................... 69
4.2.2 Compact Lines......................................................................................................... 69
4.2.3 Bundle Expansion................................................................................................... 71
4.3 Stability............................................................................................................................... 71
4.4 Thermal Limit and Voltage Drop........................................................................................ 72
4.5 Capability of a Line............................................................................................................. 74
4.6 Reactive Power Compensation............................................................................................ 74
4.7 Electromagnetic Unbalance - Transposition....................................................................... 75
4.8 Losses.................................................................................................................................. 75
4.8.1 Losses by Joule Heating Effect (RI2) in the Conductors......................................... 75
4.8.2 Dielectric Losses: Corona Losses, Insulator and Hardware Losses........................ 75
4.8.3 Losses by Induced Currents.................................................................................... 76
4.9 Reliability and Availability................................................................................................. 76
4.10 Overvoltages....................................................................................................................... 77
4.10.1 Fast-front Overvoltages (Lightning Overvoltages)................................................. 77
4.10.2 Temporary (Sustained) Overvoltage...................................................................... 114
4.10.3 Slow-Front Overvoltages (Switching Surges)....................................................... 116

Originally published by Cigré, 2014, under the ISBN 978-2-85873-284-5. Republished by Springer
International Publishing Switzerland with kind permission.
J.F. Nolasco (*) • J.A. Jardini • E. Ribeiro
Florianópolis, Brazil
e-mail: [email protected]

© Springer International Publishing Switzerland 2017 47

K.O. Papailiou (ed.), Overhead Lines, CIGRE Green Books,
DOI 10.1007/978-3-319-31747-2_4
48 J.F. Nolasco et al.

4.11 Insulation Coordination...................................................................................................... 119

4.11.1 General.................................................................................................................. 119
4.11.2 Statistical Behavior of the Insulation.................................................................... 120
4.11.3 Insulation Coordination Procedure........................................................................ 122
4.11.4 Withstand Capability of Self Restoring Insulation................................................ 127
4.12 Electric and Magnetic Fields, Corona Effect..................................................................... 128
4.12.1 Corona Effects....................................................................................................... 132
4.12.2 Fields..................................................................................................................... 140
Section 2: DC Transmission Lines
4.13 Overvoltages and Insulation Coordination......................................................................... 143
4.13.1 Overvoltages.......................................................................................................... 143
4.13.2 Insulation Coordination......................................................................................... 147
4.14 Pole Spacing Determination............................................................................................... 156
4.14.1 Case of I-Strings.................................................................................................... 156
4.14.2 Case of V-Strings................................................................................................... 157
4.15 Conductor Current Carrying Capability and Sags............................................................. 159
4.16 Tower Height...................................................................................................................... 160
4.17 Lightning Performance...................................................................................................... 161
4.18 Right-of-Way Requirements for Insulation........................................................................ 163
4.18.1 Line with I-Strings................................................................................................ 164
4.18.2 Line with V-Strings............................................................................................... 164
4.19 Corona effects.................................................................................................................... 165
4.19.1 Conductor Surface Gradient and onset Gradient................................................... 165
4.19.2 Corona Loss........................................................................................................... 168
4.19.3 Radio Interference and Audible Noise.................................................................. 171
4.20 Electric and Magnetic Field............................................................................................... 174
4.20.1 Ground-Level Electric Field and Ion Current....................................................... 174
4.20.2 Magnetic Field....................................................................................................... 182
4.21 Hybrid Corridor or Tower.................................................................................................. 183
4.21.1 Conductor Surface Gradient.................................................................................. 183
4.21.2 Radio Interference................................................................................................. 184
4.21.3 Audible Noise........................................................................................................ 185
4.21.4 Corona Losses....................................................................................................... 185
4.21.5 Electric and Magnetic Fields................................................................................. 186
References to 4.1–4.12................................................................................................................ 186
References to 4.13–4.21.............................................................................................................. 188

4.1 Electrical Characteristics

4.1.1 Introduction

Electrical parameters of transmission lines or otherwise refered to as line cons-

tants, resistance, inductance and capacitance (R,L,C) are used to evaluate the
electrical behavior of the power system. Depending on the phenomena to be
studied a different set of parameters is required. For load flow and electrome-
chanical transients the parameters used are the positive sequence. In the
short-circuit calculation the positive/negative/zero sequence parementers and
for electromagnetic transients the phase parameters and its frequency-depen-
dent parameters.
4  Electrical Design 49

For the former case, normally, the line is considered full transposed and there are
simple equations to determine the parameters. For others cases digital programs are
used like the ATP. The various procedures of calculation are discussed here-in-after,
starting with straight forward calculation for positive sequence model and comple-
ting with a general calculation.

4.1.2 Resistance

The resistance of conductors R is found in the manufacturers catalog. The values of

resistance in Ω/km for dc current at 20 °C and sometimes for ac (50 or 60 Hz) are
given as function of the conductor cross section.
R -the resistance of the bundle- is then the one sub-conductor resistance divided
by the number of them in a bundle.
It should be noted that manufacturers’ catalog indicate, normally, conductor
resistance (R20) for dc at 20 °C. For other temperature (Rt) a correction shall be applied:

Rt = R20 1 + β ( t − 20 )  (4.1)

where t is the conductor temperature and β the resistance temperature coefficient
equal to 0.00403 for Aluminum and 0.00393 for copper. Aluminum Association
provides specific values of β for every conductor section in the ranges 25-50 °C and
50-75 °C.
Example: for the conductor ACSR 954 MCM (45/7), extracting the individual
resistances from a Catalogue (Aluminum Association Handbook) and making the
calculation of β coefficients, Table 4.1 is obtained:

4.1.3 Inductance

The inductive reactance of the transmission line is calculated by (Stevenson 1962):

 GMD 
X l = 2 w10 - 4 ln   ( Ω / km ) / phase (4.2)
 GMR 

Table 4.1  Examples of coefficients of resistance variation according to temperatures

ACSR 954 MCM (Rail)
t(°C)
Unit 25 50 75
Ω/Mi 0.099 0.109 0.118
Ω/km 0.061778 0.067744 0.073337
β(25-50) 0.003863
β(50-75) 0.003303
50 J.F. Nolasco et al.

w = 2π f (4.3)

f is the frequency.
GMD and GMR are the geometric mean distance and geometric mean radius.
For a single circuit fully transposed:

GMD = 3 d ab d ac dbc (4.4)

dab dac dbc are the phase distances.
For bundle of n sub-conductors located in a circle of radius R and being a the
equal spacing between adjacent sub-conductors the equivalent radius of the bundle
or the GMR is:

nrk
GMR = R n (4.5)
R
r is the sub conductor radius

a
R= (4.6)
2 sin ( π / n )

k is a correction factor

4.1.4 Capacitance

The capacitance of a full transposed three phase line is calculated by:

0.05556
C= µ F / km (4.7)
 GMD 
ln  k1 
 GMRc 
nr
GMRc = R n (4.8)
R

k1 depends on distances between conductors and conductors to images in the soil

(equal to 0.95-1.0 for 138 kV and 0.85-0.9 for 400 kV and higher voltages).

4.1.5 Negative and Zero Sequence Parameters

Negative sequence parameters are equal to the positive parameters for transmission
lines.
There are straight forward equations also for the calculation of the zero sequence
parameters, however it is recommended to use the procedures described in
(Stevenson 1962; Happoldt and Oeding 1978; Kiessling et al. 2003).
4  Electrical Design 51

4.1.6 Representation of Lines

In this section formulae will be presented for calculating voltage, current and power
at any point of a transmission line, provided such values are known at one point.
Loads are usually specified by their voltage, power and power factor, for which
current can be calculated for use in the equations.
Normally transmission lines are operated with balanced three-phase loads. Even
if they are not spaced equilaterally and may not be transposed, the resulting dissym-
metry is slight, and the phases are considered to be balanced.
The equivalent circuit of a short line is represented by a series reactance only,
which are concentrated or lumped parameters not uniformly distributed along the
line. As the shunt admittance is neglected for short lines, it makes no difference, as
far as measurements at the ends of the line are concerned, whether the parameters
are lumped or uniformly distributed.
The shunt admittance, generally pure capacitance, is included in the calculations
for a line of medium length. The nominal Π circuit, shown in Figure 4.1 below, is
often used to represent medium-length lines.
In this circuit, the total shunt admittance is divided into two equal parts placed at
the sending and receiving ends of the line.
The voltage and current relationships used in electrical calculations under this
approach are:

 ZY 
VS =  + 1  VR + ZI R (4.9)
 2 

 ZY   ZY 
IS = Y  + 1  VR +  + 1 I R (4.10)
 4   2 

Neglecting the capacitance for short lines, the above equations become the well-­
known simple relationships:

VS = VR + ZI R (4.11)

I S = I R (4.12)

The magnitude of the voltage regulation (%Reg) for the case of medium lines is:

VS | − | VR
% Reg = 100 (4.13)
VR

Figure 4.1 Nominal Π R XL
circuit of a line.

Y Z=R+jXL Y
2 2

-j B =-j ωC = Υ
2 2 2
52 J.F. Nolasco et al.

4.1.6.1  Long Transmission Lines

As impedance and admittance are uniformly distributed along the line, the exact
solution of any transmission line is required for a high degree of accuracy in calcu-
lating long lines (for instance longer than 100 km); distributed parameters should be
used in this case.
The following nomenclature is used:

• z = series impedance per unit length, per phase

• y = shunt admittance per unit length, per phase to neutral
• l = line length
• Z = zl = total series impedance per phase
• Y = yl = total shunt admittance per phase to neutral

The following equations can be deduced:

VR + I R Z C γx VR − I R Z C − γx
V= e + e (4.14)
2 2
VR + I R Z C γx VR − I R Z C − γx
I= e + e (4.15)
2 ZC 2 ZC

z
ZC = (4.16)
y

and the propagation constant is:

γ = z y (4.17)

Both γ and ZC are complex quantities. The real part of the propagation constant γ is
called the attenuation constant α, while the quadrature part is called phase constant β.
Thus: γ = α + jβ. The above equations for voltage and current for defining V and I
turn out into:

VR + I R Z C αx j β x VR − I R Z C −αx − j β x
V= e e + e e (4.18)
2 2

V +I Z V − I R Z C −γx
I = R R C e γx + R e (4.19)
2 ZC 2 ZC

A deep analysis, beyond the scope of these highlights, will prove that the first terms
of the above equations are the incident voltage (or current), while the second term
is the reflected voltage (or current). Observe that a line terminated in its characteri-
stic impedance ZC has VR = IR ZC and therefore has no reflected wave.
Such a line is called flat line or infinite line, the latter designation arising from
the fact that a line of infinite length cannot have a reflected wave. Usually power
4  Electrical Design 53

Table 4.2  Typical line parameters and line constants for a 500 kV Line
4 × ACAR 1300 MCM (30/7) Ling length → 365  km β(b)
Parameter R(Ω/km) XL(Ω/km) XC(Ω*km) B(μS/km)) 0,001276
Z1unit 0.013172 0.220388 135411 7,385 γ
Z0unit 0.15317 1.00965 326807 B0(μS/km)) 0,001277
Eq. LT R(Ω) XL(Ω) B/2(μS) 3,060 λ(km)
“Π”nomZ1 4.808 80.442 1347.7 ZC(Ω) 4923
“Π”equiv 4.466 77.576 1372.64 172,9
“Π”nomZ0 55.906 368.521 558.43 α v(km/s)
E/E0 0.785 SIL(MW) 1447 3,81 E-05 295373
Notes: a) 1 MCM = 0.5067  mm2 b) This column refers to line constants
Line data used for the calculation above:
• voltage → 500 kV Tower type → Guyed cross rope
• phase bundle conductor → 4 × ACAR 1300 MCM (~4 × 653  mm2).
• diameter → 3.325 cm Stranding: 30/7
• sub-conductor spacing → 120,0 cm (Expanded bundle)
• phase spacing → 6.41  m
• conductor height at tower → 28.3 (average) m
• minimum distance conductor to ground → 12.0  m
• conductor sag → 22.5  m
• shield wires EHS → 3/8” and OPGW 14.4 mm S. wires spacing → 28.1  m
• shield wire height at tower → 38.0 m shield wire sag 16.5 m
• soil resistivity → 1000 Ω m

lines are not terminated in their characteristic impedance, but communication lines
are frequently so terminated in order to eliminate the reflected wave. A typical value
of ZC is 400 Ω for a single conductor line. For conductor bundles between 2 and 6,
see typical values in Table 4.2. The phase angle of ZC is usually between 0 and −15°.
ZC is also called surge impedance in power lines.
Surge impedance loading (SIL) of a line is the power delivered by a line to a
purely resistive load equal to its surge impedance.

4.1.6.2  Lumped Representation of Lines

The exact representation of a transmission line is usually made through the use of
hyperbolic functions that can treat the line with distributed electric parameters of
resistance, inductance and capacitance.
Such functions are equated in terms of the incident and reflected waves of vol-
tage and current, being summarized as follows:
VS = VR cosh γl + I R Z C sinh γl (4.20)

VR
I S = I R cosh γl + sinh γl (4.21)
ZC

VR = VS cosh γl − I S Z C sinh γl (4.22)

54 J.F. Nolasco et al.

Figure 4.2 Equivalent Zeq

circuit of a long line
(Equivalent Π).
Y -j B eq -j B eq Y
2 2 2 2

VS
I R = I S cosh γl − sinh γl (4.23)
ZC

However for the case of short or medium-length lines, equivalent circuits of trans-
mission lines have been simplified by calculating equivalent series resistance and
reactance, which are shown as concentrated or lumped parameters and not distribu-
ted along the line. The distributed capacitances are also represented by one or two
lumped parameters.
This simple circuit having the shape of the Greek letter Π is named as nominal
Π. The nominal Π does not represent a transmission line exactly because it does not
account for the parameters of the line being uniformly distributed.
The discrepancy between the nominal Π and the actual line becomes larger as the
length of the line increases. It is possible, however, to find the equivalent circuit of
a long transmission line composed of lumped parameters so that voltage and current
relations at both ends are accurate. In view of that, the true line representation that
is really a hyperbolic function can be replaced by a simplified Line Representation
having the shape of the so called Equivalent Π as shown in Figure 4.2. The values
of the equivalent parameters, Req, Xeq and Beq are determined so that the voltages and
currents are the same at the sending-end and receiving-end terminals. The calcula-
tion formulas state below.

sinh γl
Z C sin γl = z (4.24)
γl
Beq 1 tanh γl B tanh ( γl / 2 )
= = (4.25)
2 ZC 2 2 γl

Therefore, starting from the nominal Π parameters and using the circuit constants,
it is possible to calculate the equivalent Π values.

Z eq = Req + jX eq (4.26)

For short lines and low voltage lines, capacitance C is neglected and a simplified
model neglecting the capacitance can be used instead. Such simplification could be
applied to lines below 72.5 kV and for lengths below 30 to 40 km.
An example of calculation of the main parameters and line constant is shown in
Table 4.2 for a 500 kV overhead line.
4  Electrical Design 55

Line data used for the calculation above:

• voltage = 500 kV Tower type = Guyed cross rope
• phase bundle conductor = 4 × ACAR 1300 MCM (~4 × 653  mm2).
• diameter = 3.325 cm Stranding: 30/7
• sub-conductor spacing = 120.0 cm (Expanded bundle)
• phase spacing = 6.41  m
• conductor height at tower = 28.3 (average) m
• minimum distance conductor to ground = 12.0 m
• conductor sag = 22.5  m
• shield wires EHS = 3/8” and OPGW 14.4 mm s. wires spacing = 28.1 m
• shield wire height at tower = 38.0 m shield wire sag 16.5 m
• soil resistivity = 1000 Ω m.

4.1.7 General Overhead Transmission Line Models

Overhead transmission lines are modeled by electric circuit based on their parame-
ters (resistance, inductance and capacitance) and length.
The relation between voltage to ground (V), incremental length voltage drop
along the line (ΔV/Δx), and current (I) or charge (Q) are (Dommel 1986):

• Electromagnetic phenomena

[ ∆V / ∆x ] = [ Z ][ I ] (4.27)
• Electrostatic phenomena

[V ] = [ H ][Q ] (4.28)
For AC system, matrixes Z and H have one line and one column for each conductor
and shield wire. For instance, for an AC line with three phases (sub index p) and two
shield wires (sub index s) they look like:

Zpp Zps

Zsp Zss

For bipolar DC systems the matrixes are similar however with p = 2.
56 J.F. Nolasco et al.

4.1.7.1  Electromagnetic and Electrostatic Line Equations

The terms of the impedance matrix Z in (Ω/km) are:

Z ii = ( Rii + ∆Rii ) + j ( X ii + ∆X ii ) (4.29)

Z ij = ( ∆R ij ) + j ( X ij + ∆X ij ) (4.30)

Rii is the AC resistance of the bundle (one subconductor resistance divided by the
number of them in a bundle)

 1 
X ii = 2 w10 −4 ln   (4.31)
 Reqz 
 ii 
 1 
X ij = 2 w10 −4 ln   (4.32)
d
 ij 

ΔRii, ΔRij, ΔXii, ΔXij, are additional parcels (Carlson correction) for which resistance
and reactance initial term of the series are:

∆Rij = 4 w10 −4
1.5708 0.0026492 ( hi + h j ) f / ρ  (4.33)
 − + 
 4 4 

∆X ij = 4 w10 −4
 2  658.8  0.0026492 ( hi + h j ) f / ρ  (4.34)
 ln  + + 
 4  f / ρ  4 

for ii terms use hi in place of hj
For bundle of n sub-conductors located in a circle of radius R and being a the
spacing between adjacent sub-conductors the equivalent radius of the bundle is:

nr k
Reqzii = R n (4.35)
R
r is the sub conductor radius

a
R= (4.36)
2 sin ( π / n )

k = correction factor
dij = distance between the center of the bundles i and j
hi = average conductor i height (height at mid span plus 1/3 of the sag)
f = frequency
ρ = soil resistivity in Ω m
The terms of the potential matrix H in (km/μF) are:
4  Electrical Design 57

 D 
H ii = 17.976 ln  ii  (4.37)
 Reqc 
 ii 
 Dij 
H ij = 17.976 ln   (4.38)
d
 ij 

nr
Reqcii = R n (4.39)
R

Dij = distance from bundle i to the image of bundle j.

The inverse H−1 is the bus admittance matrix Y divided by w and includes the line
capacitances.
Therefore for the line parameters calculation the tower geometry has to be
known.
Notes:

• Shield wire may be grounded (ΔVs = 0), and their rows and columns can then be
eliminated (Gauss’s elimination) as shown by the equation below, and hence
their effects are included in the others lines and rows.

Z ik Z kj
Z ijnew = Z ij − (4.40)
Z kk

• for k = 4,5 and i,j = 1,2,3 in the example above.

• For isolated shield wire (Is = 0) their lines and columns are deleted.
• For asymmetrical bundle every sub-conductor has one line and one column in the
matrix Z for instance. As the sub-conductors in the same bundle have the same
voltage drop the lines and columns of one is maintained, and the others lines and
columns are substituted by the difference of their values and the corresponding
of the remained lines and columns. Now for the modified sub-conductors ΔV = 0,
It = Σ Ic, and can be eliminated like the shield wires, and their effect is kept in the
remaining one (Dommel 1986).
• If the line has phase transpositions the terms of the matrixes Z, H can be averaged
by its section length.
• Finally matrices Z, H remain with the number of lines/columns equal to the num-
ber of phases.
• For DC line the same applies being the remaining lines/columns equal to the
number of poles.
• The Electromagnetic Transients Programs that are available have routines to per-
form the necessary calculations.

4.1.7.2  Line Models

The equations indicated before, for a short line of length L are:

[ ∆V ] = [ Zu ][ I ] = [ Z ][ I ] (4.41)
58 J.F. Nolasco et al.

[I ] = jw [ H ] [V ]
−1
(4.42)

• Symmetrical components (AC lines)

Z and Y matrixes terms are all non zeros, and ΔV, V, I are phase quantities. To sim-
plify the calculation the equations above may be transformed for instance, Z, into
symmetrical components (positive, negative and zero sequences, or 1, 2, 0) by:

[ ∆V012 ] = [T ] [ Z ][T ][ I 012 ]

−1
(4.43)

Hence the symmetrical component impedance matrix is:

[ Z 012 ] = [T ] [ Z ][T ]
−1
(4.44)

If the line has a complete transposition of phases in equal sections then the symme-
trical component matrix Z012 has only the diagonal terms (the sequential impedan-
ces, Zo, Z1, Z2)
Now for the calculation, given one set of phase values, they are transformed into
symmetrical components, and the calculation is carried using the equation above.
After that, the calculated sequence components values have to be changed back to
phase components. The transformation matrix T is:

1 1 1 
1 
T= 1 a a 2  (4.45)
3
1 a 2 a 

With a = ej120.
Note that the phase components are:

 Z aa Z ab Z ac 
Z =  Z ba Z bb Z bc  (4.46)
 Z ca Z cb zcc 

The matrix Z is symmetric (Ex.: Zab = Zba). Also, if the phases have the same confi-
guration and they have complete phase transposition, then:

Zs Zm Zm 
Z =  Z m Zs Z m  (4.47)
 Z m Zm Z s 

Z= Z= Z cc = Z s (4.48)
aa bb

Z ab = Z ac = Z bc = Z m (4.49)

And

Z1 = Z 2 = Z s − Z m (4.50)

4  Electrical Design 59

Z 0 = Z s + 2 Z m (4.51)

Note: If the line is a double circuit (w,y) then Z matrix can be partitioned into four
3 × 3 sub-matrices Zww; Zyy; Zwy and the same equations can be applicable, provided
there is a complete transposition, obtaining the sequence self parameters of circuit
w, y and the mutual wy.
Therefore for two circuits w and y close together considering complete transpo-
sition the matrix has the following type.

Zws Zwm Zwm Zwys Zwym Zwym

Zws Zwm Zwym ZwyS Zwym
Zws Zwym Zwym Zwys
Zys Zym Zym
Zys Zym
Zys

And the self impedances of circuit w are:

Z w1 = Z w 2 = Z ws − Z wm
(4.52)
Z w0 = Z ws + 2 Z wm

The mutual impedances of circuit w and y are:

Z wy1 = Z wy 2 = Z wys − Z wym

(4.53)
Z wy 0 = Z wys + 2 Z wym

Similar considerations apply to the second equation and Y matrix. As example, for
single circuit being:

1
C1 = (4.54)
Hs − Hm

1
C0 = (4.55)
H s + 2Hm

Once obtained the sequence impedances the line/cable can be modeled using lum-
ped circuits, –π sections like in Figure 4.3.
For short lines (≤50 km) the above impedances are obtained by multiplying the
unit impedance with the line length. For long lines a factor < 1 shall be included as
indicated before.

a Z1 b Z0

C1 C1 C0 C0
2 2 2 2

Positive sequence impedance Zero sequence impedance

Figure 4.3  AC line model, single phase π for positive/negative sequences a) and zero sequence b).
60 J.F. Nolasco et al.

Z1

C0 C0 1 (C -C )
2 2 2 1 0

1
(Z -Z )
3 0 1

Figure 4.4  Three phase, π-model.

A three phase model can also be established and was useful for TNA calculation
(Figure 4.4).
For DC lines, similarly to the symmetrical component transformation, the matrix
T is:

1 1 1 
T= 1 −1 (4.56)
2  

That leads to two modes: ground and aerial or metallic modes (0 and 1
respectively).
Similarly for the AC equations, it results:

Z1 = Z11 − Z12 (4.57)

Z 0 = Z11 + Z12 (4.58)

1
C1 = (4.59)
Hs − Hm

1
C1 = (4.60)
Hs + Hm

The corresponding π circuit for aerial mode is (Figure 4.5):

The two poles model is that of Figure 4.6.
• For electromagnetic calculations the following equations applies

[∂V / ∂x ] = [ Z ][ I ] (4.61)

Figure 4.5 Aerial/ Z1
Metallic mode circuit; for
ground mode change Z1, C1 C1 C1
by Z0, C0. 2 2

Metallic mode impedance

4  Electrical Design 61

Z1

C0 C0 1 (C -C )
2 2 2 1 0

1 (Z -Z )
2 0 1

Figure 4.6  Two poles circuit.

[∂I / ∂x ] = jω[ H ] [V ] = [Y ][V ]

−1
(4.62)

and

∂ V / ∂x  = [ Z ][Y ][V ]

2 2
(4.63)

Now it is necessary to transform into diagonal matrix the product [Z][Y]. This is
obtained by searching for the eigenvalues and eigenvector of it (modal analysis)
(Dommel 1986).

4.1.7.3  Electrical Studies and Their Line Models

The following studies are applicable for AC and DC systems.

• Steady-state (load flow and short circuit)

• Electromechanical transients (stability, power frequency overvoltage due to load
rejection, and line/cable energization/reclosing)
• Electromagnetic transient (transient part of short-circuits, switching surge over-
voltages, DC converters commutation failure).
• Harmonic performance
• Relay protection and control coordination.

The overhead transmission line models may vary depending on the study.

• Steady state.
For load-flow calculation, the overhead AC lines and cables are represented by
-π sections based on positive sequence parameters. Overhead DC lines are repre-
sented by their resistance only.
For short circuit calculations transposed symmetrical components parameters
should be used for AC lines. For DC lines this type of calculation is not comple-
tely valid unless the converter station control action is simulated.
• Electromechanical transients
62 J.F. Nolasco et al.

AC and DC lines are represented using the same above mentioned modelling for
load flow analysis. However, for power-frequency over-voltages (rated fre-
quency ± 10%) the variation of the parameters (reactance: ωL and 1/ωC) with the
frequency shall be simulated (note that this does not refer to line frequency
dependence of parameters).
• Electromagnetic transients
AC and DC lines are represented by distributed parameter or a cascade of -π
sections. Now, overhead DC lines model considers frequency dependant
­parameters. Modal decomposition analysis is normally used in the calculation
together with non transposed Z, H−1, matrixes. This line model applies when
calculating the initial transient of a short circuit in DC and AC lines.
• Harmonic performance
AC and DC overhead lines are represented as -π sections using transposed para-
meters at the specific harmonic frequency taking into account the parameters as
a function of frequency.
• Relay protection and control
The same modeling used for electromagnetic studies applies in this case when
the transient part of the overcurrent/overvoltage phenomena is important.

4.1.7.4  Examples of Calculation DC Line

Calculations were done using ATP/EMTP-RV for the line on Figure 4.7 and are
reported below. The data for the example (arbitrarily chosen (Cigré 2009)).

• voltage = ± 500  kV
• pole conductor = 3 × 1590 MCM (~806  mm2)
• economic conductor for 1300 MW bipole
• diameter = 3.822  cm
• pole spacing = 13  m
• sub-conductor spacing = 45  cm
• minimum distance conductor to ground = 12.5 m
• conductor sag = 20.5 m (conductor height at tower 33 m)
• shield wires = EHS 3/8”
• shied wires spacing = 11 m
• shield wire sag = 20.5 m
• shield wire height at tower = 41 m
• soil resistivity = 500 Ω m.

The results using EMTP-RV are shown below (Figures 4.8, and 4.9):
After bundling (left) and elimination of the shield wires (right) the matrixes are
(Figures 4.10, 4.11, and 4.12):
or
I+ = ω 1.12E-08 V+ + ω (−1.69E-09) V−
ω = 2π f = 1 as f = 1/2π
C0 = 1.12E-08 − 1.69E-09 = > 9.51  μF
C1 = 1.12E-08 + 1.69E-09 = > 12.89  μF
Therefore
4  Electrical Design 63

pole spacing (13 m)

shield wire spacing (11 m)

shield wires

shield wire height

at tower
conductor
(41 m)
height
tower
(33 m)
shield wire height
guy wires at mid span
(20,5 m)

lattice
tower
(steel)

conductor height
at mid span
(12,5 m)

guy wire
foundation

tower foundation
(concrete)

Figure 4.7  Line geometry.

Z0 = (1.21 E-02 + 1.58 E-04) + j (2.49 E-03 + 1.59 E-03) = 0.0122 + j 0.00408 Ω

(ω = 1  rad/s)
Z1 = (1.21 E-02 − 1.58 E-04) + j (2.49 E-03 − 1.59 E-03) = 0.0120 + j 0.0009 Ω
(ω = 1  rad/s).

AC Line
Calculations were done using ATP/EMTP-RV for the line on Figure 4.13 and are
reported below. The data for the example are.

• voltage = 500  kV
• phase bundle conductor = 4 × 954 MCM (~483 mm2)
• diameter = 2.961  cm
• sub-conductor spacing = 45.7  cm.
 64

Figure 4.8  Input data of line geometry.

J.F. Nolasco et al.
Figure 4.9  Susceptance matrix, in units of [mhos/km] for the system of physical conductors.
Rows and columns proceed in the same order as the sorted input.

Figure 4.10  Susceptance matrix, in units of [mhos/km] for the system of equivalent phase con-
ductors. Rows and columns proceed in the same order as the sorted input.

Figure 4.11  Impedance matrix, in units of [Ω/km] for the system of physical conductors. Rows
and columns proceed in the same order as the sorted input.
66 J.F. Nolasco et al.

Figure 4.12  Impedance matrix, in units of [Ω/km] for the system of equivalent phase conductors.
Rows and columns proceed in the same order as the sorted input.

Figure 4.13  AC 500 kV 12.4 m

line.

5.5 m
45.7 m
0.8 m

30 m
sag 18 m 400 m
sag 12 m

• phase spacing = 11  m
• conductor height at tower = 30.0 and 30.8 m
• minimum distance conductor to ground = 12.0 m
• conductor sag = 18.0  m
• shield wires EHS = 3/8”.
4  Electrical Design 67

• shied wires spacing = 24.8 m

• shield wire height at tower = 40.0 m shield wire sag 12.0 m
• soil resistivity = 500 Ω m.
68 J.F. Nolasco et al.

4.2  urge Impedance and Surge Impedance Loading

S
(Natural Power)

The energy stored in the electric field of an overhead line can be represented as

1
Ee = CV 2 (4.64)
2
At a similar way, the energy stored in the magnetic field is:

1 2
Em = LI (4.65)
2
At the threshold condition of having electric energy equal to the magnetic energy
stored in both fields, that is as if Ee = Em, it results from the equations above (neglec-
ting resistance):

V L
= = Z0 (4.66)
I C

The ratio above has dimensions of an impedance and is called surge impedance of
the line. It can further be deduced

L
Z0
= = X L XC (4.67)
C

The surge impedance of a transmission line is also called the characteristic impe-
dance with resistance set equal to zero (i.e., R is assumed small compared with the
inductive reactance
The power which flows in a lossless transmission line terminated in a resistive
load equal to line’s surge impedance is denoted as the surge impedance loading
(SIL) of the line, being also called natural power.
Under these conditions, the sending end voltage ES leads the receiving end vol-
tage ER by an angle δ corresponding to the travel time of the line.
For a three-phase line:

Vϕϕ2
SIL = (4.68)
Zc

Where Vφφ is the phase-to-phase voltage and Zc is the surge impedance of the line.
Since Zc has no reactive component, there is no reactive power in the line,
Ql + Qc = 0. This indicates that for SIL the reactive losses in the line inductance
are exactly offset by the reactive power supplied by the shunt capacitance, or

I 2 ωL = V 2 ωC (4.69)
SIL is a useful measure of transmission line capability even for practical lines with
resistance, as it indicates a loading when the line reactive requirements are small.
For power transfer significantly above SIL, shunt capacitors may be needed to mini-
mize voltage drop along the line, while for transfer significantly below SIL, shunt
reactors may be needed.
4  Electrical Design 69

Table 4.3  Surge Impedance Loading of Typical Overhead Lines (MW)

N° of Conductors per phase bundle Z0 (Ω) Operating voltages (kV)
kV 69 138 230 345 500 765
1 400 12 48 132 298
2 320 60 165 372 781
3 280 893
4 240 1042 2438
6 162 1550 3613

4.2.1 Methods for Increasing SIL of Overhead Lines

An effort that has been made by electric industry nowadays has been directed toward
the goal of increasing the SIL of the overhead lines, especially considering the gro-
wing difficulties to acquire rights of way for new lines. For increasing the Surge
Impedance Loading of an overhead line, the following ways are possible

• Voltage increase
• Reduction of Z0 through one of the measures:
–– Reducing phase spacing (compaction)
–– Increasing number of conductors per phase bundle
–– Increasing conductor diameter
–– Increasing bundle radius
–– Introducing bundle expansion along the span but keeping the conventional
bundle spacing inside and near the tower.

Table 4.3 shows the surge impedance loading of typical overhead lines.

Table 4.3 is only illustrative of loading limits and is useful as an estimating tool.
Long lines tend to be stability-limited and give a lower loading limit than shorter
lines which tend to be voltage-drop or thermally (conductor ampacity)-limited.

4.2.2 Compact Lines

When compacted a Transmission line, the surge impedance loading can be increa-
sed. Compaction, in this case, consists of arranging the tower top geometry so that
the phases are as close as possible together. As defined by equations below, the SIL
reflects the interaction between line parameters, as follows:

V2
SIL = (4.70)
Z1

Z1 = Z s − Z m (4.71)

where:
SIL = Surge Impedance Loading (MW)
70 J.F. Nolasco et al.

6000 2800
11050 11050

2730 4400

2500

4800
4000
12990

21200
457 2450 2450 457 4500 4500

Figure 4.14  Compact Racket Tower 230 kV (left) and Compact Racket Tower 500 kV (right).

V = Operation voltage (kV)

Z1 = Positive sequence impedance (Ω)
Zs = Self impedance (Ω)
Zm = Mutual impedance (Ω).

The use of compact lines is one of the most effective methods for obtaining lines
with higher surge impedance loading or natural power figures. Reference (Fernandes
et al. 2008) shows interesting examples of 500 kV and 230 kV lines adopted by a
Utility in Brazil for having their first conventional self-supporting flat-configuration
towers which generated High SIL ratings. It was designed in the beginning of the
1980’s, but later a more recent development of compact lines was introduced into
their system. The big aim of the engineering team consisted in reducing the required
series compensation, by means of a high SIL of the lines. This represented a valuable
new tool for optimizing the new planned transmission systems
As a first real gain, the use of the compaction technology, associated with the
installation of series capacitor banks, could preclude for the transmission of
5000 MW the construction of two additional 500 kV – 800 km long each one –
transmission circuits; the adequate use of this technology could simultaneously
increase the energy transmission rate through the same corridor (MW/m2) and
improve the effectiveness of the costs associated thereof (MW/US$).
Consequently, the compaction results in an increase of the coupling between
phase conductors, so increasing the mutual impedance Zm and reducing the posi-
tive sequence impedance Z1, causing a net increase in the SIL of the line. Such
technology can provide a maximum increase of around 20 to 25% in SIL, as a
function of some limiting factors as: minimum viable phase spacing able to gua-
rantee adequate insulation coordination, asynchronous swing angles between
phase conductors, appropriate limitation of conductor surface gradient. Figure 4.14
shows a compact racket tower for 230 and 500 kV and Figure 4.15 shows a com-
pact cross-rope tower.
4  Electrical Design 71

Figure 4.15 Compact 13105 13105

Cross-rope Tower 500 kV.

3390
3175
4950
457
5500 5500

46345

H=22.0 m

H=37.0 m
6984

4.2.3 Bundle Expansion

This technique consists in designing the bundle radius R with higher value than
normal use. With that, the self impedance Zp is reduced therefore decreasing Z1 and
increasing the SIL. Similar effect is obtained by increasing the number of subcon-
ductors in the bundle, for the same total phase aluminum area.

4.3 Stability

A power system made up of interconnected dynamic elements may be said to have

stability if it will remain in stable operation following a system disturbance.

• Steady-state stability is associated with small perturbations such as slow varia-

tion on loads or generation. It depends fundamentally on the state of the system,
and on the operating conditions at the instant of the perturbation.
• Transient stability is associated with great perturbations (periodic disturbances),
such as line faults, loss of a generating unit, sudden application of a big load,
fault in equipment.
72 J.F. Nolasco et al.

It strongly depends on the magnitude and size of the perturbation and less on the
initial state of the system.
The stability limit is defined as:

V1V2
P= sin δ (4.72)
XL

Where P is the power in MW, V1 and V2 are the voltages at sending end and recei-
ving end terminals respectively; δ is the power angle of stability (between V1 and
V2).
As far as dynamic stability is concerned the power angle δ is limited to the range
30-45°, depending on the system, for the case of a generator connected to an infinite
bus, instead of a theoretical higher value near 90°, to allow stability to be kept fol-
lowing power oscillations resulting from perturbations.
The reduction of the series reactance XL is therefore considered by planning
engineers as a convenient alternative to increase the power transmitted by the line.

4.4 Thermal Limit and Voltage Drop

• As the conductor temperature increases, the following effects take place:

–– The ohmic resistance and therefore the losses increase.
–– The sags increase, reducing conductor-to-ground clearances or, conversely,
requiring higher towers.
–– As there is an increase in rating with the increase in conductor temperature, a
convenient and economic templating temperature should be chosen for every line.
–– As the conductor temperature reaches values higher than 90 °C (except for
HTLS conductor), there is a loss of its mechanical strength. The mechanical
strength reduction is cumulative with time and can cause sag increase and the
consequent reduction of conductor to ground clearances; due to safety rea-
sons a maximum value of 10% reduction in conductor UTS is usually accep-
ted along the line life.
Design temperature of a conductor is defined as the highest steady-state temperature
it can undergo under the worst (from a cooling viewpoint) meteorological conditions
(temperature, wind, solar radiation) and current. Regarding to the determination of
weather parameters for use in the case of deterministic ratings, see (Cigré TB 299).
It is usually a deterministic value. However, the determination of probabilistic
ratings is becoming more and more usual, as often significant savings can be achie-
ved. For more details, see (Stephen 1996) and also Chapter 7.
The actual recommended highest conductor (non HTLS) temperatures for line design
and spotting are 75 to 85 °C for steady-state operation and 100 to 150 °C (HTLS conduc-
tors excluded) for emergency operation. The line should be spotted considering such
temperatures and the relevant clearances to prevent the occurrence of safety problems.
It should be observed that new conductors (HTLS) recently developed or under
development stage can be operated continuously at temperatures until 150 °C to
200 °C or even more.
4  Electrical Design 73

Table 4.4  Example of Maximum Current Ratings (A) of Some ACSR Conductors and Bundles
used in Overhead Lines
Conditions Steady – State Emergency
Conductor Winter Summer Summer
N × Section (mm2) Code Day Night Day Night Day
170/28 Linnet 505 570 400 495 660
242/40 Hawk 625 715 490 620 825
322/52 Grosbeak 803 892 644 775 1055
403/29 Tern 840 965 647 840 1100
564/40 Bluejay 1030 1200 780 1045 1370
2 × 403/29 Tern 1680 1930 1290 1680 2200
483/34 Rail 957 1100 737 959 1275
2 × 483/34 Rail 1910 2200 1470 1915 2550
3 × 483/34 Rail 2870 3300 2210 2975 3825
4 × 483/34 Rail 3820 4400 2940 3830 5100
3 × 564/40 Bluejay 3090 3600 2340 3130 4100
Parameters adopted in Table above:
• Ambient temperature: winter: 20 °C
summer: 30 °C
• Wind speed: 1,0 m/s
• Latitude: 20°
• Solar radiation: winter: Day → 800  W/m2 Night → 0  W/m2
summer: Day → 1000  W/m2 Night → 0  W/m2 Conductor temperature: steady-state: 60 °C (current
indicated above) emergency: 100 °C (emergency current above)

Table  4.4 shows an example of thermal limits adopted by some utilities, for
steady-­state and emergency conditions in lines using ACSR conductors of more
widespread use.
Parameters adopted in table above:
• Ambient temperature: winter: 25 °C
summer: 30 °C
• Wind speed: 0.61 m/s
• Latitude: 20 °C
• Solar radiation: winter: 800 W/m2
summer: 1000 W/m2
• Conductor temperature: steady-state: 60 °C
emergency: 100 °C

• Voltage drop: Radial lines, especially medium and long lines, up-to 138 kV
have often their maximum transmitted powers limited by voltage drop or
regulation.
The highest limit practically recommended for the line voltage regulation is
around 10% for medium voltage lines and around 5% for EHV lines (230 kV and
above). Shunt reactive compensation (capacitors or reactors depending on the
SIL) are frequently required to reduce the voltage drop in certain cases.
74 J.F. Nolasco et al.

4.5 Capability of a Line

It is the degree of power that can be transmitted by a line as a function of its length,
considering the limitations imposed by voltage drop, stability and conductor tempe-
rature, as well as limitations inherent to substation terminal equipment, such as
circuit-breakers, current transformers etc.
The main factors determining the line capability in EHV lines are shown on Table 4.5.

4.6 Reactive Power Compensation

There are two basic types of compensation required by an electric system as a con-
sequence of the reactive power requirements, namely:

• Series Compensation, made up of capacitor banks connected in series with the

line, offsetting part of the inductive reactance (reduction of electrical length).
This compensation may be of fixed or variable value. Its main advantages are
following:
–– It improves the steady-state and transient stability
–– It allows a more economical power loading
–– It reduces the voltage drop
–– If a variable type of compensation is used it can be utilized to improve the
load distribution between circuits.

When using series compensation, especial attention should be given to other factors
affecting technically and economically the system such as, capacitor protection, line
protection and sub-synchronous resonance.

• Shunt compensation
The main shunt compensation schemes used in electric systems are:
–– Reactors, for long EHV lines for compensating line capacitive powers in
hours of light load (Ferranti Effect)
–– line connected reactors for line energization
–– Capacitors, for voltage control and power factor correction during hours of
higher demand load
–– Synchronous condensers (rarely used nowadays) that can perform the both fun-
ctions of reactors or of capacitors, depending on the instantaneous system needs.
–– Static compensators that perform the same function of the above synchronous
condensers, but have no moving parts.

Table 4.5 Determinant Line Length (km) Governing condition

factors on EHV line
0-80 Thermal limit
capability
80-320 Voltage drop
> 320 Stability
4  Electrical Design 75

4.7 Electromagnetic Unbalance - Transposition

Transpositions are made for the purpose of reducing the electrostatic and electroma-
gnetic unbalance among the phases which can result in unequal phase voltages for
long lines.
Untransposed lines can cause/increase in the following undesirable effects:
• Inductive interference with paralleling wire communication lines.
• Negative sequence currents that heat generator rotors.
• Zero sequence currents that can cause erroneous operation of protection relays.

For carrying out physically the phase transposition of the conductors, some alterna-
tives can be used such as making them in intermediate substations or near dead-­end
towers through especial conductor and insulator string arrangements or through the
utilization of special structures that allow changing phase positions by keeping the
necessary clearances to the towers and to earth.
Instead of performing phase transpositions, it is possible to adopt alternatives
that preclude them, such as:

• Use of delta or triangular phase configurations

• In rare cases.

4.8 Losses

The following types of losses have to be considered in overhead transmission lines.

4.8.1 Losses by Joule Heating Effect (RI2) in the Conductors

Those are the main losses that occur in the overhead conductors and their correct
selection and design are decisive for obtaining an economical line. Losses should be
seen as wastefull as they represent consumption of fuels or lowering of water reser-
voirs without the corresponding generation of useful work.
The power RI2 spent in the conductors and joints reduce the efficiency of the
electric system and its ability to supply new loads while the heat RI2Δt represents
burnt fuel or loss of useful water.

4.8.2 D
 ielectric Losses: Corona Losses, Insulator and
Hardware  Losses

By careful design and specifications of single or bundle conductors and accessories,

maximum conductor gradients may be limited so as to generate minimum Corona
losses under fair and foul weather conditions. Similarly a careful design of accesso-
ries and insulators can reduce to negligible values the amount of leakage currents
and the resulting losses.
76 J.F. Nolasco et al.

4.8.3 Losses by Induced Currents

The shield wires of the line are metallic conductors subjected to induced currents by
the line conductors and therefore producing losses. There are usually three alterna-
tives for reducing the shield wire losses, consisting basically in insulating them
from the towers so that only negligible currents can circulate through them:

• By insulating sections of the shield wires in the towers and just earthing an inter-
mediate point
• By totally Insulating the shield wires in the towers, i.e. not grounding them in
any point.

Certain utilities have shown that the shield wire insulation has sometimes caused
flashovers along the respective insulator, this is usually an insulator with a low
flashover capability as it must offer a free conductive path for lightning stroke cur-
rents. The current continues to flow in the shield wire until line is opened.
So, in the case the Utility decides to evaluate the economic feasibility of insula-
ting the shield wires for reducing line losses, a compromise must be found between
the savings in losses and the additional cost of insulating and maintaining the shield
wires and insulators.

4.9 Reliability and Availability

Consideration of the two important aspects of continuity and quality of supply,

together with other relevant elements in the planning, design, control, operation and
maintenance of an electric power system network, is usually designated as reliabi-
lity assessment.
Generally the past performance of a system is calculated according to some per-
formance indices.
SAIFI-System Average Interruption Frequency Index; SAIDI-System Average
Interruption Duration Index.
For the transmission lines, the unavailability is measured in terms of hours per
year or percent of time while the lines have been out.
Two considerations are more usual, namely:

• Mechanical Unavailability of the weakest component (towers), equal to the

inverse of twice the Return Period of the Design Wind Velocity, as per Reference
(Nolasco et al. 2002). The unavailability of all other components together usually
doesn’t exceed 25% of the one for the towers.
• Electrical Unavailability, considered equal to the unsuccessful reclosing opera-
tion when a lightning flashover occurs. Generally 65 to 70% of the reclosing
operations are successful. Such faults are usually caused by lightning strokes that
reach the conductors, towers or shield wires. An index that is generally used for
measuring an overhead line. Bush firing may create a similar problem.
4  Electrical Design 77

• Performance in the last case is the number of outages/100 km/year. The time

used for line maintenance (not live) is also part of the index.
• Additionally adverse weather conditions can add about for instance 0.3 events
per year with an average duration below 10 hours in general.

4.10 Overvoltages

The AC system overvoltages stresses are the input of the insulation coordination
study for the design of clearances and of the insulator string of transmission line.
The overvoltages can be classified as:

• Sustained voltages: continued power frequency voltages originated from system

operation under normal conditions; and temporary sustained overvoltages origi-
nating from switching operations such as load rejection, energization and reso-
nance conditions.
• Slow front overvoltages (switching surges): due to faults and switching operations
• Fast front overvoltages: originated mainly from lightning strikes or certain types
of switching
• Very fast front overvoltages: mainly related with gas insulated substation equip-
ment switching.

4.10.1 Fast-front Overvoltages (Lightning Overvoltages)

An important aspect to be considered in overhead transmission lines is their light-

ning performance.
Usually, the lightning performance criterion to be considered in the project of a
line or in the performance evaluation of an existing line is the maximum number of
flashovers, due to lightning, that can occur in the line per 100 km per year.
As the transmission line nominal voltage increases, the overvoltages generated
by lightning becomes less important to the specification of its insulation. This is due
to the increase in importance of other overvoltages such as switching surges.
Examples of lightning performance of real lines are shown on the Table 4.6. As
expected, the flashover rate caused by lightning is greater for lines with the lower
nominal voltages.
Lightning strokes to ground near a line or directly on it (on its conductors, towers
or ground wires) can generate high over-voltages that cause flashover in their insu-
lation and, consequently, the outage of the line.
Even though it is not the objective of the present item the detailed discussion of light-
ning phenomenon and the results of studies and researches developed to understand its
various aspects (that can be found in Cigré TB 549, 2013), a summary of its most
important parameters to the design of an overhead transmission line is presented.
To evaluate the lightning performance of transmission lines it is necessary to
considerer many additional aspects, primarily those related to the attachment
78 J.F. Nolasco et al.

Table 4.6  Examples of transmission line lightning performance (Anderson, 1975)

Lightning performance
Nominal Voltage (kV) (Flashovers/100 km-Year)
11-22 20.3
42 21.9
88 11.9
132 5.0
275 1.9
400 0.6
500 0.5
765 0.3

process of lightning channel to them, the electromagnetic surges generated in

the line when impulse currents are injected on them and the overvoltages with-
stand by their insulation.

4.10.1.1  Lightning Discharge Parameters

The primary lightning parameters are described, in Cigré TB 549, and is summari-
zed here to emphasize the primary aspects relevant to a usual transmission line
design (Cigré TB 549).
Lightning can be defined as a transient, high-current (typically tens of kA) elec-
tric discharge in air whose length is measured in kilometers. The lightning discharge
in its entirety, whether it strikes ground or not, is usually termed a “lightning flash”
or just a “flash.” A lightning discharge that involves an object on ground or in the
atmosphere is sometimes referred to as a “lightning strike”. The terms “stroke” or
“component stroke” apply only to components of cloud-to-ground discharges. Most
lightning flashes are composed of multiple strokes. All strokes other than the “first”
are referred to as “subsequent” strokes.
Each lightning stroke is composed of a downward-moving process, termed a “lea-
der”, and an upward-moving process, termed a “return stroke”. The leader creates a
conducting path between the cloud charge source region and ground and distributes
electric charge from the cloud source along this path, and the return stroke traverses
that path moving from ground toward the cloud charge source and neutralizes the lea-
der charge. Thus, both leader and return stroke processes serve to effectively transport
electric charge of the same polarity (positive or negative) from the cloud to ground.
The kA-scale impulsive component of the current in a return stroke is often fol-
lowed by a “continuing current” which has a magnitude of tens to hundreds of
amperes and a duration up to hundreds of milliseconds. Continuing currents with
duration in excess of 40 ms are traditionally termed “long continuing currents”.
These usually occur in subsequent strokes.
The global lightning flash rate is some tens to a hundred flashes per second or so.
The majority of lightning flashes, about three-quarters, do not involve ground.
These are termed cloud flashes (discharges) and sometimes are referred to as ICs.
Cloud discharges include intra cloud, inter cloud, and cloud-to-air discharges.
4  Electrical Design 79

Lightning discharges between cloud and earth are termed cloud-to-ground dischar-
ges and sometimes referred to as CGs. The latter constitute about 25% of global
lightning activity.
From the observed polarity of the charge lowered to ground and the direction of
propagation of the initial leader, four different types of lightning discharges bet-
ween cloud and earth have been identified: (a) downward negative lightning (b)
upward negative lightning (c) downward positive lightning, and (d) upward positive
lightning. Downward flashes exhibit downward branching, while upward flashes are
branched upward.
It is believed that downward negative lightning flashes (type a) account for about
90% or more of global cloud-to-ground lightning, and that 10% or less of cloud-to-­
ground discharges are downward positive lightning flashes (type c). Upward light-
ning discharges (types b and d) are thought to occur only from tall objects (higher
than 100 m or so) or from objects of moderate height located on mountain tops.
As noted above, positive lightning discharges are relatively rare (less than 10%
of global cloud-to-ground lightning activity). Positive lightning is typically more
energetic and potentially more destructive than negative lightning.
Sometimes both positive and negative charges are transferred to ground during the
same flash. Such flashes are referred to as bipolar. Bipolar lightning discharges are
usually initiated from tall objects (are of-upward type). It appears that positive and nega-
tive charge sources in the cloud are tapped by different upward branches of the lightning
channel. Downward bipolar lightning discharges do exist, but appear to be rare.
The ground flash density Ng(flashes/km2/yr) is often viewed as the primary
descriptor of lightning incidence. Ground flash density has been estimated from
records of lightning flash counters (LFCs) and lightning locating systems (LLSs) and
can potentially be estimated from records of satellite-based optical or radio-­frequency
radiation detectors. It is worth noting that satellite detectors cannot distinguish bet-
ween cloud and ground discharges and, hence, in order to obtain Ng maps from satel-
lite observations, a spatial distribution of the fraction of discharges to ground relative
to the total number of lightning discharges is needed. IEEE Std 1410-2010 recom-
mends, in the absence of ground-based measurements of Ng, to assume that Ng is
equal to one-third of the total flash density (including both cloud and ground dischar-
ges) based on satellite observations (IEEE Standard 1410-2010).
If no measurements of the ground flash density Ng for the area in question are
available, this parameter can be roughly estimated from the annual number of thun-
derstorm days Td, also called the keraunic level. Apparently the most reliable
expression relating Ng and Td is the one proposed by (Anderson et al. 1984):

N g = 0.04Td1.25 (4.73)

Another characteristic of lightning activity that can be used for the estimation of Ng
is the annual number of thunderstorm hours TH. The relation between Ng and TH
proposed by (MacGorman et al. 1984) is:

N g = 0.054Th1.1 (4.74)

80 J.F. Nolasco et al.

A typical negative cloud-to-ground flash is composed of 3 to 5 strokes (leader/return

stroke sequences), with the geometric mean inter-stroke interval being about 60 ms.
Occasionally, two leader/return stroke sequences occur in the same lightning chan-
nel with a time interval between them as short as 1 ms or less.
The observed percentage of single-stroke flashes, based on accurate-stroke-count
studies is about 20% or less, which is considerably lower than 45% presently recom-
mended by Cigré.
First-stroke current peaks are typically a factor of 2 to 3 larger than subsequent-­
stroke current peaks. However, about one third of cloud-to-ground flashes contain at
least one subsequent stroke with electric field peak, and, by theory, current peak,
greater than the first-stroke peak.
Traditional lightning parameters needed in engineering applications include
lightning peak current, maximum current derivative, average current rate of rise,
current rise time, current duration, charge transfer, and action integral (specific
energy), all derivable from direct current measurements.
Essentially all national and international lightning protection standards (IEEE
Standard 1410; IEEE Std 1243; IEC 62305), include a statistical distribution of peak
currents for first strokes in negative lightning flashes (including single-stroke flashes).
This distribution, which is one of the cornerstones of most lightning protection stu-
dies, is largely based on direct lightning current measurements conducted in
Switzerland from 1963 to 1971 (Anderson and Eriksson 1980). The cumulative stati-
stical distributions of lightning peak currents for negative first strokes, negative subse-
quent strokes and positive first strokes are presented in Figures 4.16, 4.17, and 4.18.

Figure 4.16 Cumulative %
statistical distributions of
lightning peak currents, 99
giving percent of cases
exceeding abscissa value, 95
from direct measurements 1
in Switzerland.
80

50

2
20
3

100 101 102 kΑ

1 - Negative f irst strokes
2 - Negative subsequent strokes
3 - Positive first strokes
4  Electrical Design 81

Figure 4.17 Cumulative %
statistical distributions of
crest time, giving percent 99
of cases exceeding abscissa
value, from direct 95
measurements in
Switzerland.
80
3

50

2
20
1

10-1 100 101 μS

1 - Negative f irst strokes
2 - Negative subsequent strokes
3 - Positive first strokes

Figure 4.18 Cumulative %
statistical distributions of
current rate of rise, giving 99
percent of cases exceeding
abscissa value, from direct 95
measurements in
Switzerland (Berger et al. 2
1975). 80
1

3
50

20

101 100 101 kΑ/μS

1 - Negative f irst strokes
2 - Negative subsequent strokes
3 - Positive first strokes
82 J.F. Nolasco et al.

The distributions are assumed to be log-normal and give percent of cases exceeding
abscissa value.
The log-normal probability density function for peak current I is given by:

f (I ) =
1
e
( 2)
− z
2

(4.75)
2I
Where:

ln I − Mean ( ln I )
z= (4.76)
β
and ln I is the natural logarithm of I, Mean (ln I) is the mean value of ln I, and β = σlnI
is the standard deviation of ln I.
A log-normal distribution is completely described by two parameters, the median
and logarithmic standard deviation of the variable. Logarithmic standard deviations
of lightning peak currents are often given for base 10.

P(I ) = ∫
∞
1
e
( 2 ) d
− z
2

(4.77)
I 2
Only a few percent of negative first strokes exceed 100 kA, while about 20% of
positive strokes have been observed to do so. About 95% of negative first strokes are
expected to exceed 14 kA, 50% exceed 30 kA, and 5% exceed 80 kA. The corres-
ponding values for negative subsequent strokes are 4.6, 12, and 30 kA, and 4.6, 35,
and 250 kA for positive strokes. Subsequent strokes are typically less severe in
terms of peak current and therefore often neglected in lightning protection studies.
Slightly more than 5% of lightning peak currents exceed 100 kA, when positive and
negative first strokes are combined.
Berger’s peak current distribution for negative first strokes shown in Figure 4.18
is based on about 100 direct current measurements. The minimum peak current
value included in Berger’s distributions is 2 kA.
In lightning protection standards, in order to increase the sample size, Berger’s
data are often supplemented by limited direct current measurements in South Africa
and by less accurate indirect lightning current measurements obtained (in different
countries) using magnetic links. There are two main distributions of lightning peak
currents for negative first strokes adopted by lightning protection standards: the
IEEE distribution (IEEE Standard 1410; IEEE Std 1243; Cigré WG 33-04). Both
these “global distributions” are presented in Figure 4.19.
For the Cigré distribution, 98% of peak currents exceed 4 kA, 80% exceed 20
kA, and 5% exceed 90 kA.
For the IEEE distribution, the “probability to exceed” values are given by the
following equation:

1
P(I ) = 2.6
(4.78)
 I 
1+  
 31 
4  Electrical Design 83

Figure 4.19 Cumulative
statistical distributions of
peak currents (percent
values on the vertical axis
should be subtracted from
100% to obtain the
probability to exceed.

Table 4.7  Peak current distributions adopted by IEEE

Peak current, I, kA 5 10 20 40 60 80 100 200
Percentage exceeding First strokes 99 95 76 34 15 7.8 4.5 0.78
tabulated value, P(t) 100% Subsequent 91 62 20 3.7 1.3 0.59 0.33 0.050
strokes

where P(I) is in per unit and I is in kA. This equation, usually assumed to be applica-
ble to negative first strokes, is based on data for 624 strokes analyzed by (Popolansky
1972), whose sample included both positive and negative strokes, as well as strokes
in the upward direction. This equation applies to values of I up to 200 kA. Values of
P(I) for I varying from 5 to 200 kA, computed using the previous equation are given
in Table 4.7. The median (50%) peak current value is equal to 31 kA.
In the range of 10 to 100 kA that is well supported by experimental data, the
IEEE and Cigré distributions are very close to each other (IEEE Standard 1410).
The peak-current distribution for subsequent strokes adopted is given by:

1
P (I ) = 2.7
(4.79)
 I 
1+  
 12 

84 J.F. Nolasco et al.

T10/90

T30/90
(μs) Time
IΤRΙΟ
S10
I10

I30

S10/90
k(A)
S30/90

I90
II
I100
IF

Sm

Figure 4.20  Description of lightning current waveform parameters. The waveform corresponds
to the typical negative first return stroke. Adapted from Cigré TB 63 and IEEE Std 1410-2010.

Cigré recommends for negative subsequent stroke peak currents a log-normal distri-
bution with the median of 12.3 kA and β = 0.53 (Cigré WG 33-04), which is also
included in IEEE Std 1410–2010.
In Cigré TB 549, it is discussed what it called “global” distribution of peak cur-
rent found in most lightning protection standards. Concern is expressed about using
imprecise or not homogeneous data (lumped or not in a single sample with data
considered more reliable). In this document, recent distributions of lightning peak
currents obtained from many individual studies are presented and compared.
A representative double-peaked current waveform of negative first strokes is pre-
sented in Figure 4.20, with the definition of its front parameters.
Table 4.8 are lists the values of the lightning current parameters of Figure 4.20
recommended by Cigré and IEEE.

4.10.1.2  Equivalent Impedance of the Lightning Channel

Lightning-channel impedance is an important parameter that can influence the cur-
rent injected into the object subjected to a strike.

Direct-strike Effects
Lightning is approximated by a Norton equivalent circuit. This representation inclu-
des an ideal current source equal to the lightning current that would be injected into
the ground if that ground were perfectly conducting (a short-circuit current, Isc) in
parallel with a lightning-channel impedance Zch assumed to be constant. In the case
when the strike object can be represented by lumped grounding impedance, Zgr, this
impedance is a load connected in parallel with the lightning Norton equivalent
(Figure 4.21). Thus, the “short-circuit” lightning current Isc effectively splits bet-
ween Zgr and Zch so that the current flowing from the lightning-channel base into the
4  Electrical Design 85

Table 4.8  Lightning current parameters (based on Berger’s data) recommended by Cigré and
IEEE
Parameter Description
I10 10% intercept along the stroke current waveshape
I30 30% intercept along the stroke current waveshape
I90 90% intercept along the stroke current waveshape
I100 = II Initial peak of current
IF Final (global) peak of current (same as peak current
without an adjective)
T10/90 Time between I10 and I90 intercepts on the wavefront
T30/90 Time between I30 and I90 intercepts on the wavefront
S10 Instantaneous rate-of-rise of current at I10
S10/90 Average steepness (through I10 and I90 intercepts)
S30/90 Average steepness (through I30 and I90 intercepts)
Sm Maximum rate-of-rise of current along wavefront,
typically at I90
td 10/90 Equivalent linear wavefront duration derived from IF/S10/90
td 30/90 Equivalent linear wavefront duration derived from IF/S30/90
Tm Equivalent linear waveform duration derived from IF/Sm
QI Impulse charge (time integral of current)

a Source

Igr

Isc=V0 /Zch Zch Zgr

Reference ground

b Source ρtop ρbot

TL representing
Isc =V0 /Zch Zch tall object (Zab ) Zgr

Reference ground

Figure 4.21  Engineering models of lightning strikes (a) to lumped grounding impedance and (b)
to a tall grounded object.

ground is found as Igr = Isc Zch/(Zch + Zgr). Both source characteristics, Isc and Zch, vary
from stroke to stroke, and Zch is a function of channel current, the latter nonlinearity
being in violation of the linearity requirement necessary for obtaining the Norton
equivalent circuit. Nevertheless, Zch, which is usually referred to as equivalent impe-
dance of the lightning channel, is assumed to be constant.
86 J.F. Nolasco et al.

Lightning-Induced Effects
In studying lightning-induced effects, the distribution of current along the lightning
channel is needed for computing electric and magnetic fields (Baba and Rakov).

Equivalent Impedance
The limited estimates of the equivalent impedance of lightning channel from expe-
rimental data suggest values ranging from several hundred Ω to a few kΩ.

4.10.1.3  Protection of Power Transmission Lines - Concepts

Lightning strokes can cause insulation flashover when they strike the conductors,
ground wires or even the soil nearby the transmission lines.

• Flashover caused by Induced surges.

Lightning striking to soil nearby a transmission line can induce surge overvoltages
on it. Most measurements of induced voltage have been less than 300 kV.
This level of overvoltage can cause flashover in medium voltage lines, but usually
is not a concern to high voltage transmission line.
• Flashover caused by direct strokes to conductors.
When a lightning strikes a conductor of a transmission line, a high impulse overvol-
tage is developed between the conductor and tower (in the insulator strings or air
gaps), the conductor and other phase conductors or the conductor and ground. These
impulse overvoltages can cause flashover in the line. As the insulation strings are,
usually, the elements with the lowest impulse insulation level, they are the element
with the greatest probability of occurrence of flashover.
The peak of the impulse overvoltage generated by a direct stroke in a conductor
with a surge impedance Z can be estimated, approximately, by:

Z I peak
Vsurge ≅ (4.80)
2
where Ipeak is the peak current of stroke.
Considering a surge impedance Z of 400 Ω, it is easy to see that even a low discharge
current of 10 kA can generate very high overvoltages in the conductor (2 MV).
So, when a line with high performance is desired, it is necessary to provide some
protection to reduce the probability of direct strokes to the conductors that exceed
the insulation level of the line.

• Flashover caused by direct strokes to shield wire or tower.

Even installing ground wires in a line, they cannot eliminate the probability of
flashover in the line caused by lightning.
High impulse overvoltage can still be generated, especially in the presence of a
large peak current.
Related to the lightning stroke hitting the ground wires, as the impulse impe-
dance seen from the point of incidence of the stroke is not low (it is depends on the
surge impedance of various elements: ground wires, tower, grounding system,
4  Electrical Design 87

length of span, etc.), the ground wires voltage can reach very high values. The tower
top voltage rises too.
Then, consequently, the insulation of the line is stressed by the large voltage
generated between the tower or ground wires and the conductors. If this voltage is
high enough, a flashover can occur. This flashover is called back flashover, as it
tends to occur from the grounded elements (tower or ground wire) to the energized
phase conductors.

Induced Surges
Induced surge by nearby lightning discharge is not a concern to high voltage trans-
mission lines.
In medium voltage lines, some measures can be implemented to improve the
performance of the line in respect to flashover caused by nearby strokes.

Direct Strokes to Conductors

Lightning strokes, with relatively high peak currents, directly to phase conductors
can generate very high overvoltages on them, which can cause a line outage in case
of flashover.
To reduce the probability of occurrence of such high overvoltages, the most com-
mon measure is to install shield wires on the lines. Other measure is the installation
of surge arresters. In respect to the installation of shield wires, in a specific line, it is
necessary to make a shielding analysis to determine the number and the position
relative to the phase conductors.
In both (Cigré WG 33-04) and (IEEE Std 1243), the so-called electro-geometric
model (EGM) is employed.
The basic concepts involved in this model will be explained using Figure 4.22.
Several researchers have contributed to the electro geometric model (EGM). As
the downward leader approaches the earth, a point of discrimination is reached for a
final leader step. The EGM portrays this concept with the use of striking distances.

Figure 4.22  Electrogeometric model representation of conductors and ground wires.

88 J.F. Nolasco et al.

The striking distance is of the form

rc , g = AI b (4.81)

Where A and b are constants that depend on the object and I is the stroke current.
Local electric field gradients around conductors are somewhat higher than at
ground level, so rc is usually considered to be greater than rg (the striking distance
to ground), resulting in rc ≥ rg. Arcs of circles with the radii rc are drawn centered
at the phase conductor and OHGW. A horizontal line is then drawn at a distance
rg from earth.
If a downward leader, having a prospective current I, for which the arcs were
drawn, touches the arcs between B and C (Figure 4.22), the leader will strike the
phase conductor. If the leader touches the arcs between B’s, it will strike the shield
wire. If all leaders are considered vertical, the exposure distance for a shielding
failure is Dc.
Since the final jump length in the EGM depends on current, the statistics of the
stroke-current distribution will be needed to compute the number of lightning stro-
kes to phase conductors (that depend on Dc).
At present, the following striking distance equations are recommended by IEEE
(IEEE Std 1243):

rc = 10 I 0.65 (4.82)

 3.6 + 1.7 ln ( 43 − yc )  I 0.65 yc < 40 m
rg =   (4.83)
 5.5 I 0.65 yc ≥ 40 m

where I is the stroke current (in kA) and yc is the average conductor height, given by
the height at the tower minus two-thirds of the sag.
Some researchers of EGM assume all striking distances are equal, while others
consider different striking distances to phase conductors, shield wires, and earth. In
addition, some researchers do not use a striking distance to earth. Estimates of stri-
king distance sometimes differ by a factor of two. However, this uncertainty has not
prevented the design and operation of lines with low lightning outage rates. In par-
ticular, when an engineering judgment is made to accept a low but non-zero shiel-
ding failure flashover rate (SFFOR), most models suggest similar shielding angles.
Figure 4.22 indicates an apparent possibility of perfect shielding: the possibility
to install ground wires in a position that makes Dc null for lightning stroke currents
greater than a minimal current necessary to cause flashover when it strikes directly
the phase conductors (called critical current Ic).

Strokes to Shield Wire or Towers

High impulse overvoltage can be generated in a transmission line when high inten-
sity lightning strikes its ground wires or towers. If the overvoltage stressing the
insulation of the line is greater than the voltage that it can withstand, a flashover
occurs (in this case, called back-flashover).
4  Electrical Design 89

To reduce the probability of occurrence of back flashover to an acceptable level,

in the design of a transmission line, one important aspect to be considered is the
appropriate design of structure grounding systems. It should be considered the
necessary value of resistance to achieve the desired lightning performance of the
line, but also the fact that the transient response of grounding system cannot be
expressed only by its resistance to low frequency and low amplitude currents. For
example, a long counterpoise cable can have relatively low resistance to industrial
frequency currents, but high impulse impedance (as will be discussed latter. Usually,
a number of parallel cables is better than a long counterpoise.
In areas of large flash density and high electrical resistivity of soil, sometimes it
is necessary to reduce the resistance to a level that is not possible technically or
economically. In these cases, one of the most efficient measures is the installation of
surge arresters in the line.
To identify which measures needed to be implemented, it is necessary to evaluate
the lightning performance of the line with and without those measures, even if some
approximation should be done. The evaluation of lightning performance of trans-
mission lines is then discussed.

4.10.1.4  E valuating the Lightning Performance of a Power

Transmission Line
To estimate the lightning performance of OHTL the following primary aspects
should be considered (information also in 4.10.1.1):

• Ground flash density along the line

• Lightning current parameters (primarily its peak distribution)
• Lightning stroke to the transmission line and to its individual components
• Estimate of insulation stress when lightning strikes the line
• Flashover strength of insulation to the over-voltage
• Estimate the rate of insulation flashover due to lightning striking directly the
conductors (shielding flashover) or due to backflashover phenomenon (lighting
strokes to ground wires or towers).

In terms of calculation, the fourth aspect in the above list is the most complex,
because it involves the estimation of the transient response of relative complex ele-
ments that are interconnected: conductors and ground wires (depending on the cur-
rent front of wave), towers and grounding systems. Usually, many simplifications
are done to reduce the complexity of calculation and enable the use of simple com-
putational routines in the lightning performance calculation.
Knowing the current that can reach a component of the line, the comparison of
the results of overvoltage stress with the flashover strength of insulation will indi-
cate if the flashover will occur.
In the final the lightning performance of an overhead transmission line can be
calculated. Basically, knowing the currents that can reach the line and cause a
flashover and its probability of occurrence, it should be determined:
90 J.F. Nolasco et al.

• the shielding failure flashover rate (relative to the lightning strokes directly to
conductors);
• the back flashover rate (relative to the lightning strokes on the ground wires or
towers);
• overall flashover rate (the sum of the two previous rates).

These rates usually are expressed as number of flashovers per 100 km of line per
year.
Two proposed specific procedures for estimating the lighting performance of
transmission lines are described in the documents (IEEE Std 1243; Cigré WG
33-04). In the following items the primary aspects involved in the estimation proce-
dures of lightning performance of transmission lines, as recommended by IEEE and
Cigré, are characterized, keeping in mind the practical design objective.

Ground Flash Density

The ground flash density Ng can be roughly estimated from the annual number of
thunderstorm days TD, by the equations shown in 4.10.1.1.

Lightning Current Parameter Considered

Anderson and Eriksson (1980) noted that the two sub-distributions (below and
above 20 kA) can be viewed as corresponding to the shielding failure and back-
flashover regimes, respectively. A single distribution, also shown in Figure 4.19,
was adopted by IEEE guidelines consider a triangular (2 μs/50 μs) implemented in
the software “Flash”.
For the IEEE distribution, the “probability to exceed” value of peak currents
from 2 kA to 200 kA are given by the following equation:

1
P(I ) = 2.6
(4.84)
 I 
1+  

 31 

where P(I) is in per unit and I is in kA.
Cigré guidelines consider a concave front current as shown in Figure 4.20, with
parameters listed in Tables 4.8 and 4.9.
The log-normal probability density function for peak current I is given by:

f (I ) =
1
e
( 2)
− z
2

(4.85)
2I
The probability for peak current to exceed a specified value I is given by:
∞
P(I ) = ∫
1
e
( 2)
− z
2

d (4.86)
T 2
4  Electrical Design 91

Table 4.9  Lightning current parameters (based on Berger’s data) recommended by Cigré and
IEEE
First stroke Subsequent stroke
β, logarithmic (base β, logarithmic (base
Parameter M, Median e) standard deviation M, Median e) standard deviation
Front time (μs)
td10/90 = t10/90/0.8 5.63 0.576 0.75 0.921
td30/90 = t30/90/0.6 3.83 0.553 0.67 1.013
tm = IF/Sm 1.28 0.611 0.308 0.708
Steepnes (kA/μs)
Sm, Maximum 24.3 0.599 39.9 0.852
S10, at 10% 2.6 0.921 18.9 1.404
S10/90, 10-90% 5.0 0.645 15.4 0.944
Peak (Crest) current (A)
II, initial 27.7 0.461 11.8 0.530
IF, final 31.1 0.484 12.3 0.530
Ratio, II/IF 0.9 0.230 0.9 0.207
Other relevant
parameters
Tail time to half value 77.5 0.577 30.2 0.933
th (μs)
Number of strokes per 1 0 2.4 0.96 based on
flash median
Ntotal = 3.4
Stroke charge, QI 4.65 0.882 0.938 0.882
(Coulomb)
0.057 1.373 0.0055 1.366
∫I dt ( kA ) s
2 2

Interstroke interval (ms) – – 35 1.066

Lightning to the Transmission Line

Number of Lightning Strokes that Hit the Line

IEEE guidelines use the same expression as Cigré to evaluate the number of
lightning strokes the hits a transmission line:

Ng
Nl =
10
+b ( 28h

0.6
) (4.87)

where Ng is the ground flash density (flashes/km2/yr), h is the tower height (m) and
b is the ground wires separation distance between (m).
Lightning Strokes that Hit the Phase Conductors
In IEEE and Cigré procedures, for a line with ground wires, the number of light-
ning strokes that hit directly the phase conductors are expressed as shielding failure
rate (SFFOR), calculated by:
92 J.F. Nolasco et al.

I max

SFFOR = 2 N g L ∫ D ( I ) f ( I )dI
c (4.88)
3

where:
L = length of the line (km)
Dc(I) = exposure length (m) relative to phase conductor, calculated in function of I
f(I) = statistical distribution of I
Imax = maximum current (kA) that can hit the phase conductor (current that makes
null the distance Dc(I)).
The lower limit, 3 kA, recognizes that there is a lower limit to the stroke
current.

Strength of Insulation
To identify if a flashover will occur on an insulator string stressed by an overvoltage
generated by a lightning stroke that hits a line, IEEE evaluate the voltage necessary
to cause a flashover in an insulator string with the following equations:

 710 
VD =  400 + 0.75  l ( 0.5µs ≤ t ≤ 16 µs ) (4.89)
 t 
where VD is the impulse flashover voltage in kV, t is the time to flashover in μs and
l is the insulator string length in m.
For t greater than 16 μs, IEEE recommends the use of 490 kV/m as CFO of insu-
lator strings.
Among other methods that could be used to evaluate the voltage necessary to
cause a flashover in an insulator string (Cigré WG 33-04), Cigré recommends the
use of a leader propagation model, where the leader propagation velocity is cal-
culated by:

 u (t ) 
v (t ) = K Lu (t )  − Eo  (4.90)
 d −l 
 g l 
where:
v(t) = leader velocity (in m/s)
u(t) = voltage applied to the insulator string (in kV)
Eo = electric field needed to begin the leader considered (kV/m)
dg = length (in m) of insulator strings orair (at instance t = 0)
ll = leader length (in m) at an instance t
KL = constant.
For positive surges in air gaps or insulator strings, Cigré recommends the use of
Eoas 600 kV/m and KL as 0.8 × 10−6. For negative surges, it is recommended Eo as
670 kV/m and KL as 1 × 10−6.
When voltage/time curve for standard 1.2/50 μs lightning impulse is known, the
best fitting constants may also be determined by numerical calculations for selected
combinations of flashover and time to breakdown.
4  Electrical Design 93

Estimate the Rate of Insulation Flashover

Shielding Failure Flashover Rate
Shielding failure occurs when a lightning stroke hits directly a phase conductor
of a transmission line that has shield wires. When such failure results in flashover,
ininsulator strings or in air gaps between conductor and metallic grounded compo-
nents, it is said that a shielding failure flashover occurred.
According to IEEE, the minimal or critical current Ic required to cause a flasho-
ver can be calculated as follows:

2 ⋅ CFO
Ic = (4.91)
Z surge

Z surge = 60 ln ( 2 h / r ) ln ( 2 h / Rc ) (4.92)

where
Zsurge = conductor surge impedance under Corona (Ohms)
h = average conductor height (m)
r = conductor radius (m)
RC = Corona radius of the conductor under a gradient of 1500 kV/m (m)
CFO = critical flashover voltage (kV), negative polarity, as defined in Item.
According to Cigré procedure, Ic can be calculated by a similar procedure or by one
that considers a more precise transient response of the line (using a electromagnetic
transients program, like EMTP-Electromagnetic Transients Program) and the same
or other processes of line critical flashover voltage estimation, such as:

• insulation voltage/time curve;

• integration method;
• physical models representing the Corona phase, the streamer propagation
• phase and leader phases along the line insulation.

IEEE and Cigré estimate the shielding failure rate (number of lightning strokes
directly to the phase conductors that cause flashover) as:
I max

SFFOR = 2 N g L ∫ D ( I ) f ( I ) dI
Ic
c (4.93)

where
SFFOR = shielding failure flashover rate (flashovers/100 km/yr)
L = length of the line (km)
Dc(I) = exposure length (m) relative to phase conductor, calculated in function of I
f(I) = statistical distribution of I
Imax = maximum current (kA) that can hit the phase conductor (current that makes
null the distance Dc(I)
94 J.F. Nolasco et al.

The probability that an individual subsequent stroke current Is will exceed Ic is

given approximately by:

1
P ( Ic > I s ) = 2.7
(4.94)
 I 
1+  c 
 I subs 
where
Isubs is taken as 12 kA;
Ic is also taken in kA.
The following equation gives PS, the probability of flashover on any subsequent
stroke, given that no flashover occurs on the previous strokes:

( )
n =∞
n −1
Ps = ∑Pn 1 −  P ( I s > I c )  (4.95)

n=2

where Pn is the probability that there are n strokes/flash, from data in (Tompson
1980).
The total SFFOR will be the sum of the first stroke failure rate SFFOR and the
added rate SFFORS obtained from:
Ic

SFFORs = 2 N g LPs ∫ Dc ( I ) f1 ( I ) dI (4.96)

0

If the critical current Ic is low, most shielding failures will lead to flashover, either
from the small first stroke or from the 60-70% chance that there will be a subsequent
stroke that exceeds Ic. If the critical current is higher, PS from will be lower (PS = 0.4
for space Ic of 16 kA).
The extra contribution of subsequent stroke effects to total SFFOR ensures that
perfect shielding (SFFOR = 0) will rarely be achieved. See next item.
As cited in the previous item, the estimation of shielding failure rate considering
only the lighting first strokes (number of lighting first strokes directly to the phase
conductors that cause flashover) as:
I max

SFFOR = 2 N g L ∫ Dc ( I ) f ( I ) dI (4.97)
Ic

It indicates an apparent possibility of perfect shielding: a shielding angle that makes
Imax = Ic (maximum stroke current that can be injected directly to a phase conductor
equal to the current necessary to generate an overvoltage in the phase conductor
equal to the insulator withstand), but this can be rarely achieved as it can have a
contribution of subsequent stroke effects to total SFFOR.
Considering only lightning first strokes, Cigré procedure presents the following
equation to evaluate the shielding angle where Imax = Ic:

  rg − h  r − y 
−1  g
α p = 0.5 sin −1   + sin   (4.98)

  rc   rc  
4  Electrical Design 95

where
rg, rc = calculated for the current Ic (m)
h = average height of ground wire (m)
y = average height of phase conductor (m)
An attempt to achieve a perfect shielding angle may severely handicap an econo-
mical design of lines in areas of low flash density (Ng < 2 flashes/km2/yr). It is sug-
gested to the designer evaluating the most economical configuration based on the
SFFOR required. For example, serving a critical load, a design SFFOR value of 0.05
flashover/100 km/yr may be suitable, while values of 0.1-0.2 flashover/100 km/yr are
recommended for general practice.
Rate of Insulation Flashover due to Backflashover
When a lightning strikes the tower or the overhead ground wire, the current in the
tower and ground impedances cause the rise of the tower voltage. A considerable frac-
tion of the tower and shield wire voltage is coupled by mutual surge impedance to the
phase conductors. The tower and shield wire voltages are much larger than the phase
conductor voltages. If a voltage difference from phase to tower exceeds a critical value,
a flashover occurs, called “backflash” or “backflashover”. The corresponding mini-
mum lightning current that produces such a flashover is called “critical current”.
The calculation of the critical current for back-flashover depends, in general
order of sensitivity, on the following parameters:

• Amplitude of the lightning current (generally the peak value of the first return
stroke);
• Flashover criteria for the insulation and air gaps;
• Presence of surge arresters across some or all insulators;
• Surge impedance coupling among phases and ground wires, evaluated using
transmission line models and considering the additional coupling from arrester-­
protected insulators;
• Steepness (di/dt) at the peak of the current wave, which is generally assumed to
be the maximum di/dt;
• Waveshape, including both time to peak and time to half-peak value
• Footing impedance, influenced by high frequency and soil ionization effects;
• Tower inductance or surge impedance model;
• Representation of nearby towers and grounding systems;
• Representation of nearby power system components (e.g. transformers).

Sometimes, the induction effects of the electromagnetic field of the lightning chan-
nel are additionally taken into account for the estimation of the insulator voltage.
Induction effects related to current flow in the tower past the phase conductors have
been observed and modeled in several ways.
The procedure adopted by Cigré for the calculation of the line backflashover rate
(BFR), the same as described by (Aileman 1999), is specifically aimed at calcula-
ting the critical current and the resultant BFR value. In particular, the Cigré proce-
dure analytically estimates the backflashover critical current by making reference to
the representation of the travelling wave phenomena that take place for both cases
of a lightning strike to a tower or to an overhead ground wire.
96 J.F. Nolasco et al.

The BFR is given by the probability of exceeding the critical current multiplied
by the number of flashes to the shield wires (NL), taking into account that the crest
voltage and the flashover voltage are both functions of the time-to-crest (tf) of the
lightning current. Therefore, the BRF considering all the possible time-to-crest
values is:
∞ ∞
 1
BFR = 0.6 N l ∫ ∫ ( )
f   f t f dI dt f
t
0 I c (t f )  f  (4.99)

(flashovers / 100 km / yrr )

where f(I/tf) is the conditional probability density function of the stroke current
given the time-to-crest and f(tf) is the probability function of the time-to-crest value.
Note that, in order to obtain the BFR for strokes to the tower and for stroke to the
spans, the BFR obtained for strokes to the tower is multiplied by a coefficient, equal
to 0.6.
Another, more simplified procedure for the calculation of the BFR, is also pre-
sented in Cigré procedure as the BFR resulting from the application of previous
equation using of an equivalent time-to-crest value Te. Such a value is approxima-
tely the median value of time to crest for the specific critical current. With such a
value, since a single equivalent front is used, the BFR is reduced to:
∞
(4.100)
BFR = 0.6 N l ∫ f ( I ) dI = 0.6 N l P ( I > I c )
Ic

In these equations, Ic is the minimum current that leads to insulator backflash in the
phase conductor. To consider the system voltage at the striking time, this current can
be calculated considering that such voltage is approximated 80% of the nominal
voltage.
The approach adopted by the IEEE is based on the estimation of the voltage
across the line insulation at two specific time instants namely: a first evaluation of
the full impulse-voltage waveshape peak (at 2 μs) considering only the stricken
tower, and a second evaluation on the tail (at 6 μs) considering relevant adjacent
towers. In order to estimate the backflash critical current, these values are compared
with an estimation of the volt-time curve of the line insulation.
The backflashover rate is computed according to the equation:
Nc
BFR = N l ∑ ( t1 Pi ) (4.101)

Nc

where NC is the number of phase conductors and ti is the period of time in which
each phase is dominant. This concept is related to the system voltage at the different
phases when lightning strikes, as well as the different coupling factors between each
phase and the shield wire. Pi is the probability of the lightning current exceeding a
backflashover critical value. This is evaluated with respect to each phase, taking into
4  Electrical Design 97

account the phase shift between the sinusoidal voltages and the different coupling
factors between each phase and the ground wire.
Note that the procedures to calculate both rates, SFFOR and BFR, are based on
using the local ground flash density Ng to determine the number of strikes to the line.
To calculate the Ic, the minimum current that leads to insulator backflash in the
phase conductor, it is necessary to calculate the overvoltage in the insulation of the
line and compare with the withstand voltage. In the next items, the most important
aspects involved in this calculation are discussed.
The Cigré and IEEE procedures are compared in (Nucci 2010). The main differen-
ces, when present, lie in the fact that some approaches/methods proposed so far within
Cigré can be considered to be more general than those proposed within IEEE, in that
they take into account more variables of the problem. Within the IEEE – thanks in part
to the inherently simpler approach – a computer code, called FLASH, has been made
available, which can serve either as a professional tool capable of providing an approxi-
mate, yet very useful, answer on the lightning performance of typical overhead transmis-
sion lines or as a reference for beginner researchers when simple cases are dealt with.

4.10.1.5  E  stimate of Insulation Stress Generated by Lightning

Strokes in the Line
To estimate the overvoltages generated in the insulation of a transmission line by
lightning striking on its conductors, ground wires or towers, it is necessary to model
the primary components that are responsible for the transient response of the line.
In the following items the primary aspects involved in such modeling are discus-
sed. The presentation of all the equations involved in the calculation is beyond the
scope of the present text.

Tower Surge Response Model

In the evaluation of voltages generated at the top of tower during a lightning
dischar­ge, it is necessary to consider the response of the tower to electromagnetic
transients. Usually, the tower is modeled with distributed parameters, characterized
by a surge im­pe­dance associated with an electromagnetic wave travel time. In
Table 4.10 are listed equations that enable the evaluation of theses parameters for
some self-supporting towers.
Common experience with practical structures yields typical values for tower
surge impedances in the range 150-250 Ω.
For tower with guy wires, Cigré presents a simplified approach where the mutual
coupling of guy-wires is not taken into account.
Basically, the process is:

• Evaluate the guy-wire surge impedance and travel time;

• Evaluate the equivalent inductance of all guy-wires;
• Evaluate the inductance parallel of guy-wires inductance with the tower inductance;
• Evaluate the equivalent surge impedance and travel time.
98 J.F. Nolasco et al.

Table 4.10  Tower model (impedance and travel time) (IEEE Std 1243-1997)
Cylindrical Conical

  2h    
Z = 60  ln  2  − 1 h2 + r 2
  r   Z = 60 ln  2 
 r2 
 
h
τ= h
0.85c τ=
0.85c
Waist H-Frame

  h 
Z1 = 60  ln  2 2  − 1 
  r 
π   tan −1 ( r / h )    h
60 d ln  2  + h Z1
Z= 60  ln  cot  − ln 2 
4   2 Z2 =  r
 
h+d
r1h2 + r2 h + r3 ( h − h2 ) ZZ
r= Z= 1 2
h Z1 + Z 2
h
τ= 1 hZ1 ( d + h ) Z 2
0.85c τ=
cZ hZ1 + ( d + h ) Z 3

Note (1): For tower of conical type, IEEE uses h/(0.85.c) as travel time instead of h/c indicated by
Cigré Document 63 (1991).
4  Electrical Design 99

To evaluate the parameters cited the following equations can be used:

Z guy = 60  ln ( 2 h / r ) − 1 (4.102)

τ guy = lguy / c (4.103)

Lguy = Lguy τ guy (4.104)

Lequiv _ guy _ wires = Lguy / n (4.105)

where:
h = guy-wire height (m)
r = guy-wire radius (m)
lguy = guy-wire length (m)
c = light velocity
n = number of parallel guy-wires
Lguy = inductance of a guy-wire (H)
Lequiv_guy-wires = inductance of n guy-wires (H).
Finally, for item (d), the following equations can be used:

Ltower = Z tower τtower (4.106)

Lequiv _ guy _ wires Ltower
Lequiv _ tower + guy _ wires = (4.107)
Lequiv _ guy _ wires + Ltower

cLequiv _ tower + guy _ wires
Z equiv _ tower + guy _ wires = (4.108)
HT

τequiv _ tower + guy _ wires = H T / c (4.109)

where:
Ztower = surge impedance of tower only (Ω)
τtower = travel time in the tower only (s)
Ltower = inductance of tower only (H)
HT = tower height (m)
Lequiv_tower+guy-wires = inductance of the tower and guy wires (H).
It is important to note that several approaches have been presented in the recent
literature addressing tower models.

Tower Footing Resistance

In the IEEE and Cigré procedures the tower grounding system behavior is characte-
rized by a lumped resistance (the tower footing resistance).
100 J.F. Nolasco et al.

In IEEE procedure this resistance is constant. In Cigré procedure, the effect of

soil ionization is taken into account, using the following equation when the light-
ning current amplitude exceeds the critical value Ig:

R0
Ri = (4.110)
I
1+
Ig

where:
R0 = is the low frequency non-ionized soil resistance;
Ig = is the critical value of the lightning current.
Ig = is estimated considering the soil ionization threshold field Eg, using the
equation:

Eg 
Ig = (4.111)
2 R02

Where:
ρ = is the electrical resistivity of the soil (Ω m)
Eg = is the soil ionization threshold field, considered to be, approximately,
400 kV/m for most common soils (Cigré TB 63, 1991).
In both evaluation procedures, of IEEE and Cigré, except the soil ionization, no
reference is made explicit to the transient response of structure grounding systems
or to the fact that the soil parameters vary with frequency.
Relative to these aspects, many researchers have been done. For example, the
main factors that influence the grounding behavior have been analyzed in (Visacro
and Alípio 2012), the current-dependent response of electrodes is addressed in
(Sekioka et al. 2005) and the effect of the frequency-dependent soil parameters on
this response is addressed in (Visacro and Alípio 2012).

Transmission Line Modeling

In both procedures, of IEEE and Cigré, the effects of occurrence of Corona in
ground wires are considered.
The surge impedance of each conductor or ground wire and the mutual impe-
dance between them are calculated considering as infinite the electrical conductivity
of soil and the cables in their mean height.
In IEEE procedure the wave travel time in the phase conductors and ground
wires is calculated considering an electromagnetic traveling wave velocity as 90%
of velocity of light.
To evaluate the voltage on the top of a tower where a lightning strikes, usually it
is not necessary to model more than three spans and towers on each side of the tower.

4.10.1.6  Improving Lightning Performance

As discussed in more details in IEEE Std 1243-1997, the following special methods,
among others, can improve lightning performance of a line:
4  Electrical Design 101

• Installation of additional ground wires under conductors:

Basically used to increase the common-mode coupling of voltage surges on the
ground wires to the phase conductors, and cause a reduction on the insulator
voltage at the tower.
• Installation of guy wire on the towers:
Fitting new or additional guy wires from tower to rock or soil anchors can reduce
the tower surge impedance and the grounding resistance (the latter because new
guy anchor will behave as an additional ground electrode).
• Ground wires in separated structures:
OHGWs may be supported by separate outboard towers or poles instead of being
assembled on the same structure that supports the phase conductors. This arran-
gement may give extreme negative shielding angles,which minimize induction
losses and provide excellent security from shielding failures.
Tower height and wind loading may also be reduced. While an expensive option,
OHGWs on separate structures may result inexcellent lightning performance.
Connections can be made from the OHGWs to towers, if required for ac
­fault-­current management, should be designed to have a high impedance to light-
ning through long interconnection length to minimize risk of backflash over.
• Installation of surge arresters:
With the installation of surge arresters in parallel to the insulator strings the over-
voltage on them will be reduced to acceptable levels.
The number of surge arresters can be optimized, i.e., it is not necessary to install
them in all towers and in all phases.
The use of surge arresters is covered by Cigré TB 440 (Cigré TB 440).

4.10.1.7  Grounding
Grounding systems are installed in the structures of a transmission line with the
following primary objectives:

• to provide a preferential path to earth for currents generated by faults in the line;
• to provide a grounding system with a resistance low enough to enable the over-
current protection to detect a ground fault in the line.
• to provide a preferential path to earth to lightning discharge currents;
• in urban areas, to control the step and touch voltages generated during ground
faults in the line;
• to reduce the structure ground potential rise during a lightning discharge and,
consequently, reduce the probability of occurrence of backflashover on the line.

In the following items, the primary practical aspects involved in the design of the
grounding system of the transmission line structures are discussed.

Measuring the Electrical Resistivity of Soil

To design a grounding system it is necessary to measure the electrical resistivity of
the soil where it will be installed.
As the soil resistivity may vary considerably over the surface and depth, it is
necessary to perform measurements at various locations throughout the area
102 J.F. Nolasco et al.

occupied by the grounding system using a process that enables the identification of
the variation of the resistivity with depth.
To design the grounding systems of transmission lines structures, measurements
should be done with electrodes driven along an axis coincident or near the axis of
the line and centered at the installation point of each structure. In addition to this
axis, some companies specify measurements on axes near the edges of the right-of-­
way of the line. In this case, measurements made on three axes: one in the center of
the right-of-way and the other two near their limits. The final resistivity considered
for each distance “a” (between measuring probes) is the mean value of all the mea-
surements done with that distance, except that ones those have great discrepancies
from the mean value, which are neglected.
One of the most widely used methods of measuring electrical resistivity of soil is
the Wenner four-pin method (Dawalibi and Barbeito 1991).

Soil Stratification
Usually, modeling the soil with a model of stratified horizontal layers, where each
layer has a specific resistivity and thickness, is used in grounding system design. This
can be done as most of the real soils are not homogeneous, but composed of several
layers of different electrical resistivity and thickness. These layers, due to the geolo-
gical formation, in general, are fairly horizontal and parallel to the ground surface.
From the results of resistivity measurement, it is possible to find the parameters
of the model (of a soil stratified in two or more horizontal layers).
Considering a two-layer soil model (Figure 4.23), its structure can be characterized by:

• a first layer with resistivity ρ1 and thickness d1

• a second layer with resistivity ρ2 and infinite thickness.

The stratification of the soil can be carried out by a curve fitting process, where ρ1,
ρ2 and d1 are determined.
As an example, it will be shown the results of a stratification process done with
a specific set of measured resistivity values obtained with the Wenner four-pin
method, that are presented in Table 4.11.
The two-layer soil parameters are shown Table 4.11. In Figure 4.24 are shown
the measured values and the curve ρ over a generated with the values of ρ1, ρ2 and
d1 shown on Table 4.11.

Figure 4.23  Structure of

a soil stratified in two-layer
ρ1
of different resistivities. d1

ρ2 d2= ∞
4  Electrical Design 103

Table 4.11  Measured value of apparent resistivity of soil obtained with the Wenner four-pin
method
Resistivity in Ω m for distance a between electrodes
a 2 m 4 m 8 m 16 m 32 m
ρa 1405 1173 743 553 549
Two-layer soil stratification
ρ1 (Ω m) ρ2 (Ω m) d1 (m)
1515 525 3.2

1.500
1.400
1.300
1.200
1.100
Resistivity (Ωm)

1.000
900
800
700
600
500
400
300

2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38
Distance between adjacent electrodes (m)

Figure 4.24  Measured resistivity for distance a and curve calculated with the parameters of the
two-layer soil structure obtained in the stratification process.

In this example, the deviations between measured and calculated values of resis-
tivity are small, indicating that the real soil can effectively be approximated by a
two-layer soil model.
In many practical situations, the soils cannot be perfectly stratified in two layers
as the one shown here. In such cases, conservative approximations should be done
or multi-layer layer soil model should be used.

Resistance Calculation
The resistance of a grounding system can be estimated knowing its geometry and
the resistivity of the soil where it will be installed (Table 4.12).
Usually, simplified equations are used to calculate the resistance of single electrodes
or simple grounding systems. In most case, they consider uniform soil with a resistivity
called the apparent resistivity. Such apparent resistivity can be evaluated, approxima-
tely, from the stratification of the soil and the dimensions of the grounding system.
Reference (Heppe 1979) presents the equations necessary to calculate the induc-
tion coefficients and the potential generated in each point of the soil by the currents
104 J.F. Nolasco et al.

Table 4.12  Examples of simplified equations that can be used to calculate the resistance of elec-
trodes installed in a uniform soil of resistivity ρ (IEEE Std 142)
Electrode Grounding Resistance
Vertical
ground rod ρ  4L 
R= ln − 1
2 π L  a 
Single
counterpoise
ρ  2L 
R=  ln − 1
πL  2ad 
Three point
star
ρ  2L 2L s s2 s4 
R=  ln + ln + 1.071 − 0.209 + 0.238 2 − 0.054 4 …
6πL  a s L L L 
Four point
star
ρ  2L 2L s s2 s4 
R=  ln + ln + 2.912 − 1.071 + 0.645 2 − 0.145 4 …
8πL  a s L L L 
Six point star
ρ  2L 2L s s2 s4 
R=  ln + ln + 6.851 − 3.128 + 1.758 2 − 0.490 4 …
126πL  a s L L L 
Eight point
star
ρ  2L 2L s s2 s4 
R=  ln + ln + 10.98 − 5.51 + 3.26 2 − 1.17 4 …
16πL  a s L L L 

Dimensions: Rod or wire radius → a ; Length → L ; Depth → d = s/2

injected through the segments of conductors in which the grounding system was
subdivided. Theses equations were derived considering a two-layer soil.
To calculate the grounding system resistance, it is assumed that all segments of con-
ductor are metallically interconnected and the calculations done at power frequency.
Then, it can be assumed that all the segments are at the same potential Vm. For an
arbitrary value of Vm, for example 1.0 V, the current injected in earth by each segment of
conductor can be calculated. Then, the grounding resistance can be calculated as:

Vm
Rgrounding = (4.112)
∑Ii
i =1

To calculate the potentials generated in the soil during the occurrence of ground
fault in the transmission line, it is necessary to estimate the ground potential rise of
the grounding system Vm, in the desired situation, and with it to calculate the real
currents that will be injected into the soil.

Resistance and Impedance

For low frequency currents, the behavior of typical grounding systems that are
installed in structures of transmission lines can be characterized by a resistance.
4  Electrical Design 105

For greater frequencies, especially the frequencies that are present in lightning
currents (ranging from 100 Hz to 4 MHz), the capacitance and inductance of groun-
ding systems are significant in their behavior (Visacro et al. 2011).
For such high frequencies, the relation between voltage (ground potential rise)
and current injected in the grounding system cannot be characterized by a constant
(the resistance). To be more precise, this relation should be described as an impe-
dance that varies with frequency:

V ( ω)
Z ( ω) = (4.113)
I ( ω)

The variation of the soil resistivity and permittivity with frequency is another
important aspect to be considered when high precision is required, especially for
grounding systems installed in high resistivity soils.
In the literature, some expressions are presented to describe the variation of
resistivity and permittivity with frequency. They are curve-fitting expressions that
are based on experimental results.
The following expressions were proposed in (Visacro and Alípio 2012):

{
ρ = ρ0 / 1 + 1.2 E −6 ρ00.73 ( f − 100 )
0.65
} (4.114)

ε r = 7.6 E 3 f −0.4 + 1.3 (4.115)

where ρo is the soil resistivity at 100 Hz, ρ (in Ω,m) and εr are the soil resistivity and
relative permittivity at frequency f (in Hz), respectively. The equation of ρ is valid
for frequencies between 100 Hz and 4 MHz, while the equation of εr is valid for
frequencies between 10 kHz and 4 MHz (below 10 kHz, it is suggested to use the
value of relative permittivity calculated at 10 kHz).
In Figure 4.25, the impedance Z(ω) is shown for a counterpoise of 50 m in length,
buried in 2500 Ω m soil. It considers the frequency variation of the soil parameters.
As can be seen, immediately after power frequency, Z(ω) reduces with increasing
frequencies. For even greater frequencies, it increases again, until exceeding the low
frequency resistance value.

Low frequency
+45°
resistance
(Ω)

Z (ω) Angle

120 0
|Z (ω)|

80 -45°
Minimum
Impedance
40
101 102 103 104 105 10 Frequency (Hz)

Figure 4.25  Impedance of a counterpoise of 50 m buried in a 2500 Ωm soil. Continuous line:
modulus of impedance. Dotted line: impedance angles.
106 J.F. Nolasco et al.

Soil: εr=10

2000 Ω m
50
ZP (Ω)

1000 Ω m

500 Ω m
10

10 20 30 40 50 Electrode Length L (m)

Figure 4.26  Calculated Impedance of a counterpoise buried in a uniform soil. For each soil the
effective length is indicated.

Table 4.13  Effective counterpoise length Leff (Visacro 2007)

Resistivity of Soil (Ω m) Leff (m) for fast current waves 1.2/50 μs
100 14
500 23
1000 34
2000 50

For impulse waves, primarily in lightning analysis, the behavior of a grounding

system is, usually, described by its impulse grounding impedance Zp, defined as the
ratio between the voltage and current peaks developed at the current injection point:

Vp
Zp = (4.116)
Ip

For a specific grounding system, the impulse grounding impedance Zp is a function
of resistivity of soil and the current waveform, primarily its front-time parameter.
Figure 4.26 shows calculated curves Zp over L, where Zp is the impulse impe-
dance of a counterpoise of length L installed in a uniform soil. In this same figure is
indicated the effective length of counterpoise, defined as that length beyond it Zp no
longer reduces which the increase of L. These curves were calculated considering
the soil parameters constant with frequency.
Table 4.13 shows values of effective counterpoise length for current waves of
1.2/50 μs (Visacro 2007). For current waves of 4.5/60 μs, the effective length for
these same cases are, approximately, 40% greater than for 1.2/50 μs.
The results of Figure 4.26 and Table 4.13 were obtained considering constant the
soil parameters ρ and ε; Soil ionization were not considered. In recent works, it was
identified lower values of Zp and greater values of effective length, primarily for
high soil resistivity (as 3000 Ω m) and for impulse current waves representative to
lightning first strokes wave.
Even though the precision of the values of effective length can be an object of
discussion it clearly shows that too long counterpoises should not be used with the
4  Electrical Design 107

objective of reducing its impulse impedance. Using a continuous counterpoise, for

example, is clearly a mistake.
As a design criterion, it is suggested not to install counterpoise with length much
greater than the effective length calculated with the 4.5/60 μs current wave form.
One interesting parameter that can be used to analyze the impulse behavior of a
grounding system is the impulse coefficient, defined as the relation between the
impulse impedance Zp and the low frequency resistance RLF of the grounding sys-
tem, usually called Ic.
Soil ionization occurs primarily when high current is discharged into concentra-
ted electrodes. Usually the ionization process starts when the electric field in the soil
reaches a critical value (approximately 300 kV/m for typical soils (Mousa 1994)),
and tends to reduce the ground system resistance.
In large grounding systems, as the ones used in a high voltage transmission line
constructed in high resistivity soil, which can be composed by long counterpoises,
reduction in their resistance by soil ionization occurs only for very high lightning
currents injected on them.
Although some studies have been done and simplified methodologies have been
proposed to consider the soil ionization in grounding system analyses (Mousa
1994), in practical power transmission line grounding system design, usually, the
soil ionization has not been considered explicitly.

Measuring the Structure Grounding Resistance

After the installation of the grounding system in a structure of a transmission line,
it is recommended to measure its resistance. In the following paragraphs, a basic
procedure that is widely used to do this measurement is described.
Some utilities have specification for this procedure.
In rural areas, usually, horizontal counterpoise wires, radially disposed from the
structures, with or without ground rods, are used as tower grounding system.
Usual variations done in the grounding system geometries are:

• installation of additional small wire or cables from tower (to reduce the groun-
ding system surge impedance)
• installation of a wire or cable in a form of rectangular ring around the tower base
or around guy-wires foundation
• preclusion of the cables that interconnect the guy-wires to the central mast, in
guyed tower
• installation of ground rods with or without the counterpoise cables
• installation of continuous counterpoise cables, i.e., interconnecting the counter-
poises of adjacent towers (this procedure is not recommended as reported before)
• installation of deep grounding well
• use of low resistivity materials in substitution of some portion of the local soil:
use of bentonite, for example.

Grounding systems with greater number of counterpoises or ground rods in parallel

from the points of connection to tower have lower impulse impedances.
108 J.F. Nolasco et al.

Determining the Ground System to Install in each Tower

As mentioned before, the geometry of the grounding system to install in a structure
depends on the value and distribution of the soil resistivity, the desired maximum
resistance to be obtained and the extension of the area available to install it.
When the transmission line has a large number of structures, it is common to
define basic geometries of grounding systems to be used. Usually, they are called
grounding stages as one geometry can be viewed as an extension of the previous
one. Then, the grounding stage to be installed in a specific structure is identified by
calculations or measurements.
In regions with soil of high resistivity where, even with the last stage of groun-
ding system it may occur that the desired maximum resistance is not reached. In this
case, the installation of special grounding system should be considered.
Depending on the soil resistivity, it can be considered the installation of a larger
number of horizontal radial counterpoises, ground rods, deep grounding well and
the use of low resistivity materials, like bentonite, for example.
In Figure 4.27 an example is shown of special grounding system geometry desi-
gned to be used in very high resistivity soils, with the objective of reduction of the
tower grounding impulse impedance.
In urban areas, the structures can be in regions with high traffic of people. In this
case, to guarantee the public safety, it may be necessary to design specific groun-
ding systems to control the touch and step voltage generated, primarily, during
faults on the transmission line.

Touch and Step Voltage Limits

As discussed in (IEEE Std 80-2000), the touch and step voltages generated at the groun-
ding system should not exceed the limits calculated with the following equations:

1m 1m

L3
45°

Rectangular ring
1m
OHTL axis OHTL axis

L2=0.6xL1

1m
L1
L3

1m 1m

Figure 4.27  Example of grounding system designed to reduce the grounding impulse impedance
of a tower.
4  Electrical Design 109

Vmax_ step = ( Rch + R2 Fs ) I ch (4.117)

Vmax_ touch = ( Rch + R2 Fp ) I ch (4.118)

where:
Rch = human body resistance (of order of 1000 Ω);
R2Fs = resistance of the two human feet in series;
R2Fp = resistance of the two human feet in parallel;
Ich = maximum allowable current in the human body.
The current Ich can be estimated as (for 50 kg person):

0.116
I ch = [ A] (4.119)
t
where t is the exposition time to the current.
The resistances R2Fs and R2Fp can be estimated by the = following equations:

R2 Fs = 6 Cs ρs (4.120)

R2 Fp = 1.5Cs ρs (4.121)

where ρs is the soil surface resistivity and Cs is a function of ρs, its thickness hs and
the resistivity of the soil immediately below ρs. In a natural soil ρs is equal to ρ1 and
Cs is equal to 1. If the natural soil is covered with a high resistivity material, as a
layer of gravel, asphalt or stones, ρs will be the resistivity of this material and Cs can
be calculated as:

 ρ 
0.09  1 − 1 
Cs = 1 −  ρs  (4.122)
2 hs + 0.09

With the installation of a layer of high resistivity material the step and touch voltage
generated by the grounding system can be greater, lowering its complexity and cost
or providing a greater safety margin. Another advantage is that it gives some protec-
tion to the grounding system against thieves and vandalism.
As an example, in Table 4.14 the step and touch voltage limits are shown for a
natural soil with resistivity 500 Ω m, with or without a thin layer of high resisti-
vity material: granite stones or asphalt (the installation of gravel is not recommen-
ded as it is easy to be stolen). The time of exposure t to the current was considered
equal to 1.0 s.
The step and touch voltages generated at the grounding system of a structure will
depend on the characteristics of the electrical system (basically its short-circuit current
and fault clearing time), the characteristics of the transmission line, the electrical resisti-
vity of soil and the geometry of the grounding system. In case of proximity of the struc-
ture with a substation, the influence of its grounding mat should be considered.
Typical geometries of grounding systems installed in urban areas are shown in
Figure 4.28. Basically, they are composed by ground rods and cables installed as
110 J.F. Nolasco et al.

Table 4.14  Step and touch voltage limits

Short duration step and touch voltage limits (V)
Natural soil (500 Ω m) Thin layer of granite stones (ρs : Thin layer of asphalt (ρs : 10000
5000 Ω m; thickness: 10 cm) Ω m; thickness: 10 cm)
Vtouch Vstep Vtouch Vstep Vtouch Vstep
203 464 742 2621 1073 3944
Note: Except for natural soil, all the other resistivities are for wet materials.

Figure 4.28 Typical Ground rod

grounding system
geometry for structures in
areas where it is necessary
to control the step and
touch voltages.

1.0 m

1.0 m
1.0 m

1.0 m

Four ground rods and three wires/cables installed as rectangular rings

around the base of a metallic structure at a depth of 0.5 m.

1.0 m
1.0 m

1.0 m

Four ground rods and wires/cables installed as circular rings around

the base of a concrete structure, at a depth of 0.5 m, except the last
ring, that is at 1.0 m.
4  Electrical Design 111

rectangular or circular rings, around the feet of the structure, with 1 m apart. The
installation depth of the outer ring may be greater in order to control the step voltage
in the border of the grounding rings.
In addition to the grounding system itself, usually it is necessary to install a thin
layer of high electrical resistivity material over the natural soil to increase the maxi-
mum allowable step and touch voltages and also to protect the grounding system.
Examples are parallelepipeds of granite with sides, at least, 10 cm in length or a
layer of asphalt, with a thickness of 5 cm.

Example of Design
As an example, in this item it is presented the results of the design of a grounding
system installed to control the step and touch voltages in a structure located in an
urban area. This is the grounding system of the 40th structure of a 138 kV transmis-
sion line.

• Transmission line data:

–– Nominal voltage = 138  kV
–– Length = 60  km
–– Typical span = 400  m
–– Number of structures = 149
–– Conductor = ACSR 176.9 MCM - Linnet
–– Ground wire = ACSR 101.8 MCM - Petrel
–– Typical tower = see the following figure
–– Average grounding resistance of structures = 15 Ω
–– Base of structure 40 = 5 m × 5 m
–– Distance between structure 40 and the substation at the beginning of
line = 16  km.
• Electrical system data:
–– Symmetrical ground fault current in both substations of the line = 15 kA
–– Total ground fault clearing time = 1 s
–– Resistance of the substation ground mats = 1 Ω
• Soil stratification in structure n° 40:
–– ρ1 = 500 Ωm
–– ρ2 = 1000 Ωm
–– d1 = 3  m.

The designed grounding system is shown in Figure 4.29. Its grounding resistance

was estimated in 29.3 Ω (Figure 4.30).
Also, it was recommended to cover the natural soil around the tower with granite
stones (parallelepiped) with sides, at least, 10 cm in length.
In the design process of tower 40 grounding system, it was necessary to calculate
the current distribution in the ground wire and towers of the line, for a ground fault
in tower 40 (Figure 4.31). The ground potential rise of tower 40 was estimated in
6.03 kV. The current distribution calculation was made with a software developed
112 J.F. Nolasco et al.

Figure 4.29 Typical T
tower of the line.

C C

Dimensions (mm)
T P C S f
800 3030 2900 1860 2350

specifically for this purpose (the ATP – Alternative Transients Program could also
be used). The resistance of the grounding system in design was considered in the
results shown here.
In points inside the covered area, the limits of step and touch voltages were esti-
mated considering:

• ρs = 5000 Ωm (wet granite stones)

• hs = 0.1  m
• ρ1 = 500 Ωm

With these parameters, Cs is equal to 0.72.

Then:
R2 Fs = 6.0 ⋅ 0.72 ⋅ 5000 = 21600 Ω
R2 Fp = 1.5 ⋅ 0.72 ⋅ 5000 = 5400 Ω

4  Electrical Design 113

Figure 4.30 Grounding
system of tower 40.

1.0 m

1.0 m

1.0 m

1.0 m
Legend
Tower base

Cables at a depth of 0.5 m

Cables at a depth of 1.0 m

Cables with a variable depth

I1 I2
GROUND WIRE GROUND WIRE
3.3 2.5
If

If1 If2
INICIAL FINAL
SUBSTATION SUBSTATION
4.0 1.9
0.206 I3

Figure 4.31  Current distribution near tower 40, in kA, for a ground fault on it.

and the limits are:

0,116
Vmax_ step = (1000 + 21600 ) = 2621 V
l
0,116
Vmax_ touch = (1000 + 5400 ) = 742 V
l
114 J.F. Nolasco et al.

Figure m
4.32  Equipotential curves
on the surface of the soil
near the structure n° 40.
Values in kV.

10

0 10 m

In a point outside the covered area, the limits are:

0,116
Vmax_ step _ adm = (1000 + 6 x500 ) = 464 V
l
0,116
Vmax_ touch _ adm = (1000 + 1, 5 x500 ) = 203 V
l
Figure 4.32 shows an equipotential map near tower 40.
The curve in red indicates the points where the generated touch voltage is equal
to the allowable limit for this voltage. Points inside this curve have generated touch
voltage less than the limit.
Figure 4.33 shows graphs of the calculated step voltages.
Table 4.15 lists the maximum values of the generated step and touch voltages
with the respective limit. As it can be seen, the step and touch voltages are
controlled.

4.10.2 Temporary (Sustained) Overvoltage

They are of sinusoidal type and defined by its magnitude and duration. They affect
the insulation withstand of the clearances (gaps) and other insulation and are defined
by test with an amplitude with duration of one minute. They are also important for
examining the surge arrester behavior and its energy absorption. The surge arrester
rating is chosen not to conduct significant current during these overvoltages.
The origins of temporary overvoltages are: earth faults; load rejection; line/
equipment switching; resonances (IEC 71-2).
4  Electrical Design 115

800,0
750,0
700,0
650,0
600,0
550,0
500,0
450,0
400,0
v

350,0
300,0
250,0
200,0
150,0
100,0
50,0
0,0
1 2 3 4 5 6 7 8 9 10 11 12 13 14
m

Figure 4.33  Generated step voltage at 45 degree direction.

Table 4.15  Comparison between limits and the generated step and touch voltages (for tower
n°40)
Soil with thin layer of granite stones Natural soil
Limits (V) Generated Step voltage limit (V) Step voltage
voltages (V) generated (V)
Vtouch Vstep Vtouch Vstep
742 2621 513 748 464 428

4.10.2.1  Earth Faults

These overvoltages are related to phase-to-ground faults location and the system
neutral earthing. For ungrounded earthing system the phase-to-ground overvoltage
may reach values close to the phase-to-phase voltage. For grounded neutral impe-
dance or solidly grounded system these overvoltages are much smaller.

4.10.2.2  Load Rejection

For long line systems, after load rejection overvoltages appear due to Ferranti effect,
they are bigger in the line opened end.
Shunt reactors connected to the lines reduce these overvoltages.
The condition may become worse if load rejection is combined with pre existing,
or post occurring phase-to-ground faults.

4.10.2.3  Line/Equipment Switching

Energization/reclosing of lines lead to temporary overvoltages due to Ferranti effect.
Shunt reactors reduce the overvoltage.
Capacitor switching in is also a cause of temporary overvoltages.
116 J.F. Nolasco et al.

4.10.2.4  Resonances
Temporary overvoltages may be originated by resonances, and may be mitigated by
detuning the system circuit.
Transformer energization and ferro-resonance should be of concern.
An example of sustained transient overvoltage due to load rejection is shown in
Figure 4.34.

4.10.3 Slow-Front Overvoltages (Switching Surges)

Slow front overvoltages are of oscillatory nature fast damped. They are represented
in laboratory test by a wave with time-to-peak of 250 μs and time to half-value in
the tail 2500 μs (Figure 4.35).
The switching surge overvoltages arise from:

• line energization
• line reclosing (re-energization)

100 ms

100 ms

Actual overvoltage Insulation test waveform

Figure 4.34  Sustained overvoltage – load rejection.

200 μs

10 ms

Figure 4.35  Slow front overvoltage (actual on left and laboratory test on right).
4  Electrical Design 117

• fault inception
• fault clearing
• load rejection
• capacitive switching in
• inductive load switching out.

4.10.3.1  Line Energization

During line energization, a slow-front overvoltage occurs superimposed to the power
frequency overvoltage. When the breaker closes, a travelling wave move along the
line, originating a slow-front overvoltage (after some reflections/refractions).
The peak value of the overvoltage depends on the point-on-wave switching
instant, and the Ferranti effect influenced by the presence of shunt reactors.
The overvoltage may be mitigated by synchronized switching or by the use of
pre-insertion resistor.
When pre-insertion resistor is present, the transient phenomenon has two compo-
nents: one when the resistor is inserted; and another when it is bypassed.
The resistor insertion time average value is specified (about 10 ms) but there is a
random variation of few milliseconds (2-4 ms). Synchronized switching has also a
random behavior as related to closing instant.
These energization overvoltages are determined by simulation with electromag-
netic transient model software running a set of 100-200 cases (shots) characterized
by the switching closing instant in the three phases.
It is assumed that the resistor insertion instant follows a Gaussian distribution
defined by a mean value and standard deviation. As result, the maximum value of the
overvoltage is determined (and the corresponding switching closing instants) and a set
of voltages values in the sending, receiving and some intermediate distance of the line.
The values are used to define a statistical distribution of overvoltage (Gaussian
or Weibull) through a mean, a standard deviation, and a truncated maximum value.
There are two ways for establishing the distribution of overvoltage: called “phase-­
peak” and “case-peak” methods. In the former for each shot, for one location, the peak
value of the three phases are included in the distribution; in the latter, only the highest
of the three phase peaks only is included in the distribution. Therefore they should be
considered in different ways when designing the insulation.
Energization over an existing phase-to-ground fault may lead to higher overvol-
tage; however they are not used for line insulation design but only to checking surge
arrester performance.
It should be noted that the phase-to-ground and phase-to-phase overvoltage dis-
tribution shall be obtained for insulation design.
A graph with the mean plus three standard deviation values of phase-to-ground
overvoltage along a line is depicted in Figure 4.36.

4.10.3.2  Reclosing
After line opening, one or more tentative of reenergization may occur automatically.
When the line is disconnected a trap charge is kept in the line (in the line capa-
citance) so the reclosing is an energization over the residual voltage of the line; this
should lead to higher show-front overvoltage than for energization.
118 J.F. Nolasco et al.

Energization overvoltage: 500 kV, 320 km, no shunt reactor

2,5

2,0

1,5
mean plus 3 standard deviations
pu

1,0

0,5

0,0
0 25 50 75 100
% of the line

Figure 4.36  Expected maximum switching surge overvoltage during line energization.

For lines without shunt-connected reactor the trap charge is a DC voltage with
certain damping (due to line conductance).
For line with shunt-reactor the line voltage is of oscillatory nature (with two
frequencies superimposed, a combination due to the natural line frequency and the
operating voltage frequency). In the studies, the worst instant of breaker contact
closing shall be searched. After that, a statistical calculation around this worst posi-
tion is done (random contact closing instant).
The reclosing overvoltages are mitigated by using pre-insertion resistor in the
breakers or synchronized closing system. The trap charge can be controlled through:
open resistor in the breaker; shunt reactor; inductive potential transformer, and by
closing/opening a line to ground fast switch.
The phase-to-ground and phase-to-phase over-voltage distributions are searched to be
used in the insulation coordination, in a similar way as for the energization overvoltage.
Figure 4.37 depicts a trap charge in a reactor shunt compensated line.

4.10.3.3  Load Rejection

Apart from the sustained overvoltage in the initial cycles there may occurs low-front
overvoltages, in general lower than those for energization/reclosing.
Load rejection with phase-to-ground fault (before or after breaker operation)
may be critical event for surge arrester performance.

4.10.3.4  Fault Application

When a fault occurs a travelling wave goes in the line and may cause high overvol-
tage in points of discontinuities (different surge impedances) or when summing up
waves from different pass.
In general this overvoltage has short shape and is discharged by surge arrester
without any high energy content.
Sometimes they are treated as fast-front surge.
4  Electrical Design 119

800

[kV]

460

120

-220

-560

-900
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 [s] 0.8
time
Phase A red; Phase B green; Phase C blue

Figure 4.37  Trapped charge (500 kV system) from 0.12 to 0.65 s; unsuccessful reclosing (fault
at phase A).

4.10.3.5  Fault Clearing

They are in general lower than energization/reclosing overvoltages and they depend
on the type and distance of the fault, breaker sequence of opening, and prior net-
work condition. Opening resistor may be used to mitigate them.

4.10.3.6  Inductive and Capacitive Load Switching

Capacitive load switching off does not lead to overvoltage; overcurrent during swit-
ching in is therefore of concern.
Inductive load switching off may cause local overvoltage when the breaker
forces the current to zero before natural zero crossing.
Transformer energization may cause high inrush current the could lead to reso-
nance in points of the system.
This type of overvoltage, in general, does not influence line design but substation
design. Mitigation is obtained with closing/opening resistor or synchronized switching.

4.11 Insulation Coordination

4.11.1 General

When a low voltage stress is applied in insulation there is no flow of current. When
this stress is increased to a sufficiently level, the resistivity along the pass through
the insulation changes to a low value, conducting current (breakdown).
A number of factors influence the dielectric strength of the insulation (IEC 71-2):
120 J.F. Nolasco et al.

• The magnitude, shape, duration, polarity of the voltage applied

• The electric field distribution in the insulation
• The type of insulation: air, liquid, solid, gas
• The physical state of the insulation (including ambient conditions).

Breakdown in air insulation is strongly dependent on gap configuration and polarity

and on the wave shape of the voltage stress.
This withstand capability of insulation is determined through standard test:

• Sustained overvoltage sinusoidal wave

• Fast-front 1 min 1.2/50 μs waveform
• Slow-front 250/2500 μs waveform

Withstand capability is different depending on the wave polarity.

The insulation withstand depends on the ambient conditions, and it is referred to
“standard atmospheric conditions”.

• Temperature 20 ° C
• Pressure 101.3 kPa (1013 mbar)
• Absolute humidity 11 g/m3.

4.11.2 Statistical Behavior of the Insulation

First of all it should be noted that some insulations are non regenerative (oil, paper
in a transformer for instance) and others are auto-regenerative like the air. In the
latter case the statistical behavior is discussed here-in-after.
When a certain number of shots, with the same wave, are applied in an insulation
the breakdown may occur by some of them only.
Due to this, the insulation withstand is defined by a probability function (Gaussian
or Weibull) (Figure 4.38).

Gaussian (Normal) Distribution

 −1 2
1 y
P (U ) = ∫e2
dy (4.123)
2 −∞

Where

X = (U − U 50 ) / Z

U50 being the 50% discharge voltage (P(U50) = 0,5), and Z being the conventional
deviation.
Table 4.16 shows some values.
P(y) = probability of not being exceeded
[1-P(y)] = probability of being exceeded
4  Electrical Design 121

1.0

0.0
+∞

frequency f(x) probability P (U ) = ∫ f ( x) dx

−∞

Figure 4.38  Gaussian distribution (frequency and probability).

Table 4.16  Frequency and P(y) Rounded

probability y values 1 − P(y)
−3 0.001 0.999
−2 0 02 0.98
−1.34 0.10 0.9
−1 0.16 0.84
0 0.50 0.5
1 0.84 0.16
1.34 0.90 0.1
2 0.98 0.02
3 0.999 10−3
4 0.999968 0.3 10−4
5 0.9999997 0.3 10−6

Weibull Distribution
The equations are:

 U −
− 
P (U ) = 1 − e   
(4.124)

Where δ is the truncation value, β is the scale parameter and γ is the shape
parameter.

δ = U 50 − NZ (4.125)

1
β = NZ ( ln 2 ) 2
−
(4.126)

This leads to the modified Weibull
γ
 U −U 50 
 1+ 
P (U ) = 1 − 0, 5
 ZN 
(4.127)

122 J.F. Nolasco et al.

N = number of conventional deviations

The exponents determined by

( P (U 50 − Z ) = 0,16 ) (4.128)

 ln (1  0,16 ) 
ln  
ln 0, 5 
=  (4.129)
ln (1  (1 / N ) )

With truncation at (U0 = U50 − 4Z), N = 4, results γ ≈ 5.0 and finally

( x = (U − U ) / Z )
50 (4.130)
5
 x
 1+ 
P (U ) = 1 − 0, 5  4
(4.131)

Characterization of the Insulation Withstand

The statistical behavior of the insulation (as a Gaussian distribution) is defined pro-
vided two values are known for instance, the mean U50, and the standard deviation
Z = U50 − U16. Sometimes the value U50 is substituted by U10 or U2.
When the Weibull distribution is used, the truncation value is also defined in
terms of N conventional deviations (ex: N = 4).
The conventional deviation of the insulation can be assumed as:

• For fast-front (lightning) Z = 0.03 U50

• For slow-front (switching surge) Z = 0.06 U50

IEC 71-2 considers the value U10 = (U50 − 1.3 Z), to define the withstand capabi-
lity of equipment insulation.

4.11.3 Insulation Coordination Procedure

4.11.3.1  C  ontinuous (Power Frequency) Voltage and Temporary

Overvoltage
The coordination is set based in the maximum voltage peak value phase-to-ground
that is the phase-to-phase voltage divided by 3 .
Insulation withstand of the insulator string varies depending on the pollution
level.
Table 4.17 contains the specific creepage (mm/kV), to set the recommended dis-
tance depending on the pollution level. The distance referred is the contour of the
insulator (creepage distance).
4  Electrical Design 123

Table 4.17  Recommended creepage distance (IEC 71-2)

Minimum nominal
Pollution specific creepage
level Examples of typical enviroments distance mm/kV1
I Light - Areas without industries and with low density of 16.0
houses equipped with heating plants
- Areas with low density of industries or houses but
subjected to trequent winds and/or rainfall
- Agricultural areas2
- Mountainous areas
- All these areas shall be situated at least 10 km to
20 km from the sea and shall not be exposed to winds
directly from the sea3
II Medium - Areas with industries not producing particularly 20.0
polluting smoke and/or with average density of houses
equipped with heating plants
- Areas with high density of houses and/or industries
but subjected to frequent winds and/or rainfall
- Areas exposed to wind from the sea but not too close
coasts (at least several kilometres distant)3
III Heavy - Areas with high density of industries and suburbs of 25.0
large cities with high density of heating plants
producing pollution
- Areas close to the sea or in any cases exposed to
relatively strong winds from the sea3
IV Very - Areas generally of moderate extent, subjected to 31.0
Heavy conductive duate and to industrial smoke producing
particularly thick conductive deposits
- Areas generally of moderate extent, very close to the
coast and exposed to sea-spray or to very strong and
polluting winds from the sea
- Desert areas, characterized by no rain for long
periods, exposed to strong winds carrying sand and
salt, and subjected to regular condensation
NOTE - This table should be applied only to glass or porcelain insulation and does not cover some
enviromental situations such as snow and ice in heavy pollution, heavy rain, arid areas, etc.
1
According to IEC 815, minimum creepage distance of insulatiors between phase and earth related
to the highest system voltage (phase-to-phase)
2
Use of fertilizers by spraying, or the burning of crop residues can lead to a higher pollution level
due to dispersal by wind
3
Distances from sea coast depend on the topography of the coastal area and on the extreme wind
conditions

4.11.3.2  Slow-Front (Switching Surge)

There are two methods: deterministic; and statistical approaches.
In the deterministic approach a statistical value of the overvoltage is set equal to
a statistical value of the withstand (both with certain probability)

U S 50 + N S Z S = UW 50 − NW ZW (4.132)

US50, UW50 are the means of the overvoltage and withstand capability
124 J.F. Nolasco et al.

ZS, ZW are the standard deviations (overvoltage-withstand)

NS, NW = number corresponding to a desired probability (form instance NS = 3;
NW = 4), or the truncation points.
In the statistical approach the risk of failure is evaluated. The following assump-
tions are established:

• Peaks other than the highest are disregarded

• Shape is taken as identical to the standard waveform
• All overvoltage of the same polarity (the worst).

The risk is than calculated as

U2

R= ∫ f ( u ) P ( u ) du
U1
(4.133)

where:
f(u) = probability density of the overvoltage
P(u) = discharge probability of the insulation
U1 = truncation point of the discharge probability
U2 = truncation point of the overvoltage
Figure 4.39 shows the procedure.
A simplified approach consists in the assumption that the overvoltage (US50, ZS)
and discharge voltage (UW50, ZW) are Gaussian curves.

f(u) = probability density of the overvoltage occurrence described

by a truncated Gaussian or a Weibull function
P(u) = discharge probability of the insulation described by a
modif ied Weibull function
U1 = truncation value of the overvoltage probability distribution
U50 -4Z = truncation value of the discharge probability distribution

Figure 4.39  Evaluation of the risk of failure.

4  Electrical Design 125

Failure occurs when overvoltage is greater than withstand. The combination is

also a Gaussian distribution in which the mean (R50) and the standard deviation (ZR)
are:

R50 = U S 50 − UW 50 (4.134)

Z R = Z S2 + ZW2 (4.135)

1
0
− ( )
1 2
x
Risk = 1 − ∫e
2 π −∞
2

(4.136)

x − R50
X = (4.137)
ZR

Example: Calculate the risk of failure for:

=U S 50 820
= kV Z S 82 kV or 10% (4.138)

UW 50 = 1125 kV ZW = 45 kV or 4% (4.139)

R50 = 820 − 1125 = −305 (4.140)

Z R = 82 2 + 452 = 93.5 kV (4.141)

0 − ( −305 )
X = = 3.2 (4.142)
93.5

From Gaussian table values X = 3.2:

R = (1 − 0.99931) = 0.0007.
When there are n equal insulations stressed by the same overvoltage, the risk of
failure R, of at least one insulation breakdown is:

R = 1 − (1 − R1 )
n
(4.143)

Where R1 is the individual risk
Therefore for calculating the risk in the case of:

• energization overvoltage, once known distribution at sending, middle and recei-

ving ends, calculation is made as “quasi-peak” method.

The following steps shall be followed:

• set one insulation defined by UW50, ZW

• calculate the risks Rs, Rm, Rr (at sending, middle and receiving end points)
• assume that Ns insulation are stressed by the sending overvoltage, Nm and Nr by
middle and receiving overvoltage.
126 J.F. Nolasco et al.

The total risk of a failure will be:

R = 1 − (1 − Rs ) s (1 − Rm ) m (1 − Rr ) r
N N N
(4.144)

Note: If the distributions were determined as “phase–peak”, than, the three phases
risk has to be considered as for instance.

Rr = 1 − (1 − Rr1 ph )
3
(4.145)

Rri1ph is the risk in one phase

4.11.3.3  Fast-Front (Lightning Surge)

The same concepts applied above for slow-front are valid for fast-front
overvoltages.

4.11.3.4  Influence of Atmospheric Conditions

The air pressure, temperature, and humidity affect the withstand capability of an air
gap or insulator.
Assuming that the effect of temperature and humidity cancel it other [1], than
only the effect of air pressure (altitude) is present and the correction factor Ka is
applied to the withstand capability of the insulation.
 H 
m 
Ka = e  8150 
(4.146)

Where
H is the altitude above sea level (in meters) and the value of m is as follows:
m = 1.0 for co-ordination lightning impulse withstand voltages;
m according to Figure 4.40 for co-ordination switching impulse withstand
voltages;
m = 1.0 for short-duration power-frequency withstand voltages of air-clearances
and clean insulators.
Note: The exponent m depends on various parameters including minimum
discharge path which is generally unknown at the specification stage. However, for
the insulation co-ordination purposes, the conservative estimates of m shown in
(Figure 4.146) may be used for the correction of the co-ordination switching impulse
withstand voltages. The determination of the exponent m is based on IEC 60-1 in
which the given relations are obtained from measurements at altitudes up to 2000 m.
In addition, for all types of insulation response, conservative gap factor values have
been used.
For polluted insulators, the value of the exponent m is tentative. For the purposes
of the long-duration test and, if required, the short-duration power-frequency with-
stand voltage of polluted insulators, m may be as low as 0.5 for normal insulators
and as high as 0.8 for anti-fog design.
The values of m are shown in Figure 4.40.
In Figure 4.41 the distribution of withstand value of an example of insulation
affected by atmospheric conditions is shown.
4  Electrical Design 127

Figure 4.40 Dependence 1,0

of m on the switching
surge withstand voltage. c
b

m a

0,5

0,0
1000 kV 2000 kV
Ucw
a) phase-to-earth insulation
b) longitudinal insulation
c) phase-to-phase insulation
d) rod-plane gap (reference gap)
For voltages consisting of two components, the voltage value is the
sum of the components.

40

h
30
i

20 Standard
%

condition
g

j
10 f
e
d
c
a b k
0
1020 1040 1060 1080 1100 1120 1140 1160 1180 1200 1220 1240 1280
kV

Figure 4.41  Distribution of 50% withstand value of an insulation affected by atmospheric

conditions.

The effect of atmospheric condition is then considered in the risk of one insula-
tion by applying the overvoltage distribution in the withstand capability affected, in
each interval, and then calculating the weight average of the values of risk.

4.11.4 Withstand Capability of Self Restoring Insulation

The air gaps, filled or not with insulators, are of the self-restoring type. The geome-
trical configuration of the gap influences its withstand capability.
128 J.F. Nolasco et al.

The critical flashover value (U50), in kV, for “standard atmospheric condition”,
can be estimated as function of the gap distance (d) in m by:
For slow-front

U 50 = k 500 d 0.6 2 < d < 5 (4.147)

3400
or U 50 = k 5 < d < 15 (4.148)
8
1+
d

k being the gap factor as shown in Figure 4.42.

Phase-to-phase insulation (see Figure 4.43) is also influenced by the factor α,
defined as the ratio of the negative peak and the sum of the positive and negative
peaks.
Table 4.18 shows the gap factors to be considered for α equal to 0.5 and 0.33.

• for fast-front overvoltages

U 50 = k + 500 d (4.149)

k + = 0, 74 + 0, 26 k (4.150)

K is the gap factor for slow-front overvoltages

For positive polarity and/or insulator strings in order to evaluate effect of light-
ning impinging the substation:

U 50 = 700 d

• for sustained overvoltages

The withstand characteristic is shown Figure 4.44.

Finally in a tower there are many gaps subjected to the same overvoltage:
conductor-­tower (arm); conductor-guy wire; conductor-tower (lateral); conductor-­
to-­ground; insulator string. The risk of failure in one tower Rt shall be estimated by:

Rt = 1 − (1 − Rg 1 ) (1 − Rg 2 )……. (1 − Rgk ) (4.151)

Rgk is the risk of insulation gap k.

4.12 Electric and Magnetic Fields, Corona Effect

Ground level electric and magnetic field effects of overhead power lines have
become of increasing concern as transmission voltages are increased. The electric
fields are especially important because their effects on human beings and animals
4  Electrical Design 129

Gap type Insulator Factor k

Without With
rod-plane 1.0 1.0

rod-structure (below) 1.05

conductor-plane 1.15

conductor-window 1.20 1.15

conductor-structure (below) 1.30

rod-rod (3m below) 1.30

conductor-structure 1.35 1.30

rod-rod (6m) 1.40 1.30

conductor-guy wire 1.40

conductor-tower arm 1.55 1.50

conductor rod (3 m) 1.65

conductor rod (6 m) 1.90

conductor-rod (above) 1.90 1.75

Figure 4.42  Gap factor for slow-front overvoltages with and without insulator string.
130 J.F. Nolasco et al.

Figure 4.43 Phase 1020 kV

–to–phase overvoltage. 710 kV VAB

VA

200 kV
VB
1 ms

650 kV

Table 4.18  Gap factor Configuration α = 0.5 α = 0.33

phase-to-phase insulation
Ring-ring or large smooth 1.80 1.70
electrodes
Crossed conductors 1.65 1.53
Rod-rod or conductor- 1.62 1.52
conductor (along the span)
Supported busbars (fittings) 1.50 1.40
Asymmetrical geometries 1.45 1.36

kV peak
2400 rod-rod

conductor-structure (lateral)
2000

1600 rod-plane

1200

800

400

0 d (m)
0 1 2 3 4 5 6 7 8 9 10

Figure 4.44  Withstand characteristic of gap (power frequency).

have been a concern in the last decades. Serious views still exist that prolonged
exposure to electric and magnetic fields could be associated with adverse health
effects or with increased risks. However, it is not appropriate to consider unlikely
conditions when setting and applying electric field safety criteria because of pos-
sible consequences; thus statistical considerations are necessary.
4  Electrical Design 131

The resultant electric and magnetic fields in proximity to a transmission line are
the superposition of the fields due to the three-phase conductors. Usually some
limitations, originated from the practice or researches are imposed to the maximum
electric field at the edge of the right-of-way.
The evaluation of the electric and magnetic fields across the right-of-way of over-
head transmission line can nowadays be made with high accuracy so that the possible
health effects of such fields over humans, animals and plants can be evaluated.
Although there is no evidence of harmful effects of the magnetic fields over
humans or animals, there are certain limitations imposed by the practice and by the
good sense. International organizations like Cigré and ICNIRP have undertaken
extensive investigations on such issue.
The range of maximum values expected and accepted as the usual field intensi-
ties of electric and magnetic fields are shown in Tables 4.19 and 4.20 below:
Although medical examinations in linesmen, performed in various countries,
have so far failed to scientifically prove health problems directly attributed to elec-
tric and magnetic fields produced by overhead lines, some conventional limit values
have been established for exposures from which the numbers given in Table 4.21
below gives an indication.
In general, limitations are according to Table 4.21, by ICNIRP, according to
them maximum values for general public are set for the right-of-way border, while
other values are established for occupational people.

Table 4.19  Range of maximum allowable electric and magnetic fields below overhead lines of
any voltage (example)
Exposure Type Electric Field Limit(kV/m) Magnetic Field Limit (μT)
Difficult Terrain 20 125
Non-populated Areas 15-20 100-125
Road Crossings 10-12 50-100
Frequent Pedestrian. 5 50
Circulation

Table 4.20  Range of maximum expected electric and magnetic fields below overhead lines as a
function of line voltage
TL voltage (kV) Electric field at ground level (kV/m) Magnetic field at ground level (μT)
765 8-13 5
500 5-9 3
345 4-6 3
230 2-3.5 2
161 2-3 2
138 2-3 2
115 1-2 1.5
69 1-1.5 1
132 J.F. Nolasco et al.

Table 4.21 Maximum General

electric and magnetic field Limit values public Occupational
values set by ICNIRP
Electric field 250/f 500/f
(ICNIRP 1997)
kV/m
Magnetic field uT 5000/f 25000/f

According to ICNIRP new limits have been introduced as Guidelines for limiting
Exposure to time-varying electric, magnetic, and electromagnetic fields (up to
300 GHz); the limit values for “general public” and “occupational workers” are
established as below:
where f it is the frequency in Hz.
Regarding the maximum acceptable limits for the magnetic fields, there are no
universally definitive numbers as some controversy is still worldwide existent espe-
cially about their real effects on the health of human beings and animals. While in
some countries the regulations are more permissible, in others severe rules have
been established.
Other two types of unwanted disturbances caused by overhead transmission lines
on the environment are also of importance, namely:
Radio Noise or Radio Interference (RI) that is a disturbance within the radio
frequency band, such as undesired electric waves in any transmission channel or
device. The generality of the term becomes even more evident in the frequency band
of 500 kHz to 1500 kHz (AM band). The frequency of 1000 kHz (1 MHz) is usually
taken as reference for RI calculation.
Audible Noise (AN) produced by Corona of transmission line conductors has
emerged as a matter of concern of late. In dry conditions the conductors usually
operate below the Corona-inception level and very few Corona sources are pre-
sent. Audible noise from AC transmission lines occurs primarily in foul weather.
However, in general, it can be said that transmission systems contribute very
little as compared with the audible noises produced by other sources. In the case
of rural lines, the importance of the Audible Noise (AN) as well as of the Radio-
Interference (RI) may be still lower, as the population density beside the line is
generally too small.

4.12.1 Corona Effects

When a set of voltages are applied on the conductors of a transmission line an elec-
tric field or voltage gradient appears on its surface (conductor surface gradient). If
this surface gradient is above a certain limit (Peek gradient or critical Corona or set
gradient) the Corona discharges initiate.
Electric discharge phenomena produce various effects (power loss, high fre-
quency electromagnetic fields, acoustic and luminous emission, ions and ozone
generation).
High frequency electromagnetic fields interfere with radio or TV signals in the
proximity of the lines.
4  Electrical Design 133

A person standing near an overhead line whose conductors and/or assemblies are
under Corona can sometimes hear a special noise: frying, crackling and hissing
sounds and low frequency hum.
These effects are influenced by the transmission line characteristics (conductor
size, bundle configuration, phase/pole spacing, conductor height to ground), con-
ductor surface gradient and atmospheric condition (temperature, pressure, rain,
snow, etc) the last ones a statistical behavior to the phenomenon is assigned.
The phenomenon presents different aspects if AC or DC line is under consideration.
Corona considerations in the design of transmission lines have been discussed in
the Cigré TB 61 (1996) and Cigré TB 20 (1974). This publication includes discus-
sion of Corona losses (CL), radio interference (RI) and audible noise (AN).
Factors influencing the choice of conductor bundles are discussed below. This
section provides basis for selection of the conductor bundle.

4.12.1.1  Conductor Surface Gradient

Consider the case of two conductors above soil.
For AC or DC lines, the relationship between charge (Q) and voltage (V) on the
conductors or shield wires is given by:

[V ] = [ H ][Q ] (4.152)
Hij are the Maxwell potential coefficients (refer to Figure 4.45):

1 2 hi
H ii = ln (4.153)
2 0 reqi

1 Dij
H
= ij H=
ji ln (4.154)
2 0 Dij

where:
ri = single conductor (sub index c) or shield wire (sub index sw) radius;

Figure 4.45  Symbols for l rj

j
λij calculation.
ri Dij
l
i

hj
hi D'ij

i' dij
l

j'
l
134 J.F. Nolasco et al.

nri
req = R n =equivalent radius of the bundle;
R
n = number of sub-conductors in the bundle;
R = radius of the bundle;

a
R=
2 sin ( π / n )

for a regular bundle with a distance between adjacent subconductors;
1
hij = hmin + s = average height of the conductor or shield wire;
3
hmin = minimum distance of the conductor or shield wire to ground;
s = sag of the conductor or shield wire;
Equation above can be divided into:

 Vc   H c − c H c − sw   Qc 
 =   (4.155)
Vsw   H sw − c H sw − sw   Qsw 
If the shield wires are grounded, then Vsw = 0 and the above equation system is
reduced to:

[Vc ] = ([ H c −c ] − [ H c − sw ][ H sw− sw ] [ H sw−c ]) [Qc ]

−1
(4.156)

If the shield wires are isolated from ground then the corresponding equations can be
deleted. In both cases the set of equations reduces to:

[Vc ] =  H c′ −c  [Qc ] (4.157)
The number of lines and rows of the matrix [λ′c–c] is equal to the number of phases
of an AC line or poles of a DC line. The charges are then calculated by:
−1

[Qc ] =  H c′ −c  [Vc ] (4.158)
Since it is assumed that the total charge of the bundle is equally distributed on the n
subconductors, the mean gradient of a conductor in a bundle is given by:

1 Q
ga = (4.159)
n 2 0 r

The average maximum gradient of the subconductors is defined by:

 ( n − 1) r 
g = g a 1 +  (4.160)
 R 

The critical Corona onset gradient is given by:

 k 
g c = g o δ m 1 +  (4.161)
 δ r 

4  Electrical Design 135

where:
gc = critical Corona onset gradient (kV/cm);
go = Corona onset gradient (normal ambient conditions: 25 °C, 76 cm Hg) (kV/
cm);
r = radius of the conductor (cm);
k = 0.308 for AC or DC (both polarities);
m = surface factor;
m = 1 smooth and polished surface
m = 0.6 to 0.8 actual dry weather service conductor
m = 0.3 to 0.6 raindrops, snowflakes, extreme pollution
m = 0.25 heavy rain
δ = relative air density (RAD);

ρ
δ = Kd
273 + t
P = pressure of the ambient air (cm Hg or Pa);
t = temperature of the ambient air (°C);
Kd as in Table 4.22.
In the line design the conductor surface gradient should be smaller than the Peek
gradient and including a safety factor (ex: g < 0.95 gc. This gradient is also the key
factor in the interferences that the line may cause (radio, audible).

4.12.1.2  Corona Loss

AC transmission Corona losses (in dB) can be calculated based on equations pre-
sented in the references (Chartier 1983; Maruvada 2000)

E d
P ( dB ) = 14.2 + 65 log + 40 log
18.8 3.51
(4.162)
n A
+ K1 log + K 2 +
4 300

Where:
n = number of subconductors
d = diameter of subconductors, cm
K1 = 13 for n ≤ 4 and 19 for n > 4

Table 4.22  Values of Kd Kd

Normal conditions (25 °C, 76 cm
Hg)
°C and cm Hg 3.921
°C and Pa 0.00294
IEC normal conditions (20 °C, 76 cm
Hg)
°C and cm Hg 3.855
°C and Pa 0.00289
136 J.F. Nolasco et al.

E = maximum conductor surface gradient, kV/cm

A = altitude, m

I
K 2 = 10 log for I ≤ 3.6 mm / h
1.676
I
K 2 = 3.3 + 3.5 log for I > 3.6 mm / h
3.6

I = rain rate, in mm/h
A distribution of fair and rainy weather has to be established.

4.12.1.3  Radio Interference

Radio interference (RI) is any effect on the reception of wanted radio signals due to
any unwanted disturbance within the radio frequency spectrum. Radio interference
is a concern only with amplitude modulation (AM) radio reception because
frequency-­modulated (FM) radio is inherently less sensitive to disturbances. Radio
interference is evaluated by comparing the noise level with the radio signal, i.e. the
signal-to-noise ratio (SNR), at the edge of the servitude or right-of-way, and for a
frequency of normally 0.5 or 1 MHz using a quasi-peak detector with a bandwidth
of 5 or 9 kHz according to CISPR or ANSI standards, respectively (1974; 1996).
The RI level is expressed in dB above 1 μV/m.
Since RI and AN are caused by the same phenomenon, i.e. streamer discharges
appearing on the positive conductor or during the positive half-cycle, the variation
of RI with the weather conditions is essentially the same as for AN. Thus, the RI
level is the highest in rain for AC lines, but lower in rain for DC lines
As RI is dependent on the weather conditions, it is appropriate to represent the
RI level in statistical terms for each weather condition, such as the L5 and L50 levels
in fair weather or rain.
Note: Lx is termed the exceedence level; this is defined as the level that is excee-
ded x% of the time. For example L50 is the level that is exceeded 50% of the time
and L90 is exceeded 90% of the time. The value x and the cumulative frequency are
complementary to one another, i.e. x% = 100% - cumulative frequency (%).
Alternatively, the results may be presented as an “all weather” curve considering
average climate conditions.
Cigré TB 61 (1996) reports “empirical” formulae for the radio interference level
at the reference frequency of 0.5 MHz or 1 MHz, at a given distance from a three-­
phase line, for three basic weather categories (mean fair weather, mean foul weather
and heavy rain), as a function of the main influencing parameters (conductor surface
gradient, diameter of the subconductors, number of subconductors in the bundle,
frequency, etc.). These formulae were based on the results of direct measurements
of radio interference levels performed on operating and experimental lines with
system voltages of up to 800 kV and bundles of up to four subconductors.
Sometimes “empirical” equations are given considering two approaches: BPA
and Cigré TB 61. To compare the results of the two methods, the following charac-
teristics of the two measuring standards have to be considered (Table 4.23).
For comparison of measurements performed using different receivers:
Value (CISPR) = Value (ANSI) – 2 dB
4  Electrical Design 137

Table 4.23  Basic characteristics of CISPR and ANSI radio interference measurement standards.
Receiver
Pass band Charge/discharge constant Measuring Frequency
CISPR 9 kHz 1 ms/160 ms 0.5 MHz
ANSI 5 kHz 1 ms/600 ms 1.0 MHz

For correction of measuring frequency:

Value (CISPR) = Value (ANSI) + 5.1  dB
Cigré approach consider (CISPR specification: QP: 9 kHz – 0.5 MHz) and the
“heavy rain” RI in dB is:

 D
RI hr = −10 + 3.5 g + 6 d − 33 log  
 20 

where
g = maximum surface gradient (function of the mean height)
D = radial distance from the phase to the point
D = subconductor diameter
With: 10 m < D < 60 m, and the term 6d is valid for 1 cm < d < 2.5  cm only.
It should be noted that (see Figure 4.46):

• from the heavy rain value subtract 24 dB to get mean fair weather value
• from the heavy rain value subtract 7 dB to get “mean foul weather”
• from the heavy rain value subtract 3.5 dB to get “mean stable rain”.

The equation applies for all phases, however the total RI can be calculated by:
3
Ei = 20 log ∑E
k =1
2
ki

(4.163)

RI kj

Ekj = 10 20 (4.164)

Ranking of the phase-fields:

RI a ≥ RI b ≥ RI c (4.165)

(dB (1 μV/m))

If RI a − RI b ≥ 3 dB : RI = RI a (4.166)

RI a + RI b
If RI a − RI b < 3 dB : RI = + 1.5 (4.167)
2

Design criteria for RI from transmission lines are generally based on signal to noise
ratios (SNR) for acceptable AM radio reception. Studies carried out on Corona-­
generated RI from AC and DC transmission lines indicate that the SNRs for accep-
table radio reception are in the Table 4.24.
138 J.F. Nolasco et al.

Fair weather dry conductors

σ=6 dB
99.5
99
98
97
96
95

90
CUMULATIVE RELATIVE FREQUENCY %

80
70
60
50
40 20 dB
30
20 Heavy rain
int. ≥ 1 mm/h
10
σ=2.5 dB
5

2
1
0.5
-40 -30 -25 -20 -15 -10 -5 0 5
RELATIVE RADIO NOISE (dB)

Figure 4.46  RI and weather.

Table 4.24  Quality of radio reception

Signal noise ratic
Logarithmic (dB) Code Reception quality Subjective impression
30 5 Interference not audible
24 4 Interference just perceptible
18 3 Interference audible but speech perfectly received
12 2 Unacceptable for music music but speech intelligible
6 1 Speech understandable only with great concentration
0 0 Spoken word unintelligible: noise swamps speech totally

Minimum radio station signal requirement in many countries (ex: Brazil) is

66 dB for cities with population from 2500 to 10000 inhabitants. Similar condition
probably applies to other countries and is used here as part of the criteria.
At present, there are no established design criteria for RI from DC transmission
lines; so the tentative guidelines are for limiting the RI at the edge of the right of
way to (66-18) = 48 dB or to keep a reception code 3 at the reception. The SNR
above may be referred to average fair weather noise. For more stringent criteria, the
noise shall be below 48-6 = 42 dB for 85% probability of not being exceeded (in fair
weather), meaning that in 15% of the time the reception will be classified as bet-
ween the code 2 (in fair weather). The reference frequency is considered here as
0.5 MHz, if 1 MHz is considered the noise is 6 dB lower.
4  Electrical Design 139

4.12.1.4  Television Interference

Investigations regarding conductor Corona interference above 30 MHz are not as
extensive as for RI in the frequency range up to 30 MHz. In practice, interferences
to television reception (TVI) are more often caused by microgap discharges on
power line hardware or by polluted insulators than by conductor Corona. Because
of the higher frequency range, attenuation along a power line as well as that away
from the line is considerably larger. Consequently, effects of local noise sources are
much more pronounced for TVI. There are no indications that TVI should be of
special concern to AC or DC lines; therefore, if a line has an acceptable RI level,
then TVI need not be considered, and is not further discussed here (1974).

4.12.1.5  Audible Noise

The audible noise emanates from the air pressure variations that are caused by the
Corona discharges, more specifically the streamer discharges created under positive
DC voltage or during the positive half-cycle of the AC voltage. The audible noise is
the result of numerous uncorrelated Corona discharges, resulting in a broadband
noise spectrum covering the entire range of audible frequencies. AN from AC lines
also contain a hum component (100 or 120 Hz) caused by space charge movement
close to the conductor, correlated with the power frequency.
The human ear has a different response to each of these frequencies; therefore
weighting filter networks are used for measuring the human response. The most
common is the A-weighting network, in which case the audible noise level is stated
as dBA above 20 μPa.
For the AC conductors, (1996) presents empirical formulas for direct calculation of
AN from different sources. A Cigré formula for AN L5 levels, derived from the original
formulas, is also given. A formula developed by BPA formula is also presented and will
be used here.
For “wet conductor” the average excitation function is:
ΓA50 = K1 + 120 log g + K 2 log n +
q (4.168)
55 log d +
300

K1 K2
n < 3 −169.7 0
n ≥ 3 −182.7 26.4
For average fair-weather subtract 25 dBA
Range of validity: 230 – 1500 kV, n ≤ 16, 2 ≤ d ≤ 6.5  cm
The AN level in dBA is:
LA = ΓA + 54.3 − 11.4 logD (4.169)

ΓA5 = ΓA50 + 3.5 (4.170)

The sum of the individual pressure values from all phases results in the total sound
pressure at the point:
3 LAi

LATot = 10 log ∑10 10 (4.171)

i =1
140 J.F. Nolasco et al.

Note: The hum component of Corona is usually low, but might be taken into consi-
deration separately in special “foul weather” conditions with large Corona loss.
The AC lines design criteria are defined based on subjective evaluation obtained
from group of people.
Low complaints :< 52 dBA (equivalent to business office noise)
Moderate: (some) complaints = 52-58 dBA
Many complaints: >58 dBA
Moreover the weather condition has to be established and in some countries it is
defined as maximum noise not to be exceeded at average rain = 42 dBA (average).
This level corresponds to suburban living room noise (EPRI Transmission).

4.12.2 Fields

4.12.2.1  Electric Field

For the calculation of the electric fields close to the ground, first the V-Q-Maxwell
potential coefficient are used (EPRI Transmission).
The charges in all phases and eventually grounded shield wires (Q) are calcula-
ted by:

[Q ] = [ H ] [V ]
−1
(4.172)

It should be noted that V = (Vr + j Vi) is a complex number and so Q = (qr + j qi)
The electric field in a point N with coordinates (xN,yN) due to the charge qa and its
image qa is:
 ∼  ∼ 
Ea = E x , a u x + E y , a u y (4.173)

 
Where u x and u y are the unit vectors along the horizontal and vertical axes and Ẽx,a
and Ẽy,a are given by:

∼ ( qra + jqia ) ( xN − xa )
E x,a = −
2 π ( xa − xN ) + ( ya − y N ) 
2 2

 
(4.174)
( qra + jqia v ) ( xN − xa )
2 π ( xa − xN ) + ( ya + yN ) 
2 2

 
And

∼ ( qra + jqia ) ( yN − ya )
E y,a = −
2 π ( xa − xN ) + ( ya − y N ) 
2 2

 
(4.175)
( ra ia ) ( N a )
q + jq y + y
2 π ( xa − xN ) + ( ya + y N ) 
 2 2

 
4  Electrical Design 141

qa = qra + jqia

ya
Ẽ

xa N

xN

Image -qa

Figure 4.47  Calculation of the electric field at ground level.

The horizontal and vertical components, Ẽx and Ẽy of the electric field are calculated
by adding the contributions of all the conductors (a,b,…):
∼ ∼ ∼

E x = E x , a + E x ,b +… (4.176)
∼ ∼ ∼
E y = E y , a + E y ,b +… (4.177)

The results is an equation where there are the real and imaginary parts of the charge
and components in the x and y axis. To find the maximum values an specific calcu-
lation is needed (EPRI Transmission); one way is tabulating Ẽx, Ẽy and Et = Ex + E y
2 2

as function of the time and getting the maximum value of Et.

For the electric field at ground level the x axis components of a charge and its
image cancel themselves (Figure 4.47).
The equation is then:

qa ya
E y,a = (4.178)
π ε0 ( xa − xN )2 + ya2

Similar equations apply to other phases and the total field becomes:

Et = E y , a + E y ,b + …… (4.179)

4.12.2.2  Magnetic Field

The phenomena and calculation procedure are presented in (EPRI Transmission)
and is summarized here-in-after (Figure 4.48).
The variables to be considered are the flux density B and the magnetic field
H. They are related by the equation

B=µ H
(4.180)
µ = µ0 µr

142 J.F. Nolasco et al.

Conductor

r
h

θ Observer
L HX

HY
H

Figure 4.48  Magnetic flux at ground level.

The permeability of the free space is μ0 = 4π10−7 and μr the relative permeability is
unity for all except ferromagnetic materials.
Assuming a circle in a plane perpendicular to the conductor whose center is on
the conductor, the magnetic field along the circle is a constant given by:
i
H=
2 πr
(4.181)
µ i
B=
2 πr
Where i is the current and r the radius of the circle.
Simplified calculation of the magnetic field at ground level is given by:
i B
H= = (4.182)
2π h + L2 2 µ

h
H x = H sin ( θ ) = H
r

and
L
H y = H cos ( θ ) = H (4.183)
r
For multi-conductor the superposition applies in a similar way presented for electric
field. The maximum value has to be searched taking into consideration also that the
current has real and imaginary parcels.
Another consideration relates to the image of conductors, they have to be located
at great depth d that in a first approximation (being ρ the soil resistivity and f the
frequency) d is given by:

ρ
d ≅ 660 (4.184)
f

Accurate calculation may call for the use of Carson’s equations. However for many
purposes the earth current may be disregarded.
4  Electrical Design 143

4.13 Overvoltages and Insulation Coordination

First, the reader shall realize that the line constants consideration are included in the
section related to AC lines (Section 4.1.7) and is not repeated here reason why this
section starts with overvoltages.
The selection of the optimum transmission line (bipole) alternatives encompas-
ses the different components of the line, so that a global optimization can be achie-
ved. The optimum choice only has a real meaning when electrical, mechanical,
civil and environmental aspects are taken into account as a whole set, for which a
satisfactory performance and reasonable costs are simultaneously looked for (Cigré
TB 388 2009).
Regarding the transmission line itself, its design includes at first the electrical
requirements such as power transfer capability and voltage which are specified
from which the tower-top geometry, the electric field effects, the Corona effects,
the overvoltage and insulation coordination and the required right-of-way are esta-
blished. Then the mechanical design of the towers and foundations, the determina-
tion of conductors and shield wires stresses are carried out; finally the economics
including direct costs, cost of losses, operation and maintenance cost along line
life, are evaluated.
The design process is iterative as the electrical parameters can be met with a
variety of solutions. The optimum solution is derived from interaction with planners
and designers.
Note: this text is based on (Cigré TB 388 2009) that may be consulted if more
details are necessary. See also Chapter 7 and for electrical constants details
Section 4.1.7 (that includes AC and DC line constants).

4.13.1 Overvoltages

4.13.1.1  Types of Overvoltages

The definition of the insulation levels is dependent on different voltage stresses that
reach the air-gaps and are so chosen as to result in the best compromise between a
satisfactory electrical performance and reasonable costs.
To define the tower-top-geometry of the towers, in the case of a DC line, the
following voltage stresses are considered: sustained due to operating voltage, and
transient due to lightning and switching surge overvoltages. Therefore, the scope of
this clause is an evaluation of the overvoltages in the HVDC system aiming at the
DC line insulation design required.
The switching-surge overvoltages in a HVDC system occur in the DC as well as
in the AC part of the system.
In the latter one, overvoltages are the result of the following switching opera-
tions: line energization; line reclosing, load rejection, fault application, fault clea-
ring and reactive load switching, and all should be evaluated.
144 J.F. Nolasco et al.

As related to HVDC system, the above mentioned overvoltages are also conside-
red for the converter station insulation design; by the use of surge arresters, the
overvoltages are limited to values corresponding to the arrester Maximum Switching
and Lightning Surge Voltage Levels. The surge discharge capability of the arrester
needs to be verified as part of the overvoltage studies for equipment specification.
Regarding switching surges fault application is the only one type of overvoltage
to be considered because of the intrinsic process of the HVDC system. For line
energization and reclosing, the DC voltage is ramped up smoothly from zero, and in
the reclosing process the line de-energization process eliminates the trapped charge.
As for load rejection, it generally does not transfer overvoltages to the DC side.
DC filter switching does not cause overvoltages.
Lightning overvoltages may start a fault in the DC line, however its effect is
smaller as compared with AC system faults due to the fact that the fault current will
be limited by HVDC station controls, the line voltage is ramped down and after a
sufficient time for the trapped charge discharge, the voltage is ramped up to the
nominal value or to a reduced voltage value (around 80% for example).
Shield wires are normally installed in the lines for reducing the number of faults,
by providing appropriate shielding. The major point in the design is then to locate
the shield wires in the right position. Shield wires may also be used as a communi-
cation medium for control of the converters, their design needs to take both func-
tions into account.
Sustained overvoltages in the DC side of HVDC systems do not occur due to the
intrinsic control process of the HVDC operation. It should be noted that overvolta-
ges in the DC side may appear due to harmonic/filter/smoothing reactor resonance.
It is considered here that this is a problem to be solved by the design of appropriate
elements, and so such kind of stresses will not be considered herein for the insula-
tion design of the DC line.

Determination of Switching Surge Overvoltage (Fault Application)

Switching surge due to fault application in a DC line, being the most important
voltage stress to be applied to its insulation, will be evaluated hereafter.
Modeling
The overvoltages hereinafter are calculated with PSCAD/EMTP (Electromagnetic
Transient Program) using models such as the one shown on Figure 4.49. The data of
the Base Case are here also represented.

• Generator/receiving system
They are modeled as a short circuit power, providing enough power as required.
The short-circuit capacities used are: 9400MVA for single-phase short-circuit
and also for three-phase short-circuit.

• The converter transformers of both terminals are specified in this model as:
Two transformers per pole herein modeled with the following characteristics:
Power = 400 MVA each
4  Electrical Design 145

Mid Point
Smoothing
± 500
reactor
kV Line
500 kV
φ–φ rms

DC DC
Filter Filter

DC DC
Filter Filter

Figure 4.49  HVDC system modeling for fault application calculation (1300 MW).

Reactance xcc = 18%
Turn ratio = 500/199 kV at rectifier or 194 kV at inverter

• DC filters/smoothing reactor
The values used here are (EPRI EL 3892 1985):
Smoothing reactor of 200 mH;
DC filter as below (Figure 4.50)

• Simulation time step = 5  μs

• Converter stations

They are modeled using PSCAD/EMTP blocks. Converter control system are
modeled according to (Szechtman et al. 1991)
There is a system voltage control in the inverter (γmin = extinction angle control),
line current control at rectifier (α = firing angle control). At fault inception in the
HVDC line the current tends to increase and the control acts reducing the line cur-
rent (different from HVAC systems). Also the line protection sensing the fault
change α, γ, to values above 90° and the current goes to zero.

• DC line
The line model is composed of eight sections, each one modeled as lossless line
traveling wave equations. Line losses (resistance) are represented in the model at
section end. Electrical parameters (resistance and inductance) are modeled as
frequency dependent or constant. For more detailed analysis see Section 4.1.7.
146 J.F. Nolasco et al.

Figure 4.50  DC filter

parameters.

0.7 µF 0.7 µF

0.00429 H
0.8 µF 0.053 H

3200 Ω

0.0079 H 0.02992 H

Fault Application Phenomena

For the initiation of the fault in the negative pole, a positive surge of value equal
to the pre-fault voltage is injected in the fault point, and the resulting surge travels
in both line directions, reflecting in the line end and coming back to the fault point.
The traveling wave is coupled to the positive pole resulting in an overvoltage which
values are due to the composition of the forwarded and of the reflected waves.
The maximum overvoltage occurs for a fault initiated in the middle of the line,
within a time close to the travel time to the line end and back to the mid point of the
first reflections. Faults in other locations produce smaller overvoltages. Due to this,
the overvoltage profiles down the line are similar for every line length, as will be
shown later. Line end equipment (filters, smoothing reactor and source) play an
important role, as they define the traveling wave reflection coefficients.

Calculation Results
For the Base Case calculation, the following points were taken into account: a
line 1500 km long; equal sources at both ends (rectifier and inverter) and line
­parameters variable with the frequency.
Figure 4.51 shows the maximum overvoltage profile in the sound pole for a fault
initiated at mid point of the other pole, and Figure 4.52 the voltage versus time in
the mid/end point of the sound pole.
The maximum overvoltage reaches 2.0 pu, however the overvoltages are below 1.6 pu
(20% lower) at 1/4 of the line. Standard deviation for insulation switching surge withstand
is 6%, this means that the overvoltage in the major part of the line does not contribute to
the risk of failure and therefore the line is designed considering mainly the maximum
value (2.0 pu in this case). As example in the insulation coordination calculation the
envelope shown in blue in the Figure 4.51 may be used to address risk of failure.
 4  Electrical Design 147

Fault at the middle of the line

2.00

1.80
Overvoltage (pu)

1.60

1.40

1.20

1.00
0 250 500 750 1000 1250 1500
Line length (km)

Figure 4.51  Overvoltage profile along the sound pole for a fault initiated at midpoint of the other
pole.

Sound pole overvoltage (middle of line) Sound pole overvoltage (3/8 of line)
1200 900
800
1000
700
800 600
Voltage (kV)

Voltage (kV)

500
600
400
400 300
200
200
100
0 0
1.998 2 2.002 2.004 2.006 2.008 2.01 2.012 2.014 2.016 1.998 2 2.002 2.004 2.006 2.008 2.01 2.012 2.014 2.016
time (s) time (s)

Figure 4.52  Fault at mid point, overvoltage in the sound pole (middle and 3/8 of l) (1,500 km
line).

From here on, the line is split in several segments, identified as a fraction of its
length (1/8, 1/4, 3/8 and so on). Figure 4.53 shows the overvoltage profile for fault
initiated at other line positions. It can be seen that all values are below 1.8 pu and so
do not contribute so much to the risk of failure.

4.13.2 Insulation Coordination

This section aims at designing the clearances and at defining the number and type
of insulators to be used in the insulator strings.
The number of insulators is initially selected based on the maximum DC voltage
withstand and on the assumption of a certain pollution level. The number of insula-
tors obtained by these criteria is then verified by considering the overvoltage values.
The clearances to be determined are: conductor-to-tower cross-arm, conductor-­to-­
tower or objects (lateral), conductor-to-ground or objects (at the ground), and con-
ductor to guy wires.
148 J.F. Nolasco et al.

Fault at midpoint of line (4/8 L)

2.00 Fault at 3/8 of line
Fault at 2/8 of line
Fault at 1/8 of line
1.80
Overvoltage (pu)

1.60

1.40

1.20

1.00
0 250 500 750 1000 1250 1500
Line length (km)

Figure 4.53  Overvoltage profiles, Base Case, fault in different positions.

Table 4.25  Clearances for Operating Voltage (kV) Clearance (m)

operating voltages (m)
+500 1.20
+800 1.90

They are calculated for switching surge overvoltage withstand. However, the cle-
arance to tower and guy wires as well as to edge of right-of-way shall be verified in
the condition of insulation string swing due to wind in order to prevent flashovers
and the touch of objects (such as trees) at the border of the right-of-way.

4.13.2.1  Operating Voltage Withstand

Air Clearances
For determining the minimum necessary conductor-structure clearances for opera-
ting voltage insulation, the following premises are considered:

• Withstand voltage regarding the most unfavorable condition: positive polarity,

conductor-to-structure;
• Maximum operating voltage and correction for the atmospheric conditions: 1.15 pu.

The distances conductor-to-structure were obtained according (EPRI 1977) and are
shown on Table 4.25.

Number of Insulators
In AC or DC system the number of insulators in a string is determined by adopting
an environmental condition (pollution, air density) and choosing a creepage dis-
tance criterion for AC or for DC (Figure 4.54).
As example, by using a creepage distance (pole-to-ground) equal 30 mm/kV, the
number of insulators and the respective insulator string lengths are determined and
Reference

EPRI DC very
light moderate heavy
light

20 - 25 25 - 32 32 – 40 mm/kV pole-gr 40 - 70
4  Electrical Design

very light light moderate heavy

SHEE DC Clean (agricult/outskirt) industrial heavily

23 – 40 mm/kV pole-gr 50 70

SHEE AC clean slight moderate severe salt

15 mm/kVrms ph-ph 19.2 25.5 30.9

IEC 6815 very

AC light medium heavy
heavy

12 - 16 mm/kVrms ph-ph 20 25 31

Figure 4.54  Creepage design criteria.

149
150 J.F. Nolasco et al.

Table 4.26  Number of Insulator and String Length

Creepage distance Number of
Operating Voltage (kV) Insulators 30 mm/kV String Length (m) (*)
± 500 30 5.20
± 800 48 8.17
Notes:
The following type of insulator was considered:
- Anti-fog insulator, pitch of 165 mm and leakage distance of 508 mm
- Hardware length: 0.25 m
- Porcelain type or glass. Composite can be used in any area and is robust against vandalism and
pollution.
It should be noted that the suitability of the insulator string length is verified considering switching
surge and the gap conductor cross arm.

shown in Table 4.26. The creepage distances adopted are adequate with a good saf-
ety margin to zones with a pollution level classified as “light contamination”.
For agricultural areas and woodlands 23 mm/kV is recommended, and for outs-
kirts of industrial areas 40 mm/kV is recommended. Some references recommend
as acceptable even lower creepage distances down to 20 mm/kV (for area classified
as with “very light pollution”); however a higher figure is here considered as more
appropriate.
As a reference, the Itaipu lines (“light pollution - agricultural area”) were desi-
gned for 27 mm/kV and have shown adequate performance in more than 20 years of
operation.

Insulator String Swing Angle

The swing angle of the conductor due to wind, as example, was calculated accor-
ding Cigré/IEC (Cigré TB 48 1995) recommendation, using the following data:

• Line altitude = 300  m
• Average temperature:=16 °C
• Minimum ratio of vertical/horizontal span = 0.7
• Wind return period = T = 50  years
• Mean value of the distribution (m/s)= V =18.39 (10 min integration)
• Standard deviation =3.68 m/s
• Wind distribution with 30 years of measurements
• Terrain classification = B

Considering a Gumbel distribution (extreme values), the wind velocity to be con-

sidered, depending on the return period, is determined by:

X
Vt = V + (Y − C2 ) (4.185)
C1

  1 
Y = − ln  − ln  1 −   (4.186)
  T 
4  Electrical Design 151

where:
Vt = Wind velocity (m/s) with return period T.
V = Wind velocity - mean (m/s).
S = Standard deviation (m/s).
C1 = 1.11237 and C2 = 0.53622 are coefficients, for a sample of 30 years [35].
T = return period (years).
It Results = Vt = 29.52  m/s
The insulator string deflection is calculated by:
The swing angle of an insulator may be related to the wind velocity by:

−  ( / 2 )VR2 k D Lw + Fwi / 2 
Φ = tan −1   (4.187)
 Wc + Wi / 2 

In this formula the following symbols are used:
ϱ air density
VR reference wind speed
k correction factor taking into account the effect of wind span
D conductor diameter
Lw wind span
Fwi wind load insulator
Wc effective conductor weight taking into account the differences in the level of
conductor attachments
Wi weight of insulator
VR = 1.05 Vt (5 min integration) = 30.99
Q = 1.18  kg/m3
k = 1.0
D = 0.0382  m
LW = 450  m
By disregarding insulators parameters:
1.18*0.5*(30.99)2 *1*450*0.0382 = 9748.5 Newtons
WC = 2.671  kg/m × 450  m × 0.7(ratio LP/LW) × 9.81 = 8253.8 Newtons
φ = 49.7°
The calculations were done based on (Cigré TB 48 1995) Cigré TB 48, for a set
of ACSR-Aluminum Conductor Steel Reinforced conductors; the results are shown
on Table 4.27.
4.13.2.2  Switching Surge Withstand

Calculation Procedure
Once the switching surge overvoltages (as determined before) are known, the clea-
rances are calculated based on the risk of failure considering the withstand capabi-
lity of the gaps. This is estimated by:
152 J.F. Nolasco et al.

Table 4.27  Swing Angle to be used together with the respective Clearances for the Operating
Voltage
Conductor code Aluminum/steel mm2/mm2 Aluminum MCM* Swing Angle (°)
Joree 1.274/70 2.515 44.5
Lapwing 806/57 1.590 49.7
Bluejay 564/40 1.113 53.4
Tern 403/29 795 56.7
* 1 MCM = 0.5067 mm2
Note: The conductor types and stranding taken as examples in this report can be further optimized
in the case of a real project. There are cases were other conductor types (ASC Aluminum
Conductor; AAC-Aluminum-Alloy Conductor, ACAR – Aluminum Conductor Aluminum-Alloy
Reinforced; AACSR-Aluminum-Alloy Steel Reinforced) may be more adequate. These, however,
will not be covered here. The entire methodology does however apply to them.

V50 = k 500 d 0.6 (4.188)

where:
V50 = Insulation critical flashover (50% probability), (kV)
d = gap distance (m)
k = gap factor:
k = 1.15 conductor – plane
k = 1.30 conductor – structure under
i = 1.35 conductor – structure (lateral or above)
k = 1.40 conductor – guy wires
k = 1.50 conductor – cross arms (with insulator string)
Note: The standard deviation σ of the withstand capability is 6% of the mean.
The latter equation applies to Extra High Voltage System when 2 < d < 5 m.
An alternative equation when 5 < d < 15 m, is:

3400
V50 = k (4.189)
1+ 8 / d
The clearances are determined based on the fault application overvoltage profiles,
aiming at a certain flashover failure risk target (design criteria). It is proposed here
a failure rate of 1 in 50 or 1 in 100 years. It will also be assumed, as design criteria,
that 1 fault per 100 km per year (mainly due to lightning) can occur. The overvolta-
ges shown on Figure 4.51 are used for this purpose. The following steps are carried
out (together with an example):

• Select one line length and one rated voltage (Ex: 1500 km; 500 kV as the Base
Case);
• Select one gap type and size (Ex: conductor-structure lateral = 3.0 m);
• Select the overvoltage profiles in the sound pole for fault in the middle of the
other pole (Ex: maximum value is 2.0 × 500 = 1000  kV);
• Calculate the risk of flashover for the tower in the mid-point of the line for 1 gap (Ex:
The critical flashover value of the gap is V50 = 1.35 × 500 × (3.0)0.6 = 1305  kV) the
4  Electrical Design 153

overvoltage 1000 kV is at (−1305 + 1000)/(0.06*1305) = −3.9 σ, hence correspon-

ding to a risk of failure of 5 × 10−5 (see Gaussian probability (2-9);
• Calculate the flashover risk of failure in the central section (Ex: suppose 300 km,
the envelope of Figure 2-3, or 600 gaps in parallel subjected to the same overvol-
tage of the tower in the mid-point of the line, leading to a risk of failure of
600*5 × 10−6 = 3 × 10−3);
• Extend the flashover risk calculation for parallel gaps (towers) for the whole
overvoltage profile (Ex:1.5 pu, see Figure 4.53 envelope, or 750 kV that is at
(−1305 + 750)/(0.06*1305) = −7.0 σ and the risk < 1.0 × 10−8. Considering 800 km
or 1600 towers the composite risk is <1.6 × 10−5,,therefore negligible contribution
to the value on item V above. For the third step of figure 2-3 the overvoltage is
1.4 pu also not contributing to the risk of failure);
• Repeat calculation of the flashover risks of failure for the gap, for fault at other
points: or send, or 1/8, or 3/8, or 5/8, or 7/8, or receiving end of the line (ex:
disregard this parcels as very few overvoltage values are above 1.6 pu);
• Calculate the weighted flashover average risk of failure, considering that each
profile represents fault occurring in a section of (1/8) of the length of the line
except seeding/receiving end profiles that correspond to (1/2) × (1/8) of the
length. The total flashover risk R is then determined (Ex: (3.0/8) × 10−3);
• Consider the number of occurrences (faults) and determine the probability of
flashover (Ex: 1 fault per 100 km per year or 15 faults per year). Check against
1 in 50 - 100 years (Ex: 15× (3/8) × 10−3 = 0.005 or once in 200 yr); if the flasho-
ver risk is different, then select another gap size (Ex: 2.8 m as the risk can be
increased) and go to step III above;
• Repeat for all gaps.

It should be noted that, if the line is designed with I-strings (as opposed to V-strings),
then it is recommended to consider in the risk calculation the effect of possible
winds simultaneously with the overvoltages.
There are two approaches for taking this point into account: first, by calculating
the clearances for an established risk and admitting that such clearances shall be
maintained with a certain swing due to wind (of about 15°); or second, considering
the simultaneous occurrence of wind and overvoltage, and finally calculating the
composite risk (to lead to 1 failure in 50 years).

Clearances for an Established Flashover Risk of Failure

The following Figures (4.55, 4.56, 4.57, and 4.58), taken from (Cigré TB 388 2009)
in slight different conditions established here, show the clearances for the gaps
above mentioned as a function of the line voltage. They were designed for a flasho-
ver risk of failure of 1/50 yr, no displacement due to wind, and the overvoltages
were calculated using the J. Marti line model and the software EMTP-RV.

Switching Overvoltages with Conductor Displacement due to Wind

Cigré TB 48 (Cigré TB 48 1995) recommends the adoption of a swing angle caused
by a wind intensity corresponding to 1% probability of being exceeded in a year
154 J.F. Nolasco et al.

Figure 4.55  Conductor to Conductor to Tower

8.0
tower clearances.
7.0
1500 km
6.0 3000 km
5.0

Clearance (m)
4.0
3.0
2.0
1.0
0.0
300 400 500 600 700 800
Voltage (kV)

Conductor to Cross-Arm
6.0

5.0
1500 km
4.0 3000 km
Clearance (m)

3.0

2.0

1.0

0.0
300 400 500 600 700 800
Voltage (kV)

Figure 4.56  Conductor to cross-arm clearance.

Conductor to Guy Wires

7.0

6.0
1500 km
5.0 3000 km
Clearance (m)

4.0

3.0

2.0

1.0

0.0
300 400 500 600 700 800
Voltage (kV)

Figure 4.57  Conductor to guy wires clearance.

4  Electrical Design 155

Figure 4.58  Conductor to Conductor-to-ground (object; 4.5m; under)

8.0
object clearance (add 4.5 m
to get Conductor-to-ground 7.0
1500 km
distance). 6.0 3000 km

Clearance (m)
5.0
4.0
3.0
2.0
1.0
0.0
300 400 500 600 700 800
Voltage (kV)

Table 4.28  Swing angle to Swing

be used together with ACSR Conductor code MCM* Angle (°)
Switching Surge Clearances
Joree 2.515 13.4
Lapwing 1.590 15.3
Bluejay 1.113 17.0
Tern 795 18.6
* 1 MCM = 0.5067 mm2

together with the occurrence of switching surge overvoltages. Using the wind distri-
bution as per item 2.2.1, the wind intensity is 13.54 m/s.
The swing angles caused by this wind are shown on Table 4.28.
It should be noted that considering simultaneously: the conductor swing due to
the wind with 1% probability of being exceeded in one year, and the clearances
corresponding to a risk of 1/50 years; the final flashover risk will be much smaller
than 1/50, therefore the stated criteria is conservative.
An alternative approach is to find a clearance considering the composite risk for
overvoltage distribution and a swing due to the wind distribution.
Note: It should be alerted here that the results obtained in this example and others
are applicable only to the parameters used, i.e. wind speed, probability functions, etc.

Composite Risk Calculation

Using this approach a swing angle lower than 8° is obtained.

Atmospheric Conditions
The calculations presented consider standard laboratory test conditions, however
corrections should be considered if atmospheric conditions are different from stan-
dard one.
156 J.F. Nolasco et al.

Table 4.29  Values of Kd Kd

Normal conditions (25 °C, 76 cm ----
Hg)
°C and cm Hg 3.921
°C and Pa 0.00294
IEC normal conditions (20 °C, 76 cm ----
Hg)
°C and cm Hg 3.855
°C and Pa 0.00289

n
 δ
Vcorrect = Vs tan dard   (4.190)
H
where:
δ is the relative air density (RAD);

ρ
δ = Kd (4.191)
273 + t
ρ = pressure of the ambient air (cm Hg or Pa);
t = temperature of the ambient air (°C);
Kd = as in Table  4.29:
H is the humidity correction; factor function of steam pressure that is calculated
using maximum saturated steam pressure, humid bulb temperature, air pressure,
and dry bulb temperature (also a correction curve is needed).
n = exponent function of the gap length
Atmospheric correction has a statistical behavior and this shall be taken into
consideration in the risk of failure calculation. This can be simplified by changing
V50 by the average value of the correction or by changing the gap withstand capabi-
lity standard deviation (σ) by composite standard deviation that includes the stan-
dard deviation of the correction factor.

4.14 Pole Spacing Determination

The pole spacing requirements are determined considering the use of I- or V-strings.

4.14.1 Case of I-Strings

For the pole spacing evaluation, the swing angles of the insulator strings as deter-
mined before will be used.

4.14.1.1  Pole Spacing Required for Operating Voltage

The minimum pole spacing DPTO is:
4  Electrical Design 157

Table 4.30  Assumed Tower Tower

Widths Operating Voltage (kV) Width (m)
±500 1.7
±800 2.5

Table 4.31  Pole Spacing (m) for Operating Voltage/strings

Pole Spacing (m)
ACSR Conductor Cross Section (MCM)* ±500 kV ±800 kV
Joree 2.515 12.5 18.8
Lapwing 1.590 13.1 19.8
Bluejay 1.113 13.6 20.6
Tern 795 14.0 21.1
*1 MCM = 0.5067  mm2

DPTO = ( R + d min + ( L + R ) sin ( θ ) ) 2 + w (4.192)

where:
dmin = Operating voltage clearance;
a
R = bundle radius R =
2 sin ( π / N )
a = subconductor spacing (as general rule, 45 cm is adopted);
N = number of subconductors in the bundle (N = 4 is adopted for all calculations
here), leading to R = 0.32  m;
L = insulator string length;
θ = swing angle for the maximum wind speed with 50 year return period;
w = tower width at conductor level, as per Table 4.30.
The pole spacing values are shown on Table 4.31.

4.14.1.2  Pole Spacing Required for Switching Surges

The minimum pole spacing required for switching surges is calculated in a similar
manner as before, except that the swing angles are those from Table 4.28. The
results for ±800 kV bipole lines are shown on Figure 4.59.
It can be seen that the operating voltage criteria governs the pole spacing for
±800 kV voltages and of course for the other voltages as well.
Therefore, the values of Table 4.31 shall be used as pole spacing for I-string
configurations.

4.14.2 Case of V-Strings

In this case there will be no swing angles due to wind at the towers the clearance
requirements for switching surges will determine the pole spacing. However, the
158 J.F. Nolasco et al.

Pole Spacing for 800 kV

22
OV
21
SS 1500 km

20 SS 3000 km
Pole Spacing (m)

19

18

17

16

15

14
500 1.00 1.500 2.00 2.500 3.000

Conductor Cross Section (MCM)

Figure 4.59  Pole Spacing (±800 kV, 750 to 3000 km). Nomenclature: OV → Operating Voltage;
SS → Switching Surge.

V-strings having length (L) shall be inserted in the tower, meaning that the mini-
mum pole spacing (PSmin) for installation will be:

PSmin = 2 L cos ( 45° ) + w (4.193)

where:
w = tower width;
It is assumed here that the V-string angle is 90 degrees, however this opening can
be reduced.
The pole spacing requirement is otherwise calculated by:

DPTO = ( d min + R ) 2 + w
(4.194)

( provided that DPTO > PSmin )
The results are shown on Table 4.32.
In summary the pole spacing distances are:

• ±500 kV = > 9.3 m
• ±800 kV = > 14.4 m for line length
15.6 for line length equal to < 2250 km
16.8 for line length equal to 3000 km
It should be noted that clearances for insulation is not the only criteria to choose
between I- or V-strings, for instance I-string offers less surface for pollution from
4  Electrical Design 159

Table 4.32  Pole spacing requirements

Operating Clearance Conductor Bundle Tower

Voltage Structure (m) Radius Width Pole Spacing (m)
(kV) 1500 km 2250 km 3000 km (m) (m) 1500 km 2250 km 3000 km PSmin
±500 2.55 2.83 3.06 0.32 1.70 7.4 8.0 8.5 9.3
±800 5.62 6.25 6.81 0.32 2.50 14.4 15.6 16.8 14.3

birds excretion, the Corona protection rings are simpler, and of course are less
expensive as they have less insulators.

4.15 Conductor Current Carrying Capability and Sags

The current carrying capability of ACSR conductors were calculated based on Cigré
recommendation (TB 207), that relates to AC current. It should be noted that the DC
current has a lower heating effect than AC current due to the absence of the transfor-
mer and eddy current effects, however this will not be considered here. Therefore, the
methodology of calculation can be considered as the same for both AC and DC lines.
The following assumptions are considered:

• Wind speed (lowest) 1 m/s

• Wind angle related to the line 45 degree
• Ambient temperature 35 °C
• Height above sea level 300 to 1000 m
• Solar emissivity of surface 0.5
• Cond. solar absorption coefficient 0.5
• Global solar radiation 1000 W/m2

The maximum temperature of the conductor will be limited here to 90 °C (as
design criteria commonly used in many countries) for steady state and in emergency
or short duration conditions, although it could be accepted temperatures even above
100 °C for non-special conductors (Thermal Resistant Conductors may withstand a
higher temperature in the steady state condition). However, the conductor is selec-
ted based on economic criteria (cost of line plus losses) leading to a maximum
operating temperature in normal conditions much lower (~55 to 60 °C). Therefore
90 °C will eventually apply to pole conductors at abnormal conditions as well as to
electrode lines and metallic return conductors. Figure 4.60 shows the current capa-
bility for some conductors, so that the corresponding values for intermediate sizes
can be interpolated.
The sags are presented on Figure 4.61 for conductor temperatures in the range
from 50 to 90 °C. The sag calculation was based on the following conditions:

• Span = 450  m
• EDS = Every Day Stress condition
160 J.F. Nolasco et al.

Figure 4.60 Conductor Conductor Current Carrying Capability

2500
Current Carrying
90°
Capability for alternatives 70°
2000
maximum temperature 60°
50°
criteria.

Current (A)
1500

1000

500

0
0 500 1000 1500 2000 2500 3000
Conductor Cross Section (MCM)

Figure 4.61 Conductor Conductor Sag

23
sags.
22
50°
60°
21 70°
Sag (m)

90°
20

19

18

17
500 1000 1500 2000 2500 3000
Conductor Cross Section (kcmil)

• Tension of 20% of the RTS (this is a simplification - ideally the EDS should be
selected based on fixed H/w horizontal-tension/weight, the catenary’s parameter);
• Temperature: 20 °C.

It can be seen that the sags vary from 18 to 22 meters, depending on the conductor
temperature and type of conductor. It should be noted that the conductors conside-
red in this graph are those of the tables shown before. Conductors with the same
aluminum but different steel contents will have different sags.

4.16 Tower Height

The following distances are defined hereunder.

The conductor height at the tower (hp) is:
hp = CS + sc + Ext + R (4.195)

where:
hp = distance from the center of the bundle to ground at tower;
CS = clearance to ground at mid-span =12.5 and 19.5 m for ±500 and ±800 kV,
respectively, determined by electric field criteria;
4  Electrical Design 161

Table 4.33  Conductor and Voltage (kV) hp (m) hg (m)

shield wire heights at tallest
±500 42.8 50.8
tower (Two shield wires; for
one add 2.5 m to hg) ±800 50.8 61.8

sc = conductor sag at 90 ° C (criteria adopted), as per Figure 4.61 (22 m adopted

for all conductors in this clause);
R = bundle radius;
Ext = tower extensions up to 3 × 3  m = 9  m
The shield wire height (hg) at the tower is:

hg = hp + R + dis + DG (4.196)

where:
dis = insulator string and hardware length: 5.2 and 8.17 m for ±500 and ±800 kV,
respectively;
The assumed values for shield wire to cross arm distance DG are:
DG = 2.5 m (for the case of two shield wires),
or
DG = 5 m (for the case of only one shield wire).
Table 4.33 shows the values to be used in the calculations which follow.

4.17 Lightning Performance

In order to get a good performance under lightning strokes, the design of HVDC
lines should include the use of shield wires (one or two).
The shield wires reduce the direct strokes to the conductors. For the strokes that
hit the shield wires, there will be an overvoltage that is coupled to the pole conduc-
tors and can cause flashovers or not.
To set a good design, some conditions shall be considered:

• The current of the stroke that hit the pole conductors should not produce an over-
voltage greater than the insulation withstand of the line.
• The closer are the shield wires to the pole conductor, the better will be the per-
formance due to strokes hitting the shield wires.
• The tower footing resistance and the corresponding tower footing surge impe-
dance should be low, therefore requiring the use of an adequate grounding sys-
tem, generally counterpoises at the towers.

In regions with ice, the second condition may be conflicting with the requirements of
keeping a safety distance from the shield wire to the pole conductors during icing events.
The clearances at the tower are designed to withstand switching overvoltages
with a pre-established risk of failure, or the operating voltage.
Once defined the required clearances, the Critical Impulse Flashover Capability
E of the insulation (50% probability) for lightning surges (fast front overvoltages)
are known.
162 J.F. Nolasco et al.

With E, Vop- the operating voltage- and the conductor surge impedance Z, the
critical “threshold current” Ioc, into the conductor for which a flashover will start is
determined by:

2 ( E − Vop )
Ioc = (4.197)
Z

The striking distance rsc is a function of Ioc and is calculated by:

rsc = k 6.7 I 0oc.8 (4.198)

where:
rsc = in (m)
Ioc = in (kA)
k is a factor different from 1 eventually adopted for shield wires or ground.
The horizontal distance X between conductor and shield wire is:

X = rsc  1 − ( k − T ) − 1 − ( k − R ) 
2 2
(4.199)
 
where:
rsc = striking distance (m)
k = factor
T = hg*/rsc
R = hp*/rsc
hg* = average shield wire height (m)
hp* = average conductor height (m)
Three types of terrain may be considered, namely:

• Flat: in this case the following parameters are used in the equations above.

h*p = hp − Sc ( 2 / 3 ) (4.200)

h = hg − Sg ( 2 / 3 )
*
g (4.201)

hp, hg are conductor or shield wire heights at tower; and Sc, Sg are the conductors and
shield wire sags.
• Rolling: in this case:

h*p = hp (4.202)

b* = ( hg − hp ) + ( Sc − S g ) ( 2 / 3 ) (4.203)

hg* = h*p + b* (4.204)

• Mountainous
hp* = 2 hp (4.205)

4  Electrical Design 163

Table 4.34  Protection for direct strokes

2 shield wires
Voltage (kV) E (kV) hg* (m) hp* (m) Ioc (kA) rsc (m) X (m) θ (°)
±500 3000 55.2 42.8 14.3 56.2 1.6 11.5
±800 4850 66.2 50.8 23.1 82.7 4.7 23.3

hp*, hg* as in the rolling case.

In this report the evaluations will be done considering rolling terrain, average
tower (no extensions) and k = 1.
The protection angle θ is then:

 X 
θ = arc tan   (4.206)
h −h
 g p 

The line surge impedance Z is assumed here as 350 Ω.

When the lightning activities are low (and on icing regions where it is desired
that the shield wires should not be in the same vertical line as the conductors), one
shield wire may be a preferable design for economical reasons.
From Table 4.34 it can be seen that the minimum protection angle θ can be set at
values from 11 to 23 degrees. The closer are shield wires to the conductors, the
better is the lightning performance for back flashovers due the higher coupling
factor.
As a consequence, the protection angle can be adopted as 10 degrees, when using
two shield wires.
Note: Only EHS steel wire is considered for shielding purpose. However other
material or characteristics may be used if one intend for instance to provide dual
function: lightning shielding and communication (carrier or fiber optics).

4.18 Right-of-Way Requirements for Insulation

The Right-of-Way width (ROW) is defined considering the following aspects:

Conductor swing and clearances to objects at the border of ROW, Corona and field
effects. At this point, only the first condition is examined and the results will be
partial.
In the ROW determination, clearances for operating voltage and I-type insulator
string length are used.
The swing angles are calculated using the same parameters as clauses before,
except that the ratio vertical to horizontal span is equal 1.0, and the span length
should not exceed 600 m. It should be reminded that the wind intensity corresponds
to 50 year return period. The swing angles are shown on Table 4.35.
The conductor sags (Table 4.36) were obtained by starting from EDS conditions
and considering the wind load with the coincident temperature.
164 J.F. Nolasco et al.

Table 4.35  Swing angles Conductor Swing

for ROW width definition Angle
ACSR Code Section (MCM)* (degree)
Joree 2515 34.1
Lapwing 1590 39.1
Bluejay 1113 43.5
Tern 795 47.5
* 1 MCM = 0.5067 mm2

Table 4.36  Sags for ROW Conductor

width definition
ACSR Code Section (MCM)* Sag (m)
Joree 2515 36.5
Lapwing 1590 34.9
Bluejay 1113 34.5
Tern 795 33.6
* 1 MCM = 0.5067 mm2

Table 4.37  Right Of Way (I-strings) in (m)

Conductor
ACSR Code Section (MCM)* ±500 kV ±800 kV
Joree 2515 62.1 73.2
Lapwing 1590 66.7 78.5
Bluejay 1113 71.3 83.8
Tern 795 74.3 87.2
* 1 MCM = 0.5067 mm2

4.18.1 Line with I-Strings

The minimum ROW when using I-strings is determined by:

ROW = ( R + L + S ) sin θ + d min  2 + PS (4.207)

Where:
dmin = operating voltage clearance
R = bundle’s radius (m)
L = insulators string length
S = conductor sag
θ = swing angle due to wind (50 year return period)
PS = pole spacing
Table 4.37 shows the ROW width as function of the conductor type.

4.18.2 Line with V-Strings

The minimum ROW widths (“V strings”) are calculated according to the same equa-
tion before but disregarding insulator string length. The results are shown in Table 4.38.
4  Electrical Design 165

Table 4.38  Right of Way (V-Strings)

ACSR Conductor SECTION ±500 kV 750 ±800 kV
CODE (MCM)* to 3,000 Km <2,250 km 2,250 km 3,000 km
Joree 2515 53.2 59.6 60.7 61.8
Lapwing 1590 56.3 62.7 63.8 65.0
Bluejay 1113 59.9 66.3 67.4 68.5
Tern 795 61.9 68.4 69.5 70.6
*1 MCM = 0.5067  mm2

Note that the results (for I- or V-strings) are partial as Corona effects were not
yet considered. Also note that only horizontal design is considered (vertical design
will led to smaller ROW).

4.19 Corona effects

4.19.1 Conductor Surface Gradient and onset Gradient

4.19.1.1  Conductor Surface Gradient

The parameter that has the most important influence on Corona performance is the
conductor surface electric field or conductor surface gradient. Electrostatic princip-
les are used to calculate the electric field on the conductors of a transmission line.
The procedure of calculation may be the same indicated for AC lines. However
for DC lines simplified approach can also be used.
When bundled conductors are used, the electric field around the sub-conductors
of the bundle is distributed non-uniformly, with maximum and minimum gradients
occurring at diametrically opposite points and the average gradient at a point in
between. Using the method known as Markt and Mengele’s method, the average
and maximum bundle gradients of a bipolar HVDC line, with n-conductor bundles
on each pole, are given as (Maruvada 2000).
V
Ea = (4.208)
  2 
 2H 
n r l n  2H /  req   +1
   S  
  
where:
V = voltage applied (actually ± V) to the conductors of the line, kV
r = conductor radius, cm
H = conductor height, cm
S = pole spacing, cm

 r
Em = Ea 1 + ( n − 1)  (4.209)
 R
Where:
r = sub-conductor radius, cm
R = bundle radius, cm
166 J.F. Nolasco et al.

req = equivalent bundle radius, cm

a
R= (4.210)
2 sin ( π / N )

1
n r  n
req = R   (4.211)
 R 

a = distance between adjacent subconductors, cm

Equations above give reasonably accurate results for the maximum bundle gra-
dient, for n ≤ 4 and for normal values of H and S.
Consider the line geometry indicated below as Base Case example for calcula-
tions in this session.

Voltage ± 500 kV
Conductor MCM 3 × 1590
Code Lapwing
Diameter 3.822 cm
Bundle spacing 45 cm
Pole spacing 13.0 m
Minimum conductor-ground clearance 12.5 m

The conductor surface gradient (conductor considered parallel to ground at mini-

mum height).

45
R= = 26 cm (4.212)
2 sin ( π / 3 )

1
 31.911  3
req = 26   = 15.7cm (4.213)
 26 

500
Ea =
21250
31.911 ln
2
 2 * 1250  (4.214)
15.7  +1
 1300 

= 20.297 kV / cm

 1.911 
Em = 20.297 1 + (3 − 1) = 23.28 kV / cm (4.215)
 26 

As a general formulation the equations of the electrostatic phenomena described in

Section 4.1 can be used. The charge-voltage equation in the matrix form is:

[V ] = [ H ][Q ] (4.216)
4  Electrical Design 167

where:
V = voltages on the conductors and shield wires [kV]
Q = charges [kV*F/km]
H = Maxwell’s potential coefficients [km/F]
The inverse equation is:

[Q ] = [C ][V ] (4.217)
Where:
C = the admittance coefficient (F/km).
Once known the charge in a bundle Qb the average charge in one sub conductor
is Qb/n and the average field is
Qb / n
Ea = (4.218)
2 π ε rc

rc = sub conductor wire radius
The maximum electric field Em is then calculated by the equation above.
When looking for the electric field in the shield wires the procedure used is as
follows.
If the shield-wires are grounded at the towers, their voltages are zero and their
charges are calculated by:
QSw1 = CSW 1− C1 V+ + CSW 1− C 2 V− (4.219)

Where CSW1−C1 is the mutual coefficient between shield wire 1 and pole 1(positive)
and CSW1−C2 from shield-wire 1 and pole 2 (negative).
The electric field in the shield wire surface is:
QSW 1
ESW 1 = (4.220)
2π ε r

where:
r = shield wire radius
1 1
ε= 10 −9 F / m or 10 −6F / km
36 π 36 π

In reference (Cigré TB 388 2009) it is found a calculation of shield wire surface gradient
ESW for a  ± 800 kV two twelve pulse converter per pole. In normal operation condition,
for both poles at 800 kV, ESW = 13.7 kV considering the shield wires and conductors
position as it is in the tower. However, ESW = 13.7 kV, if the calculation is carried with the
position as it is in the mid-span. When in one pole the voltage is 800 kV and in the other
400 kV (emergency of one converter) ESW = 18.9 kV and with 800 kV and zero isthe
other pole ESW = 24.1 kV (conductors and shield wire position as they are in the tower).

4.19.1.2  Corona Onset Gradient

When the electric field at the surface of a transmission line conductor exceeds a
certain value, partial electrical breakdown of the surrounding air takes place, giving
rise to Corona discharges.
168 J.F. Nolasco et al.

500 kV Surface Gradient

40

35
1 cond
2 cond
30 3 cond
4 cond
Ec
kV/cm

25 0.95Ec

20

15

10
1 1.2 1.4 1.6 1.8 2 2.2 2.4
Radius (cm)

Figure 4.62  Conductor Surface Gradients for ±500 kV.

For the Base Case example in this session, 3 × 1590 MCM conductor Lapwing
(1 MCM = 0.5067  mm2) and assuming m = 0.82; δ = 0.915; r =1.911 cm, the critical
gradient is:

 0.301 
Ec = 30 ⋅ 0.82 ⋅ 0.915 ⋅ 1 +  = 27.6 kV / cm
 0.915 ⋅ 1.911  (4.221)

Reference (Cigré TB 388 2009) shows graphs indicating conductor surface gradient
(kV/cm) and critical gradient Ec as function of conductor radius. The figures for 500
and 800 kV are reproduced here (Figures 4.62 and 4.63).

4.19.2 Corona Loss

Corona losses on both AC and DC transmission lines occur due to the movement of
both positive and negative ions created by Corona. However, there are basic diffe-
rences between the physical mechanisms involved in AC and DC Corona loss
(Maruvada 2000; Cigré TB 1974). On AC lines, the positive and negative ions cre-
ated by Corona are subject to an oscillatory movement in the alternating electric
field present near the conductors and are, therefore, confined to a very narrow region
around the conductors. On DC lines, however, ions having the same polarity as the
conductor move away from it, while ions of opposite polarity are attracted towards
the conductor and are neutralized on contact with it. Thus, the positive conductor in
Corona acts as a source of positive ions which fill the entire space between the con-
ductor and ground, and vice-versa, for the negative conductor.
4  Electrical Design 169

800 kV Surface Gradient

40

35 3 cond
4 cond min PS
5 cond
30 6 cond
kV/cm

Ec
0.95Ec
25

20

15
1 1.2 1.4 1.6 1.8 2 2.2 2.4
Radius (cm)

Figure 4.63  Conductor Surface Gradients for ±800 kV.

The case more widely used is the bipolar HVDC transmission line, the positive
and negative conductors in Corona emissions having the same polarity as the res-
pective conductor. Unipolar space charges fill the space between each pole and
ground while ions of both polarities mix in the bipolar region between the two poles
and are subject to some amount of recombination.
Theoretical calculation of Corona losses from HVDC transmission lines requires
analysis of the complex electric field and space charge environment in the unipolar
and bipolar regions. Such an analysis determines in the first step the electric field
and ion current distributions on the surface of the conductors and ground plane and
then evaluating Corona losses of the line. Ambient weather conditions have a large
influence on Corona losses from the line. The losses are lower under fair weather
conditions than under foul weather conditions such as rain, snow etc. However, the
ratio of foul weather to fair weather Corona losses on a DC line is much lower than
in the case of an AC line.
Because of the complexity of theoretical calculations and the large number of
factors influencing Corona on practical HVDC transmission lines, it is often prefer-
able to obtain empirical formulas derived from a large amount of data on long-term
Corona loss measurements made on experimental lines with different conductor
bundles and under different weather conditions. However, the amount of data avai-
lable for CL from DC lines is much more limited than in the case of AC lines and,
consequently, the accuracy and applicability of empirical formulas may be limited.
For unipolar DC lines, Corona losses may be calculated using an empirical for-
mula derived from measurements made on an experimental line in Sweden (Knudsen
et al. 1974), which is given as:
0.25( g − g0 )
P = Vu kc nrc 2 10 −3 (4.222)

170 J.F. Nolasco et al.

where:
P = Corona loss, kW/km
Vu = line voltage, kV
n = number of sub-conductors in the bundle
rc = sub-conductor radius, cm
g = maximum bundle gradient, kV/cm
g0 = reference value of g, and kc is an empirical constant
The reference value is given as g0 = 22 δ kV/cm, where δ is the relative air den-
sity. The empirical constant is given as kc = 0.15 for clean and smooth conductors,
kc = 0.35 for conductors with surface irregularities and kc = 2.5 for the calculation of
all-weather Corona losses.
For bipolar DC transmission lines, some empirical formulas have been developed
for Corona losses in different seasons of the year and under different weather con-
ditions. However, the following empirical formulas are recommended since they are
derived using available experimental data from a number of different studies
(Corbellini et al. 1996), for evaluating fair and foul-weather Corona losses of bipo-
lar HVDC transmission lines:

 g d
Pfair = P0 + 50 log   + 30 log  
 g0   d0 
(4.223)
 n  H S 
+20 log   − 10 logg 
 n0   H 0 S0 

 g d
Pfoul = P0 + 40 log   + 20 log  
g
 0  d0 
(4.224)
 n  H S 
+15 log   − 10 logg 
 n0   H 0 S0 

Where P is the bipole Corona loss in dB above 1 W/m, d is conductor diameter in

cm and the line parameters g (conductor surface gradient), n (number of conduc-
tors), H (height) and S (pole spacing) have the same significance as described above.
The reference values assumed are g0 = 25  kV/cm, d0 = 3.05  cm, n0 = 3, H0 = 15  m and
S0 = 15 m. The corresponding reference values of P0 were obtained by regression
analysis to minimize the arithmetic average of the differences between the calcula-
ted and measured losses. The values obtained are P0 = 2.9 dB for fair weather and
P0 = 11 dB for foul weather.

P ( W / m ) = 10 P ( dB) /10
bipole losses in W / m

In the economic evaluation it will be considered 80% of time fair-weather and 20%
as foul-weather.
4  Electrical Design 171

For the Base Case example:

 23.28   3.822 
Pfair = 2.9 + 50 log  + 30 log 
 25   3.05 
 3  12.513 
+20 log   − 10 log  = 5.7
 3  15 ⋅ 15 

Pfair ( W / m ) = 10 = 3.7 W / m 5.7 /10

Similarly: Pfoul (W/m) =20.6 W/m and Ptot = 0.8 · 3.7 + 0.2 · 20.6 = 7.1  W/m.

4.19.3 Radio Interference and Audible Noise

While Corona losses occur due to the creation and movement of ions by Corona on
conductors, radio interference and audible noise are generated by the pulse modes
of Corona discharges. The current pulses induced in the conductors and propagating
along the line produce RI, while the acoustic pulses generated by these modes of
Corona and propagating in ambient air produce AN.
The characteristics of Corona-generated RI and AN on DC transmission lines dif-
fer significantly from those on AC lines. Firstly, while all three phases of an AC line
contribute to the overall RI and AN of the line, only the positive pole of a DC line
contributes to the RI and AN level. Secondly, the RI and AN levels of DC transmis-
sion lines under foul weather conditions such as rain etc., which produce rain drops on
conductors, are lower than those under fair weather conditions. This is contrary to the
case of AC lines on which foul weather conditions produce the highest levels of RI
and AN, much higher than in fair weather. These two distinguishing features play
important roles in predicting the RI and AN performance of DC transmission lines
and in establishing the design criteria necessary for conductor selection.

4.19.3.1  Radio Interference

Both analytical and empirical methods may be used for calculating the RI level of
DC transmission lines.
Some empirical methods have been developed for predicting the RI level of DC
transmission lines under different weather conditions. Based on data obtained on
experimental as well as operating lines, a simple empirical formula (for the bipo-
le-­as negative pole contribution can be neglected) has been developed (Cigré TB
1974; Chartier et al. 1983) for predicting the average fair weather RI level for bipo-
lar HVDC transmission lines as:

g d
RI = 51.7 + 86 log + 40 log +
g0 d0
(4.225)
19.9 q
10 1 − log (10 f ) + 40 log
2
+
D 300
172 J.F. Nolasco et al.

Where:
RI = radio interference level measured at a distance D from the positive pole with
a CISPR instrument, dB above 1 μV/m
g = maximum bundle gradient, kV/cm
d = conductor diameter, cm
f = frequency, MHz
D = radial distance from positive pole, m
The contribution of the negative pole is 4 dB lower. The noise under foul weather
is 3 dB lower
The reference values are g0 = 25.6  kV/cm and d0 = 4.62  cm.
Adequate statistical information is not presently available to determine the diffe-
rence in the RI level between the average and maximum fair-weather values or
between the fair and foul-weather values.
However, based on the results of some long-term studies (Maruvada 2000), the
maximum fair weather RI may be obtained by adding 6 dB and the average foul
weather RI may be obtained by subtracting 5 dB from the average fair-weather value.
Design criteria for RI from transmission lines are generally based on signal-to-­
noise ratios (SNR) for acceptable AM radio reception, similarly as presented for AC
line (Section 4.1).
For the Base example the following value is obtained:
23.28 3.822
RI = 51.7 + 86 log + 40 log +
25.6 4.62
19.9 600
10 1 − log 2 (10 ⋅ 1) + 40 log + = 41.8 dB
30 300

Note that it was assumed: f = 1  MHz; D = 30  m; q = 600  m.

4.19.3.2  Audible Noise

As in the case of RI, analytical treatment of AN from transmission lines requires
knowledge of a quantity known as generated acoustic power density, which can be
obtained only through extensive measurements on an experimental line using a
number of conductor bundles and carried out in different weather conditions.
Based on measurements made on experimental as well as operating DC lines and the
general characteristics of Corona-generated AN, an empirical formula has been developed
(Chartier et al. 1981) for the mean fair weather AN, in dBA, from a DC line as:
AN = AN 0 + 86 log ( g ) + k log ( n ) +
q (4.226)
40 log ( d ) − 11.4 log ( R ) +
300
where:
g = average maximum bundle gradient, kV/cm
n = number of sub-conductors
d = conductor diameter, cm
R = radial distance from the positive conductor to the point of observation
4  Electrical Design 173

The empirical constants k and AN0 are given as:

k = 25.6 for n > 2

k=0 for n = 1, 2
AN 0 = −100.62 for n > 2
AN 0 = − 93.4 for n = 1, 2

The maximum fair-weather AN (probability 10% of not being exceeded) is calcu-

lated by adding 5 dBA to the mean fair weather value obtained above, while the
mean AN during rain is calculated by subtracting 6 dBA from the mean fair
weather AN.
As in the case of RI, there are presently no regulations for AN from HVDC trans-
mission lines. The Environmental Protection Agency (EPA) in the US recommends
that the day-night average sound level Ldn (U.S. EPA 1974) be limited to 55 dBA
outdoors. The level Ldn is defined as:

 1  Ld Ln +10
 
Ldn = 10 log  15 ⋅ 10 10 + 9 ⋅ 10 10   (4.227)
 24   
where Ld and Ln are the day and night time sound levels, respectively. However,
since the highest level of AN from DC lines occurs in fair weather, it may be pru-
dent to limit the Ldn (10%) of AN from HVDC transmission lines to 55 dBA, and
this corresponds to 50 dBA for Ldn(50%). Reference (Chartier et al. 1981) indicates
that the night and the all time distribution are close together by 1.5 dBA. Therefore
assuming Ld = Ln = 42 to 44 dBA, it results Ldn ~ 50 dBA.
As a conclusion, the AN calculated by the equation above (average value) shall
be limited to ~42 dBA at the edge of the right-of-way.
For the Base Case example the following value is obtained:

AN = −100.62 + 86 log ( 23.28) + 25.6 log (3) +

600
40 log (3.822 ) − 11.4 log( 12.52 + (30　 6.5) ) +
2

300
AN = 38.2 dBA
Note that: D = 30  m; q = 600  m
In the Figure 4.64 the values of AN and RI as function of the lateral distance are
shown.
It can be seeing that RI will govern the right-of-way width requirements. It
should be remembered that the conductor position in the calculations were those in
the mid-span. Sometimes the equivalent distance to ground (minimum distance plus
1/3 of the sag) may be used.

4.19.3.3  Final ROW Width

The final right-of-way of a HVDC line is chosen as the largest requirements for
insulation coordination and Corona effect.
174 J.F. Nolasco et al.

RI and AN function of distance from tower center

60
3x1590 MCM → RI
55 3x1590 MCM → AN
AN criterium
50 RI criterium

45

40
dB

35

30

25

20

15
0 10 20 30 40 50 60
Distance from tower center (m)

Figure 4.64  RI and AN values for the Base Case example and criteria.

Figure 4.65 (from (Cigré TB 388 2009)) illustrates what defines the (1/2 ROW)
for ±500 kV, 3 conductors per pole. In this case, RI governs for conductors larger
than 1400 MCM (insulation requirements are always smaller in this case).

4.20 Electric and Magnetic Field

4.20.1 Ground-Level Electric Field and Ion Current

4.20.1.1  Introduction
Induction effects under AC transmission lines are defined mainly in terms of the
magnitude and frequency of the alternating electric fields at the ground level. In the
case of DC transmission lines, however, the magnitudes of both the electric field and
the Corona-generated ion currents at ground level are required to characterize any
induction effects.
Corona-generated ion space charge fills the entire space between the conductors
and the ground plane. In the cases of both unipolar and bipolar DC transmission
lines, only positive or negative unipolar space charge exists at ground level. The
combined presence of DC electric field and ion space charge is generally known as
space charge field (Maruvada 2000).
Both unipolar and bipolar space charge fields are defined in terms of a set of
coupled non-linear partial differential equations. Solution of these equations, with
appropriate boundary conditions, provides a description of the electric field, space
charge density and ion current density at every point and, consequently, at the sur-
face of the ground plane.
Unipolar DC space charge fields are defined by the following equations:
4  Electrical Design 175

RoW 3 conductor per pole 500 kV

75
RI (m)
65 NA (m)

gradient (kV/cm)
55
m or kV/cm

insulation clearance (m)

45

35

25

15
1000 1500 2000 2500
MCM

Figure 4.65  Half ROW requirements and gradient for ±500 kV (bipole having three conductors
per pole).

ρ
∇E = (4.228)
ε0

J = µρ E (4.229)

∇J = 0 (4.230)

where E and J are the electric field and current density vectors at any point in
space, ρ is the space charge density, μ is the ionic mobility and ε0 is the permitti-
vity of free space. The first is Poisson’s equation, the second defines the relations-
hip between the current density and electric field vectors, and the third is the
continuity equation for ions. Solution of these equations, along with appropriate
boundary conditions, for the conductor-ground-plane geometry of the HVDC
transmission line, determines the ground-level electric field and ion current distri-
butions (Maruvada 2000).
Corona activity on the conductors and the resulting space charge field are influ-
enced, in addition to the line voltage and geometry, by ambient weather conditions
such as temperature, pressure, humidity, precipitation and wind velocity as well as
by the presence of any aerosols and atmospheric pollution. It is difficult, if not
impossible, to take all these factors into account in any analytical treatment of space
charge fields. Information on the Corona onset gradients of conductors, which is an
essential input in the analytical determination of electric field and ion current
environment, is also difficult to obtain under practical operating conditions. For
these reasons, it is necessary to use analytical methods in combination with accurate
long-term measurements of ground-level electric field and ion current distributions
under experimental as well as operating HVDC transmission lines, in order to
develop prediction methods.
176 J.F. Nolasco et al.

4.20.1.2  Calculation Methods

• The first method for solving equations above for multiconductor dc transmission
line configurations was developed by Maruvada and Janischewskyj (Maruvada
2000). The method, originally developed to calculate Corona loss currents,
involves the complete solution of the unipolar space charge modified fields and,
consequently, the determination of the ground-level electric field and ion current
density distributions. The method of analysis does not include the influence of
wind.
The method of analysis is based on the following assumptions:
–– The space charge affects only the magnitude and not the direction of the elec-
tric field
–– For voltages above Corona onset, the magnitude of the electric field at the
surface of the conductor in Corona remains constant at the onset value.

The first assumption, often referred to as Deutsch’s assumption, implies that the
geometric pattern of the electric field distribution is unaffected by the presence of
the space charge and that the flux lines are unchanged while the equipotential lines
are shifted. Since HVDC transmission lines are generally designed to operate at
conductor surface gradients which are only slightly above Corona onset values,
Corona on the conductors generates low-density space charge and the ions may be
assumed to flow along the flux lines of the space-charge-free electric field. This
assumption is much more valid for dc transmission lines than for electrostatic pre-
cipitators where Corona intensity and space charge densities are very high.
The second assumption, which was also implied in Townsend’s analysis, has
been justified from theoretical as well as experimental points of view.

• Gela and Janischewskyj (Janischewsky et al. 1979) developed the first Finite
Element Method (FEM) for solving the unipolar space charge modified field
­problem without recourse to Deutsch’s assumption. This method has been used
and improved by many authors. However the method is quite complex and diffi-
cult to be used for line designers
• A simplified method of analysis was developed at the Bonneville Power
Administration (BPA) for determining the ground-level electric field and current
density under unipolar and bipolar dc transmission lines. In addition to the two
assumptions mentioned, other simplifying assumptions were made to develop
the computer program ANYPOLE which was made available in the public
domain. One of the simplifying assumptions made in this program was the repla-
cement of bundled conductors by an equivalent single conductor.

The input data and the results of calculation for the Base Case example follows
(Figures 4.66 and 4.67).
According to (Chartier et al. 1981) the values obtained using the default values
for Corona and Ion information lead to results with 90% probability of not being
exceeded.
4  Electrical Design 177

Figure 4.66  Input data Base Case example.

• Empirical methods are generally derived from extensive experimental data

obtained preferably on operating HVDC transmission lines, but also on full-scale
test lines. In order to derive valid and accurate methods, the experimental data
should be obtained for lines of different voltages and configurations with a wide
range of values for parameters such as conductor bundle, conductor height and
pole spacing in the case of bipolar lines. The validity is usually restricted to the
range of values of line parameters for which the experimental data, used to derive
the empirical method, was obtained.

A semi-empirical method, called the “degree of Corona saturation” method, was pro-
posed (Johnson et al. 1987) for calculating ground-level electric fields and ion currents
under bipolar dc lines. The basic principle of the method is given by the equation,

Q = Qe + S (Qs − Qe ) (4.231)

where Qe is the electrostatic value of any parameter (electric field, ion current den-
sity or space charge density), QS is the saturated value of the parameter and S is the
degree of saturation. The electrostatic value Qe of the parameter can be calculated
178 J.F. Nolasco et al.

Electric field
30.0
Electric field
Y axis: Field with charge (kV/m)

20.0 500 kV, Triplex

-500 kV, Triplex
10.0

0.0

-10.0

-20.0

-30.0

-40.0
-60.0 -40.0 -20.0 0.0 20.0 40.0 60.0
X axis: Distance to Center

Ion current
60.0
Ion current
40.0 500 kV, Triplex
-500 kV, Triplex
Y axis: Ion current (nA/m2)

20.0

0.0

-20.0

-40.0

-60.0

-80.0

-100.0
-60.0 -40.0 -20.0 0.0 20.0 40.0 60.0
X axis: Distance to Center

Figure 4.67  Results of calculation for the Base Case example.

using the well-established electrostatic field theory. It should be noted that the elec-
trostatic values of current density and charge density are zero.
Equations were derived for the saturated values of QS of the electric field, ion
current density and charge density, based on laboratory tests on reduced-scale
models with thin wires of unipolar and bipolar dc line configurations. The degree of
Corona saturation factor S was derived from full-scale tests carried out on a number
of bipolar dc line configurations.
The reference (Johnson et al. 1987) presents the equations and parameters to
carry on the calculations and are reproduced here

• Maximum saturated values (positive or negative), within right of way (in the
ground, close to conductors, bipolar lines) [4.6].
4  Electrical Design 179

(
E = 1.31 1 − e −1.7 P / H V / H ) (4.232)

( )
J + = 1.65 × 10 −15 1 − e − 0.7 P / H V 2 / H 3 (4.233)

J − =　 2.15 × 10 −15 (1 − e − 0.07 P / H
)V 2
/ H3 (4.234)

P = pole spacing (m)

H = conductor height (m)
V = Voltage (kV)
E = Electric field (kV/m)
J = Ion flow (A/m2)

• Maximum saturated values in the ground, at any distance “x” from the tower
center provided that 1 < (x − P/2)/H < 4

−　　0.7 ( x −　P / 2 ) / H
E = 1.46 1 − e −　2.5 P / H  .e V /H (4.235)

− 1.75( x − P / 2 ) / H
J + = 1.54 10 −15 1 − e − 1.5 P / H  .e V2 / H3 (4.236)

− 1.75( x − P / 2 ) / H
J − = 2. 10 − 15 1 − e − 1.5 P / H  .e V2 / H3 (4.237)

• Electrical Field without space charge

2V H
E=
 4 H  1  4 H 2 + P2 
n   − n  
 Deq  2  P2  (4.238)
 1 1 
 2 − 2 
 H + ( x − P / 2 ) H + ( x + P / 2 ) 
2 2

• Saturation factor

− k ( G − Go )
S = 1− e (4.239)

k = empirical coefficient
G = surface gradient (kV/cm)
Go = empirical coefficient

• Values considering the degree of saturation

180 J.F. Nolasco et al.

Table 4.39  Parameters Go and k for the weather conditions

summer spring fall summer summer
fair fair fair humidity fog rain snow
Go 9 14.5 12 7.5 6 12
positive 50% K 0.037 0.041 0.039 0.06 0.058 0.03
Go 3 11 10 3 6 11
positive 95% K 0.067 0.086 0.092 0.086 0.087 0.045
Go 9 14.5 12 8.5 6 12
negative 50% K 0.015 0.021 0.017 0.045 0.058 0.03
Go 3 11 11 3 6 11
negative 95% K 0.032 0.065 0.07 0.063 0.087 0.045

Q = Qe + S (Qs − Qe ) (4.240)

Q = value of a quantity (electrical field, ion flow, etc.)
Qs = saturated value
Qe = electrostatic value
S = degree of saturation
Information obtained from field tests (Johnson et al. 1987), parameters for the
calculation considering the effect of weather, are shown in Table 4.39.
Moderate weather can be represented by “fall-fair” and extreme by “summer
high humidity and fog”.
Below the results (with intermediate calculation values) for the Base Case
example are shown (Tables 4.40 and 4.41).

4.20.1.3  Design Criteria

Reference (Cigré TB 473reports the analysis done by Cigré JWG B4/C3/B2.50
related to electric fields near HVDC lines and concluded:
“None of the scientific weight-of-evidence reviews conducted to date concluded
that any adverse health effects of exposure are likely but micro-shocks under some
conditions under a DC transmission line could be annoying or provoke startle”.
In the absence of any significant induction effects and the lack of evidence lin-
king exposure to dc electric fields and ion currents with any health hazards, percep-
tion thresholds for dc electric fields and ion currents are generally used as criteria
for the design of dc transmission lines. Figure 4.68 summarizes the results of an
investigation conducted to evaluate the perception of electric field.
From the figure it can be seeing that:

• for 25 kV/m and 100 nA/m2, 1/3 of the persons perceived the existence of the
field
• for 15 kV/m and 15 nA/m2, less than 10% perceived the field
4  Electrical Design 181

Table 4.40  Data and Worst

intermediate values value at X = 20  m
Voltage (kV) +/−500 +/−500
Conductor diam (cm) 3.822 3.822
Number of cond. 3 3
Bundle spacing (cm) 45 45
Pole Spacing P (m) 13 13
Height H (m) 12.5 12.5
X distance center (m) 10 20
P/H 1.04 1.04
H/V 0.025 0.025
H/Deq 39.81 39.81
Es maximum sat values 43.46
J+ 10−15 max sat values 109.21
J− 10−15 max sat values 142.30
X − P/2 3.5 13.5
X + P/2 16.5 26.5
(X − P/2)/H 0.28 1.08
Es sat at X 25.39
J+ sat at X 19.26
J− sat at X 30.55
G without space charge at X 10.5 5.2

Table 4.41 Results: E (kV/m); J (nA/m2)

Worst value at X = 20  m Worst value at X = 20  m
Spring Spring Summer high humidity/fog Summer high humidity/fog
S 50% pos 0.302 0.302 0.612 0.612
S 95% pos 0.652 0.652 0.825 0.825
S 50% neg 0.168 0.168 0.486 0.486
S 95% neg 0.550 0.550 0.721 0.721
E 50% pos 20.4 11.3 30.7 17.6
E 95% pos 32.0 18.4 37.7 21.9
E 50% neg 16.0 8.6 26.5 15.0
E 95% neg 28.6 16.3 34.3 19.8
J 50% pos 33.0 5.8 66.8 11.8
J 95% pos 71.2 12.6 90.1 15.9
J 50% neg 24.0 5.1 69.1 14.8
J 95% neg 78.2 16.8 102.6 22.0
182 J.F. Nolasco et al.

DC f ield
(kV/m) n=0.050 n=0.067 n=0.100
50
n=0.133

40

30 n=0.200

50%

20 33%
25%
16%
10 10%

0
0 30 60 90 120 150
Iron current density (nA/m2)

Figure 4.68  DC field intensities and ion current densities detected by various of the more sensi-
tive subjects.

• Therefore a pair E and J can be selected as design criteria associated with awe-
ather condition, for example:

• in any place inside the ROW: 25 kV/m; 100 nA/m2, spring 50% values (average);
or40 kV/m; 90 nA/m2 summer high humidity/fog
• at the ROW edge 10 to 15 kV/m; 10 to 15 nA/m2, summer high humidity/fog
95% values.

4.20.2 Magnetic Field

The magnetic field of transmission lines is calculated using two-dimensional analy-

sis assuming conductors parallel over a flat terrain.
The magnetic field Hji at a point(xi, yi) at a distance rij from the AC and DC con-
ductor or shield wires with a current Ii (real part only) is calculated as if for AC
lines.
Figure  4.69 shows the magnetic field for the Base Case example considering
1300 MW and ± 500 kV.
Note that the currents in the poles are of different polarity, therefore reducing the
effect. For monopolar transmission the magnetic fields are higher. The magnetic
field is of the order of the earth magnetic field ~50 μT.
4  Electrical Design 183

magnetic f ield Base Case

25

bipolar 1.3 kA
20 monopolar 1.3 kA

15
f ield (µT)

10

0
0 5 10 15 20 25 30 35 40 45
distance from center (m)

Figure 4.69  Magnetic Field.

4.21 Hybrid Corridor or Tower

Today important discussion has been carried out related to conversion of AC to DC

lines. In these cases there comes the possibility of having AC and DC line close
together or in the same tower.
In this case there will be interaction of electric and magnetic field and change in
the conductor surface gradient due to induction from one circuit to the other, there-
for influencing the Corona phenomena.
It is well accepted practice that Corona and field effects need to be taken into
account when designing new AC and DC power lines, when up-rating the voltage of
an existing AC line, or when converting existing AC lines to DC operation. The aim
of this item is therefore to give a sufficiently detailed description of the characteri-
stics and prediction of DC Corona and field effects and how to integrate these into
converted line designs. The various technical descriptions have also attempted to
emphasize those aspects which relate specifically to the conversion of AC lines to
DC operation (Cigré WG B2 41).

4.21.1 Conductor Surface Gradient

The calculation procedure shall include the concept that the AC conductor surface
voltage gradients are biased by the presence of the DC conductors, while the DC
conductor surface gradients include a ripple emanating from the presence of the AC
conductors, as illustrated in Figure 4.70. The calculation procedure proposed covers
only electrostatic effects; any influence of space charges on the electric fields on the
surface of the AC conductors is tentatively disregarded.
184 J.F. Nolasco et al.

AC CONDUCTOR BUNDLE
SURFACE GRADINET ON G=Peak value Gac=AC Component Gdc=DC Component

Gdc
Gac
Gp

Time-varying surface gradient on the ac conductor

b
AC CONDUCTOR BUNDLE
SURFACE GRADINET ON

Gac

Gdc

Gp=Peak Value
G

Gac=AC Component
Gdc=DC Component

Time-varying surface gradient on the dc conductor

Figure 4.70  Conductor surface voltage gradients in hybrid configurations.

As discussed later, the Corona performance of a hybrid configuration can then be

established by applying conventional calculation methods in the following way:

• Calculate the peak values of the combined AC and DC electric field distribution
on the surfaces of the AC and DC conductors.
• The Corona performance of the AC conductors may be calculated as for conven-
tional AC lines after dividing the peak values of the calculated conductor surface
gradients by 2 .
• The Corona performance of the DC conductors may be calculated using conven-
tional DC empirical methods by applying the peak values of the conductor sur-
face gradient.

4.21.2 Radio Interference

The calculation of RI from hybrid configurations presents some peculiarities.

• RI levels from DC lines are higher in fair-weather than in rain, and therefore prima-
rily considered as a fair-weather phenomenon. On the other hand, RI levels from AC
4  Electrical Design 185

lines are considered as both a fair weather and a foul weather phenomenon. For
hybrid configurations, it is therefore necessary to study both weather conditions.
• RI from AC conductors occurs around the peak of the positive half-cycle of the
power frequency voltage, whereas RI from DC conductors for an equivalent gra-
dient occurs all the time, primarily from the positive conductor.
• The nuisance value appears to be higher for AC than for DC.

The empirical formulas presented can be used to calculate the RI level of a hybrid
configuration at any particular distance from each AC phase and DC pole conductor
by applying the appropriate surface voltage gradients. With this information, the
total RI level is determined by adequately adding those levels for the different
weather conditions (Cigré WG B2 41).
Remind that:
For AC contribution: Foul weather RI = heavy rain-7 dB and fair = heavy rain -24 dB
For DC: fair negative weather RI = fair positive- 4 dB and foul = fair-3 dB.

4.21.3 Audible Noise

As with RI, the calculation of AN from hybrid configurations presents some

peculiarities:

• AN levels from DC lines are higher in fair-weather than in rain, and therefore
primarily considered as a fair-weather phenomenon. In contrast, AN levels from
AC lines are much higher in rain and therefore considered as a foul weather phe-
nomenon. For hybrid configurations, it is therefore necessary to study both
weather conditions.
• AN from AC conductors occurs around the peak of the positive half cycle of the
power frequency voltage, whereas RI from DC conductors for an equivalent gra-
dient occurs all the time, and practically only from the positive conductor.

The empirical formulas presented can be used to calculate the AN level of a hybrid
configuration at any particular distance from each AC phase and DC pole conductor
by applying the appropriate surface voltage gradients. With this information, the
total AN level is determined by adequately adding those levels for the different
weather conditions.
Remind that:
For AC contribution: Fair weather AN = foul- 25 dB
For DC: foul weather AN positive = fair positive- 6 dB and foul negative = zero.

4.21.4 Corona Losses

Formulas for calculating Corona losses on AC lines and isolated DC lines are presented
previously. However, due to the lack of full-scale test results there is no information
available how to apply these formulas to hybrid configurations (one may use the concept
of equivalent surface gradient as input in the presented equations for AC and DC lines.
186 J.F. Nolasco et al.

4.21.5 Electric and Magnetic Fields

The AC electric field at ground is calculated by the conventional method described,

with the DC conductors at zero potential. The result is an increase as compared with
DC line only (Cigré WG B2).
The DC electric field and ion current density at ground level may be estimated in
the same way as for isolated DC lines, with the AC conductors at zero potential.
The magnetic fields are calculated by the conventional method described. The
AC and DC magnetic fields can be treated separately since the effects on humans
are different (there are no induction effects from DC magnetic fields).

References to 4.1–4.12
Abramowitz, M., Stegun, I.A.: Handbook of Mathematical Functions. Dover Publication
Aileman, A.R.: Insulation coordination for power systems. Marcel Dekker, New York (1999). 767 p
Aluminum Association Handbook
Anderson, R.B., Eriksson, A.J.: Lightning parameters for engineering application. Electra 69,
65–102 (1980)
Anderson, R.B., Eriksson, A.J., Kroninger, H., Meal, D.V., Smith, M.A.: Lightning and thunderstorm
parameters. In: Lightning and Power Systems. London: IEE Conf. Publ. No. 236, 5 p (1984)
Baba, Y., Rakov, V.A.: On the use of lumped sources in lightning return stroke models. J. Geophys.
Res. 110, D03101. doi:10.1029/2004JD005202
Brown, G.W., Whitehead, E.R.: Field and analytical studies of transmission line shielding II. IEEE
PAS 88, 617–626 (1969)
Chartier, V.L.: Empirical expressions for calculating high voltage transmission line Corona phen-
omena. First Annual Seminar Technical Program for Professionals Engineers, BPA (1983)
Chartier, V.L., Stearns, R.D.: Formulas for predicting audible noise from overhead high voltage
AC and DC lines. IEEE Trans. PAS-100(1), 121–130 (1981)
Chartier, V.L., Sarkinen, S.H., Stearns, R.D., Burns, A.L.: Investigation of corona and field effects
of AC/DC hybrid transmission lines. IEEE PAS-100(1), 72–80 (1981)
Cigré TB 20: Interferences produced by corona effect of electric systems (1974)
Cigré TB 20: Interferences produced by corona effect of electric systems; description of pheno-
mena, practical guide for calculation (1974)
Cigré TB 21: Electric and magnetic fields produced by transmission systems (1980)
Cigré TB 48: Tower Top Geometry. WG 22-06 (1995)
Cigré TB 61: Addendum to interferences produced by corona effect of electric systems (1996)
Cigré TB 74: Electric power transmission and the environment. Field, noise and interference
Cigré TB 299: Guide for weather parameters for bare conductors to use for the deterministic rating
calculation
Cigré TB 388: Impacts of HVDC lines in the economics of HVDC projects (2009)
Cigré TB 440: Use of surge arresters for lightning protection of transmission lines
Cigré TB 473: Electric field and ion current environment of HVDC overhead transmission lines
Cigré TB 549: Working Group C4.407 – “Lightning Parameters for Engineering Applications” –
Cigré TB 549, Aug (2013)
Cigré TB No. 61: Addendum to Cigré Document No. 20 (1974). Chapter 7 (1996)
Cigré WG 33-04, TB 63: Guide to Procedures for Estimating the Lightning Performance of
Transmission Lines. Paris (1991)
Cigré WG B2-41: Guide to the conversion of existing AC lines to DC operation
Corbellini, U., Pelacchi, P.: Corona losses on HVDC bipolar lines. IEEE Trans. PWRD-11(3),
1475–1480 (1996)
4  Electrical Design 187

Dawalibi, F., Barbeito, N.: Measurements and computations of the performance of grounding sys-
tems buried in multilayer soils. IEEE Trans. Power Deliv. 6(4), 1483–1490 (1991)
Dommel, H.W.: Electromagnetic Transient Program – Reference Manual (EMTP Theory Book
prepared for BPA) (1986)
Douglass, D.: Alternating Current (AC) resistance of helically stranded conductors. Cigré TB 345
2008 SC B2 Overhead lines WG B2 12
EPRI EL 3892: HVDC Converter Stations for voltages above 600 kV. Project 2115-4 (1985)
EPRI Transmission line reference book 345 kV and above 2nd edition
EPRI: Transmission line reference book HVDC to 600 kV. EPRI report (1977)
Fernandes, J.H.M., et al.: Eletronorte and the challenge of long-distance transmission in Brazil.
Cigré (2008)
Fink, D.G., Beaty, H.W.: Standard Handbook for Electrical Engineering. McGraw-Hill, fourteen
Ed., (SHEE)
Happoldt, H., Oeding, D.: Elektrische Kraftwerke und Netze (Electrical Power Plants and
Systems). Springer (1978)
Heppe, R.J.: Computation of potential at surface above an energized grid or other electrode, allo-
wing for non uniform current distribution. IEEE PAS PAS-98 (1979)
HVDC Reference Book, TR-102764 (EPRI DC) (1993)
ICNIRP: Guidelines for limiting exposure to time varying electrics, magnetics, and electroman-
getic fields (1997)
IEC 61815: Guide for the selection of insulators in respect to polluted conditions. (1986)
IEC 62305: Protection against lightning – part 1: general principles
IEC 71-2: Insulation coordination part 2: application guide (1996)
IEEE Standard 1410: IEEE guide for improving the lightning performance of electric power over-
head distribution lines (2010)
IEEE Std 80: IEEE guide for safety in AC substation grounding (2000)
IEEE Std 142: IEEE recommended practice for grounding of industrial and commercial power
systems (1991)
IEEE Std 1243: IEEE guide for improving the lightning performance of transmission lines (1997)
Janischewsky, W., Grela, G.: Finite element solution for electric fields of coronating DC transmis-
sion lines. IEEE PAS-98(3) (1979)
Johnson, G.B.: Degree of corona saturation for HVDC transmission lines. IEEE PWRD-2(2)
(1987)
Kiessling, F., et al.: Overhead Power Lines. Springer (2003)
Knudsen, N., Iliceto, F.: Contribution to the electrical design of HVDC overhead lines. IEEE
Trans. PAS-93(1), 233–239 (1974)
MacGorman, D.R., Maier, M.W., Rust, W.D.: Lightning strike density for the contiguous United
States from thunderstorm duration records, NUREG/CR-3759, Office of Nuclear Regulatory
Research, U.S. Nuclear Regulatory Commission, Washington, DC, 44 p (1984)
Maruvada, P.S.: Corona Performance of High-Voltage Transmission Lines. Research Studies
Press, Baldock (2000)
Mousa, A.M.: The soil ionization gradient associated with discharge of high current into concen-
trated electrodes. IEEE PWRD 9(3), 1669–1677 (1994)
Nolasco J.F., et al.: Assessment and Improvement of availability of overhead lines and compo-
nents. Cigré B2 Session -n°-22-107 (2002)
Nucci, C.A.: Survey on Cigré and IEEE procedure for the estimation of the lightning performance
of overhead transmission and distribution lines. Asia–Pacific International Symposium on
EMC (2010)
Popolansky, F.: Frequency distribution of amplitudes of lightning currents. Electra 22, 139–147
(1972)
Sekioka, S., Sonoda, T., Ametani, A.: Experimental study of current-dependent grounding
resistance of rod electrode. IEEE Trans. Power Del. 20, 1569–1576 (2005)
Stephen R.: The thermal behaviour of overhead conductors. Sections 1 and 2. Cigré SC:22
Overhead lines. Electra. No: 144 pp. 107–125 (1992)
188 J.F. Nolasco et al.

Stephen, R.: Probabilistic determination of conductor current rating – Cigré Electra 164, 103–119
(1996)
Stevenson, W.D.: Elements of Power System Analysis (1962)
Szechtman, M., Wess, T., Thio, C.V.: First benchmark model for HVDC control studies. Electra,
No. 135 (1991)
TB 207: Thermal behavior of overhead conductors (2002)
Tompson, E.M.: The dependance of lightning return stroke characteristics on latitude. J. Geograph.
Res. 85 (1980)
U.S. EPA., 550/9-74-004: Information on levels of environmental noise requisite to protect public
health and welfare with an adequate margin of safety (1974)
Visacro, S., Alípio, R.: Frequency dependence of soil parameters: experimental results, predicting
formula and influence on the lightning response of grounding electrodes. IEEE PWRD 27(2),
927–935 (2012)
Visacro, R.: A comprehensive approach to the grounding response to lightning currents. IEEE
PWRD 2(2) (2007)
Visacro, S., Alípio, R., Murta, M.H., Pereira, C.: The response of grounding electrodes to lightning
currents: the effect of frequency-dependent soil resistivity and permittivity. IEEE EMC 52(2),
401–406 (2011)
Wagner, C.F.: The relation between stroke current and the velocity of the return stroke. IEEE
Trans. Power Syst. 82, 609–617 (1963)
Young, F.S., Clayton, J.M., Hileman, A.R.: Shielding of transmission lines. IEEE Trans. PAS 82,
132–154 (1963)

References to 4.13–4.21
Abramowitz, M., Stegun, I.A.: Handbook of Mathematical Functions. Dover Publication
Chartier, V.L., Stearns, R.D.: Formulas for Predicting Audible Noise from Overhead High Voltage
AC and DC Lines. IEEE Trans. PAS-100(1), 121–130 (1981)
Chartier, V.L.: Empirical Expressions for Calculating High Voltage Transmission Line Corona
Phenomena. First Annual Seminar Technical Program for Professional Engineers, Bonneville
Power Administration (BPA) (1983)
Cigré TB 473: Electric Field and Ion Current Environment of HVDC Overhead Transmission Lines
Cigré TB 20: Interferences Produced by Corona Effect of Electric Systems; Description of
Phenomena, Practical Guide for Calculation (1974)
Cigré TB 48: Tower Top Geometry. WG 22-06 (1995)
Cigré TB No. 61: Addendum to Cigré Document No. 20 (1974). Chapter 7 (1996)
Cigré TB 388: Impacts of HVDC lines in the economics of HVDC projects. (2009)
Cigré WG B2-41: Guide to the conversion of existing AC lines to DC operation
Corbellini, U., Pelacchi, P.: Corona Losses on HVDC Bipolar Lines. IEEE Trans. PWRD-11(3),
1475–1480 (1996)
EPRI EL 3892: HVDC Converter Stations for voltages above 600 kV. Project 2115-4 (1985)
EPRI: Transmission Line Reference Book HVDC to 600 kV. EPRI Report (1977)
Fink, D.G., Beaty, H.W.: Standard Handbook for Electrical Engineering. McGraw-Hill, fourteen
Ed., (SHEE)
HVDC Reference Book, TR-102764 (EPRI DC) (1993)
IEC 61815: Guide for the selection of insulators in respect to polluted conditions. (1986)
Janischewsky, W., Grela, G.: Finite Element Solution for Electric Fields of Coronating DC
Transmission Lines. IEEE PAS-98(3) (1979)
4  Electrical Design 189

Johnson, G.B.: Degree of Corona Saturation for HVDC Transmission Lines. IEEE PWRD-2(2) (1987)
Knudsen, N., Iliceto, F.: Contribution to the Electrical Design of HVDC Overhead Lines. IEEE
Trans. PAS-93(1), 233–239 (1974)
Maruvada, P.S.: Corona Performance of High-Voltage Transmission Lines. Research Studies Press
Ltd., Baldock (2000)
Szechtman, M., Wess, T., Thio, C.V.: First Benchmark Model for HVDC Control Studies. Electra
(135) (1991)
TB 207: Thermal Behavior of Overhead Conductors (2002)
U.S. EPA., 550/9-74-004: Information on Levels of Environmental Noise Requisite to Protect
Public Health and Welfare with an Adequate Margin of Safety (1974)

Joao F. Nolasco  graduated in Electric & Electronic Eng­ineering

and M.S. in Transmission Systems, as well as in Statistics in
State University of Minas Gerais (UFMG), Brazil. Engineer in
Siemens Brazil (Jan to June 1967) and then in Siemens Germany
(1967-1969). Professor in the UFMG University (1968-1970),
then started his career at the Electric Utility Cemig, where has
been in sequence Planning Enginer, Design Engineer, Head of
Electric Studies of T. Lines and finally Head of Transmission
Line Department. From 1994 until today, he has acted as
Consultant in both AC and DC Transmission Systems and
Planning, providing services to several Utilities and private
Transmission Companies in Brazil, Argentine, Chile, Peru,
Bolivia, Suriname, Mexico, USA, South Africa etc. In Cigré he
has been Member of SC B2 (Overhead Lines) - 1984 to 1998 -
and then Convener of some WG’s and TF’s covering AC & DC lines (1998-2014). Author of more
than 100 papers, several technical reports and two Technical Brochures. One of the authors of the
Book “Overhead Power Lines” (Springer Verlag, Germany - 1993). He wrote and provided courses
on Transmission and Distribution Lines at the University of Cape Town in 1989 and in other coun-
tries afterwards.

José Antonio Jardini  has received his B.Sc. degree from the

Polytechnic School at the University of Sao Paulo (USP) in
1963. Subsequently, he obtained his M.Sc. and Ph.D. degrees in
1970 and 1973, all from the same institution. From 1964 to 1991
he worked at Themag Eng. Ltd (Engineering Company) in the
area of Power Systems, Automation and Transmission Lines
projects. Currently, he is Full Professor in the Department of
Engineering of Energy and Electric Automation at University of
São Paulo. He is a member of CIGRE and was the Brazilian
representative in the SC38 of CIGRE and member of WG B4
(HVDC & FACTS power electronics). At IEEE he is Fellow
Member and Distinguished Lecturer of PES/IEEE and was also
for IAS/IEEE. He is one of the authors of several Cigré
Brochures on HVDC Lines Economics, Convener of JWG Electric of HVDC Transmission Lines,
and Converter modelling. He had many volunteer activities at IEEE as: South Brazil section
Chairman; region treasurer, IEEE Fellow Committee, nomination boards, Education Society Fellow
Committee. Awards: IEEE HVDC Uno Lamm (2014); IEEE MGA Larry K Wilson(2010); IEEE
innovation (2004); CIGRE B4 Technical (2011).
190 J.F. Nolasco et al.

Elilson E. Ribeiro  graduated in Electrical Engineering (1984)

and M. S. in Electric Power Systems (1987) in State University
of Minas Gerais (UFMG). Engineer of MF Consultoria (1987-
1998). Professor in the Pontifical Catholic University of Minas
Gerais (PUC Minas; from 1999 until today) and Director of
NSA Consultoria. He has acted as Consultant, primarily, in the
areas of electromagnetic and thermal modelling in equipment
development (air core power reactors), grounding system design
(for transmission lines, power substation and industrial installa-
tions), analysis of electromagnetic transients in electric power
­systems, analysis of electromagnetic interference from trans-
mission lines and transmission line design.
Structural and Mechanical Design
5
Elias Ghannoum

Contents
5.1 Introduction......................................................................................................................... 192
5.2 Deterministic and Reliability Based Design (RBD) Methods............................................ 193
5.2.1 Historical Background.......................................................................................... 193
5.2.2 The Need for Reliability Based Design in Overhead Line Design....................... 194
5.2.3 How RBD Methods Address the Deficiencies of Deterministic
Design Procedures................................................................................................ 196
5.2.4 How to Apply IEC 60826..................................................................................... 197
5.2.5 Conclusions........................................................................................................... 204
5.2.6 References............................................................................................................. 204
5.3 Comparison of RBD Methods............................................................................................ 205
5.3.1 Introduction........................................................................................................... 205
5.3.2 Documents Compared and References................................................................. 205
5.3.3 Basis of Design..................................................................................................... 206
5.3.4 Basic Design Equation.......................................................................................... 207
5.3.5 Combination of Loads.......................................................................................... 207
5.3.6 Load Factors for Permanent and Variable Loads.................................................. 207
5.3.7 Wind Loads........................................................................................................... 210
5.3.8 Drag Coefficient.................................................................................................... 214
5.3.9 Span Factors.......................................................................................................... 214
5.3.10 Ice Loads............................................................................................................... 215
5.3.11 Combined Wind and Ice Loads............................................................................. 215
5.3.12 Failure and Containment Loads (Security Loads)................................................ 216
5.3.13 Construction and Maintenance Loads (Safety Loads).......................................... 216
5.3.14 Other Loads........................................................................................................... 216
5.3.15 General Comparative Overview............................................................................ 217
5.3.16 Conclusions........................................................................................................... 218

Originally published by Cigré, 2014, under the ISBN 978-2-85873-284-5. Republished by Springer
International Publishing Switzerland with kind permission.

E. Ghannoum (*)
Montreal, Canada
e-mail: [email protected]

© Springer International Publishing Switzerland 2017 191

K.O. Papailiou (ed.), Overhead Lines, CIGRE Green Books,
DOI 10.1007/978-3-319-31747-2_5
192 E. Ghannoum

5.4 Tower Top Geometry and Mid-span Clearances................................................................. 225

5.4.1 Introduction........................................................................................................... 225
5.4.2 Part 1: Existing National Practices....................................................................... 226
5.4.3 Part 2: Swing Angles............................................................................................ 228
5.4.4 Part 3: Required Clearances.................................................................................. 229
5.4.5 Part 4: Coordination of Conductor Positions and Electrical Stresses................... 230
5.4.6 Part 5: Application Example................................................................................. 231
5.4.7 Conclusion............................................................................................................ 234
5.4.8 References............................................................................................................. 234
5.5 Load Control Devices......................................................................................................... 234
5.5.1 Summary............................................................................................................... 234
5.5.2 Introduction........................................................................................................... 236
5.5.3 Classification of LCD........................................................................................... 236
5.5.4 Technical Data Related to Available LCD............................................................ 237
5.5.5 Specification for an Ideal LCD............................................................................. 248
5.6 Mechanical Security of Overhead Lines Containing Cascading
Failures and Mitigating their Effects.................................................................................. 254
5.6.1 General.................................................................................................................. 254
5.6.2 Exceptional Loads, Accidental Loads and Mechanical
Security of Overhead Lines.................................................................................. 255
5.6.3 Line Cascade or Multiple Support Failures?........................................................ 255
5.6.4 Learning from Recent Major Tower Cascading Failures...................................... 258
5.6.5 Current Understanding of Dynamic Line Cascading............................................ 258
5.6.6 Recent Developments in OHL Cascading Mitigation........................................... 259
5.6.7 Security Design Criteria to Prevent OHL Cascades............................................. 261
5.6.8 Framework for Successful Design to Limit Overhead Line Cascades................. 263
5.6.9 Conclusions and Recommendations for Future Action........................................ 263
5.6.10 References............................................................................................................. 265
5.7 Influence of Design Parameters on Line Security.............................................................. 265
5.7.1 Introduction........................................................................................................... 265
5.7.2 The Need for Unbalanced Longitudinal Loads..................................................... 266
5.7.3 Requirements of Standards, Design Codes, and Utility Practices........................ 266
5.7.4 Comparison of BCL and UIL with Weather Load Cases..................................... 267
5.7.5 Recommendations................................................................................................. 274
5.7.6 Conclusions........................................................................................................... 275
References................................................................................................................................. 276

5.1 Introduction

This chapter describes the studies made by Cigré in the area of structural design of
overhead lines.
The first part is devoted to the Reliability Based Design (RBD) method, particu-
larly IEC 60826 where Cigré SC22-B2 Committees played an important role in
supporting and contributing to this international standard nowadays widely used
worldwide. In this part, the principles of RBD are explained as well as the require-
ments of IEC 60826. This is followed in 2006 by a comparison of RBD standards to
IEC 60826.
In the second part “tower top geometry and mid-span clearances” the methods to
establish the distances between conductors and the supporting structure either in
mid-span or at the tower are given as in the reference Cigré studies and Brochures.
5  Structural and Mechanical Design 193

The third part covers load control devices that were developed to reduce dynamic
and/or static load on structures in case of a failure in order to reduce possibility of
cascading events.
The fourth part covers the Mechanical Security of Overhead Lines Containing
Cascading Failures and Mitigating Their Effects.
Finally, the fifth part covers the influence of line design parameters on line
security.

5.2  eterministic and Reliability Based Design (RBD)

D
Methods

5.2.1 Historical Background

During recent decades, the IEC (International Electrotechnical Commission) Com­

mittee TC11 and Cigré Study Committee SC 22 (now SC B2) pioneered impro-
vement of deterministic overhead transmission lines design criteria as well as the
introduction of reliability/probability based design concepts, in, for example Cigré
1990 [2], Cigré 2001 [3], Ghannoum-Orawski, 1986 [4].
The technical work on reliability based transmission line design started in Cigré
more than 40 years ago under the sauspices of Study Committee 22. Many experts,
to name a few, such as Commellini, Manuzio, Cojan, Paris, White, Schjetne, Wood,
Orawski, Kieβling, Henrioul, Rogier, Ghannoum, etc., contributed to this work eit-
her in the SC22 or in Cigré papers and publications.
Further technical development was pioneered by a joint effort between Cigré
and IEC through both technical working groups SC22-06 from Cigré, initially
chaired by Georges Orawski from the UK, and IEC/TC11/WG08 chaired by Elias
Ghannoum.
As a result of the above technical work, a milestone occurred in 1991 when IEC
8261 entitled “Loading and Strength of Overhead Lines” was published by IEC in
1991 as a technical report type II, i.e., a pre-standard document to be reviewed in a
few years for the purpose of converting it to an IEC standard. This publication intro-
duced reliability and probabilistic concepts for calculation of loading and strength
requirements for overhead line components.
Subsequent work by Cigré SC22-WG06 helped in modifying the IEC report and,
in October 2003, Standard IEC 60826 [1] was published by the IEC after a unani-
mous positive vote by National Committees.
This chapter aims to increase understanding of IEC 60826 as well as its applica-
tion to overhead lines. It demonstrates why Reliability Based Design (RBD) methods
constitute a major improvement compared to deterministic methods and summarizes
the main features of IEC 60826.

1
 The new number of this publication in the IEC catalogue is 60826 and the older version 826 is not
anymore available.
194 E. Ghannoum

5.2.2 T
 he Need for Reliability Based Design in Overhead Line
Design

Many current standards and design practices in the world are deterministic in nature
and some of them are even imported from other countries, despite major differences
in climatic and terrain conditions.
However, it is fair to recognize that deterministic methods have evolved, and
many utilities or countries tried to address deficiencies they have identified in their
deterministic design practices by specifying additional loading cases and strength
requirements, often much more critical than their basic standards.
Despite improvements, experience and technical studies have shown that determi-
nistic methods have inherent deficiencies that cannot be addressed by minor changes
in the design requirements, and can only be resolved by migrating to RBD. Some of
the potential deficiencies of deterministic methods are given hereafter and practical
examples are provided.

5.2.2.1  P  ossible Inconsistencies and Unbalance in Strengths

of Components
The objective of deterministic methods is to comply, as a minimum, with the safety
(or overload) factors required for each loading case specified. There are usually no
requirements for a preferred sequence of failure in these standards and any sequence
can be expected as a result of deterministic design.
In a number of reviews of standards and line failure analyses performed by the
Author, it was found that the withstand of some angle and dead-end towers were in
fact less than that of tangent towers. This is particularly true in lines located in icing
areas where tangent towers are designed with wind and weight spans significantly
exceeding average spans, while angle towers may be used at their maximum angles.2
Cases where tangent tower can withstand as much as 40–50 mm of radial ice,
while angle towers can fail due to 35 mm of ice are not uncommon in locations with
important ice loads.

5.2.2.2  Unknown Reliability Level

Except by a general inference from very long experience, it is very difficult to attri-
bute a reliability value to any design based on deterministic principles. In fact these
principles do not even recognize that load can exceed strength (both values are
deemed to be constant) and often imply that whoever follows such principles should
obtain a safe and reliable line.
The past experience is not always a good descriptor of reliability, especially if the
experience with these design criteria is not long enough and the transmission line
system covers only a small fraction of the total service area. For example, if the real
probability of failure of existing lines designed according to deterministic loads is
low, say 3 % every year, there is about 50 % chance that no failure will occur during
a 20–25 year life span. Thus, the fact that no failures occurred during this period is
not always a good measure of the actual line reliability.

2
 This is usually the case in many lines, because line routes and line deviation angles often aim to
use expensive angle towers at their maximum angles.
5  Structural and Mechanical Design 195

5.2.2.3  Difficulty to Adjust the Overall Line Reliability

If a designer wants to increase the reliability of a given line because of its import-
ance in the network, he has few means to assess the quantum of increase in the
safety (or overload) factors he currently uses to improve the reliability. He simply
does not have sufficient information to decide if an increase, say of 10 % or 30 % in
the safety factors, would be significant enough or not.

5.2.2.4  D  ifficulty to Design Heterogeneous Structures such

as Combination of Steel and Wood
When designing a structure using different materials, such wood poles and steel
crossarms, if the same overload factors are applied to loads, they will lead to une-
qual reliability between these structural parts of the same structure.
For example, the compressive stresses of wood pole stresses in North America
were given as average values. In recognition of this fact and the large strength dis-
persion of wood strength, a safety factor of 4 was specified for compression loads.
If a wood crossarm is substituted by a steel section, the deterministic design methods
do not have the appropriate tools to provide for a substitute and equivalent safety
factor for steel crossarms.

5.2.2.5  Difficulty to Evolve with new Technologies

The above example of a structure involving different material can be extended to a
more general issue of the equivalence between safety factors/overload factors to be
applied to different structural components. For example, steel and concrete pole
manufacturers recently challenged the large safety factors imposed on them by stan-
dards in North America arguing that steel pole properties are quite predictable con-
trary to wood poles.
The same could be extended to new types of materials such as poles made of
fiber reinforced concrete or fiberglass, etc. Deterministic methods cannot provide
designers with reliable guidelines to establish equivalent requirements between new
material technologies and established ones.

5.2.2.6  Difficulty to Adjust Design to Local Conditions

Deterministic methods cannot efficiently cope with the variations in weather loa-
ding that occur in a service area. For example, not so long ago, North America was
divided in only a handful of loading zones, each one as large as many European
countries combined. Another major problem arises because the specified climatic
loadings are not related to any specified return period. For example, the combined
ice and wind load of 12.7 mm of radial ice and 385 Pa of wind pressure3 specified
in CSA C22.3 for Canada was estimated, using RBD, to correspond to return peri-
ods of 3 to 500 years, depending on the actual location.

3
 This pressure corresponds to a 10 min wind speed of about 65 to 70 km/h for conductors at 10 m
of height in a terrain type B.
196 E. Ghannoum

5.2.3 H
 ow RBD Methods Address the Deficiencies
of Deterministic Design Procedures

The following basic principles are used by RBD methods in order to address the
deficiencies stated above in deterministic methods:

5.2.3.1  Limit Loads are Specified

Contrary to deterministic methods that incorporate safety and/or overload factors,
RBD specifies the limit loads4 that the line has to withstand without damage. When
limit loads are used, these are transferred to the conductors and then to hardware, to
towers (suspension or dead-end) and to foundations. Therefore, each line compo-
nent will be designed for the effects of the same limit loads considered for the line,
without the risk of mismatch between components.

5.2.3.2  L  oads and Strength are Recognized as Random Variables

and Treated as such
There is no disagreement between proponents of RBD or deterministic methods that
loads (e.g. ice, wind, temperature and their combinations) and strengths are random
variables. When the randomness of loads and strengths are taken into account, there
is an immediate recognition that absolute reliability cannot be achieved and that
there is always a risk that design loads can be exceeded or component’s strength can
be less that design values.
Therefore, RBD design standards will have to specify an acceptable maximum
probability of failure (or its complement to one, the minimum reliability) and pro-
vide means to modulate these probabilities if warranted by economics or the import-
ance of the project.

5.2.3.3  D  esign Loads are Selected Based on the Required

Return Period
It is widely accepted that yearly maximum climatic loads such as ice and/or wind,
follow extreme distribution functions. With such statistical functions, designers can
associate a value of loads for any selected return period.
The return period is an important parameter for qualifying reliability. For
example, if a 150 year return period load is selected as the limit load to design for,
the probability of exceeding this load is 1/150 ≈ 0.67 % per year. In a 50 year life
span, the probability of exceeding the same load is equal to [1-(1-1/150)50] ≈ 28 %.

5.2.3.4  T  he Characteristic Strength to be Associated with Load QT is

also Dependent on Strength Dispersion and is Equal
to the 10 % Exclusion Limit
Similarly, the strength is also a random variable, typically described by normal or
log-normal distribution functions. The design strength is usually taken as the aver-
age minus 1 to 3 standard deviations. If the strength were chosen with an exclusion
limit of 10 %, this value would correspond to the average strength minus 1.28×
standard deviation in the case of a Normal distribution function.

 These loads are sometimes called ultimate loads, but the wording “limit” is a preferred one.
4
5  Structural and Mechanical Design 197

5.2.3.5  T  he Association of QT and (10 %) R will Yield an almost

Constant Yearly Reliability of the order of 1/(2 T)
The line reliability and unreliability threshold corresponds to the condition where
the load effect is equal to the strength. If load does not exceed strength, then the
system (transmission line) is reliable. In the opposite case, the line would be in a
failed condition.
If a load QT is associated with a strength corresponding to 10 % exclusion limit,
the resulting reliability is almost constant and equal to 1/(2 T) in the normal range
of variation of Q and R. This very significant relationship was uncovered and pro-
ven in previous papers and Cigré reports written by Ghannoum.

5.2.3.6  Y early Reliability can be Customized by Varying the Return

Period T of Design Loads
Increasing reliability with RBD can be done very easily, either by selecting loads
with a higher return period, or strength with a lower exclusion limit. The former
approach is more practical and accurate and was selected by Cigré and IEC.

5.2.4 How to Apply IEC 60826

5.2.4.1  Design Steps of IEC 60826

The design methodology as per IEC60826 can be summarized in Figure 5.1. It is
noted that activities (a) and (h) listed in this figure are not within the scope of IEC
60826.

a- Preliminary design : route selection, conductors, insulation design, supports, foundations, climate data, etc.

b1- Select reliability level b2- Select security requirements b3- List safety requirements (compulsory)

c- Calculate climatic variables

d1- Calculate climatic limit d2- Calculate loads related to d3- Calculate construction & maintenance
loads security loads

e- Determine strength
coordination

f- Select load and strength

factors

g- Calculate required
characteristic strength of
components

Check compliance with safety requirements of

national and regional regulations

h- Detailed design of line

components

Figure 5.1  Transmission line design methodology according to IEC 60826.

198 E. Ghannoum

5.2.4.2  Source of Design Requirements

The design according to IEC 60826 (see boxes b1, b2 and b3 in Figure 5.1) origina-
tes from the following requirements:

• Reliability: These requirements consist of climatic loads (wind, ice, temperature

and their combinations) and aim to provide lines with satisfactory service perfor-
mance. Statistical tools are used to quantify these loads.
• Security: These requirements aim to prevent or reduce risk of uncontrollable or
cascading failures.
• Safety: These requirements aim to prevent human injury.

5.2.4.3  Reliability Levels

Three Reliability levels (I, II, III) are proposed in IEC 60826. These levels corres-
pond to return periods of design loads of 50, 150 and 500 years. In general,

• Level I is considered minimum for all permanent lines

• Level II applies to lines with voltages equal or exceeding 230 kV
• Level III applies to important lines in excess of 230 kV that are a unique source
of supply.

Other levels can be selected based local conditions or on an economical optimiza-

tion between cost of increased reliability and present worth of future failures.

5.2.4.4  Security Requirements

Security requirements relate to behavior of lines once failure is initiated. They aim
to prevent uncontrolled propagation of failures (cascading). In such case, compo-
nents are allowed to reach stresses very close to their ultimate limit state (failure). It
is noted that in IEC 60826, security is a deterministic concept, while reliability is
probabilistic.
Security and reliability requirements are interrelated because both tend to increase
the required strength of components. Security measures, if more critical than cli-
matic loads (reliability requirements), can also increase reliability.

5.2.4.5  Safety Requirements

These are required to protect people from injury. They consist of construction and
maintenance loads. It is aimed that the probability of failure under such loads should
be very low.

5.2.4.6  Design Equation, General Format

Load effect < Strength

or:
5  Structural and Mechanical Design 199

QT < RC

or: Load corresponding to a return period T < Characteristic strength R C
The above equation has been expanded in the IEC standard to the form below:

γ QT = φR RC

where:

γ factor for span dispersion, default value equal to 1.0 for new lines
QT load corresponding to a return period T
ϕR global strength factor equal to the product of ϕS ϕN ϕQ ϕC
ϕS factor related to coordination of strength (sequence of failure)
ϕN factor related to number N of components
ϕQ factor related to the difference between tested and installed component
ϕC factor related to the statistical parameters of the characteristic strength

It is important to note that the load QT shall be the maximum along the space covered
by the line. Furthermore, not only the maximum load intensity is important, but also
its spatial coverage, as both affect design requirements and line reliability. Dir­
ectional tendencies of wind or ice loads can be taken into account if confirmed;
otherwise, it should be assumed that load direction always occurs in the most criti-
cal direction.

5.2.4.7  Loading Cases and Limit States

Limit states of strength of line components are defined for each component: a damage
limit state (serviceability) and a failure (ultimate) limit state. Each group of loading
requirements is associated with one of the limits states given below (Table 5.1).

5.2.4.8  Differences between Theoretical and Actual Reliabilities

IEC 60826 recognizes that the actual reliability may differ from the theoretical
reliability when factors such the ones listed below are not properly accounted
for:

• Actual use factors of components, particularly towers, are quite different from the
assumed value of 1.0;
• The degree of correlation between loads and strengths
• Direction of wind speed in relation to that of the line
• Exclusion limit of strength different from the assumed 10 %
• Number of components subjected to maximum load intensity
• Quality control during fabrication and construction.

Methods to take into account the above factors are covered in the subject standard.
200 E. Ghannoum

Table 5.1  Loading conditions and strength limits

Condition Loads Strength Limit State
Reliability climatic, ice, wind, wind + ice, with a damage limit
return period T
Security failure limit (torsional and longitudinal) failure limit
Safety construction and maintenance loads damage limit

5.2.4.9  Use Factor of Components (U)

The use factor in IEC 60826 is defined as the ratio of the actual load (as built) to the
limit design load of a component. For tangent towers, it is virtually equal to the ratio
of actual to maximum design spans (wind or weight), and for angle towers, it also
includes the ratio of the sines of the half angles of deviation (actual to design angles).
Use factor cannot exceed 1.0 and its influence on line reliability has been covered
in the IEC standard. The use factor variation in overhead lines is inevitable because
of the following reasons:

• Line components are mass fabricated

• Components are not specifically designed for each tower location or use
• Their design parameters reflect maximum usage along the line
• Effective loads on line components are location dependent (span and tower height
at each location)

Globally, the use factor variation increases reliability. However, a large dispersion
of use factor U may be an indication of a poor optimization (e.g. not enough tangent
tower types or their parameters incorrectly selected). It is important to recognize
that the preferred sequence of failure could also be altered if the use factor variation
is not taken into account.

5.2.4.10  The Characteristic Strength RC

IEC 60826 makes reference to the characteristic strength which is defined as the
strength value guaranteed in relevant Standards. Sometimes, it is also called the gua-
ranteed strength, the minimum strength, or the minimum failing load, and usually
corresponds to an exclusion limit, from 2 to 5 %, with 10 % being an upper practical
(and conservative) limit.
The strength distribution function is usually Normal (Gaussian). With stringent
quality control, it tends to become a Log-normal function.
The characteristic strength can thus be calculated from the following equation,
assuming it corresponds to a 10 % exclusion limit:

R C = (10%) R = R (1 − k VR ) , where

k = 1.28 for Normal distribution

k = 1.08 to 1.26 for Log-normal distribution.

In case the maximum intensity of load is widespread and covers a large number (N) of
structures, the strength distribution becomes that of a chain or a series of N components
5  Structural and Mechanical Design 201

whose strength is controlled by the weakest. Although the original distribution of

strength can be Normal, that of the series of N structures will tend to be an Extreme
(minima) type. Correction factors are provided in order to take into account the effect
of the spatial coverage of the maximum load event on reliability (ϕN factor).

5.2.4.11  Strength Coefficient ϕS Related to Sequence of Failure

In IEC 60826, line components can be designed to fail (with a 90 % probability) in a
preferred mode called “preferred sequence of failure”. The best (or least damaging)
failure mode is the one where the consequences of the first failure on the line are mini-
mized. Strength factors allowing to target a preferred sequence of failure are provided
in the standard. It is generally accepted that angle towers, dead-end towers, conductors
or foundations should not fail first, thus leaving tangent towers as the one to fail first.
Table 5.2 specifies the strength factors applicable to the strength of the component not
to fail first.

5.2.4.12  Wind Loads and Limitations of Wind Calculations

Wind loads on conductors and tower structures are the source of important and cri-
tical loading requirement for overhead transmission lines. Methods to calculate
wind forces, starting with a reference wind speed, are provided in the standard for
the following conditions:

• Spans between 200 m and 800 m

• Height of supports less than 60 m
• Altitude below 1300 m.

5.2.4.13  Ground Roughness

For the purpose of calculating wind pressure and forces, four categories of ground
roughness (also called terrain types) are provided:

• A- Flat coastal areas and deserts

• B- Open country, cultivated fields
• C- Numerous low height obstacles
• D- Suburban areas.

5.2.4.14  Reference Wind Velocity

The reference wind velocity VR considered in IEC 60826 consists of a 10 min. aver-
age, at 10 m height, in a terrain type B. The standard provides for conversion from
other wind data, having different averages or located in a different terrain category,
to the above reference value.

Table 5.2  Values of strength coefficient ϕS

Coefficient of Variation (COV) of R1
5 % 7.5 % 10 % 20 %
COV of 5 - 10 % 0.92 0.87 0.82 0.63
R2 10 - 40 % 0.94 0.89 0.86 0.66
Note: in the above Table, R2 is the component designed more reliable than R1
202 E. Ghannoum

5.2.4.15  Wind Speed Design Cases

High wind is combined with average minimum daily temperatures, and a reduced wind
(60 % of the reference value) is combined with the 50-year minimum tem­perature.

5.2.4.16  Wind Load Model

The conversion from wind speed V to wind forces follows the equation:
Wind force = k (½ τμ V2), where k is the product of:

• height factor
• span factor
• response factor
• shape factor.

In IEC 60826, the k factor in the above equation takes the form of:

Wind force = A Cx Gc GL (½ τμ V2), with

Gc = the combined wind factor dependent on spans, height and terrain roughness
category
GL = Span factor
Cx = Drag (or force) coefficient

μ is the air mass per unit volume = 1.225 kg/m3 (this is a default value, but adjust-
ments of μ for different temperatures and altitudes are provided), τ is the air density
correction factor given in Table 5 of IEC 60826.
A similar equation provides for calculation of wind forces applied to various
types of transmission structures such as those made of angle sections, round pipe
sections or steel poles. Drag coefficients are also provided for these tower types and
take into account the compactness (or solidity ratio) of the windward face to reflect
the shielding of wind on the leeward face.

5.2.4.17  Icing Types

Ice accretion on conductors and structures are the source of important loads, and
often control the design in many northern countries. The standard covers three types
of ice accretion: precipitation icing, wet snow, and in-cloud icing. Methods to cal-
culate design icing are provided and cover a range of cases with various availabili-
ties of statistical data.

5.2.4.18  Ice Loading Cases

Once design ice thickness or weight of ice per unit length of conductors has been
statistically defined, this value is used in the following loading cases:

• Uniform ice formation

• Non uniform ice (longitudinal unbalanced icing, with all phases in a span subjec-
ted to the same unbalanced conditions)
• Torsional condition (unbalanced icing conditions occurring in opposite longitu-
dinal directions thus creating a torsional moment on the structure).
5  Structural and Mechanical Design 203

5.2.4.19  Combined Ice Loads with Wind

The presence of wind during or after icing episodes requires special loading cases and
combinations of ice and wind loads in order to account for their combined effects.
The calculation of combined forces due wind on ice covered conductors are pro-
vided in the standard and take into account: the ice thickness or ice weight per unit
length of conductor, ice density, wind speed during icing, and drag coefficient of ice
covered conductor.
Two combinations of ice, wind speed during icing, ice density/drag coefficient
are provided for in the standard and consist of combining an extreme value of one
variable (such as the 50 year value) with the average values of the other variables.

5.2.4.20  C  onstruction and Maintenance Loads (Safety

Requirements)
The loading conditions provided in the IEC standard supplement national regula-
tions and safety codes. They are focused on reducing the risk of injuries to person-
nel working during construction and maintenance of the lines. These requirements
should result in a very high reliability (risk of failure practically nil). The approach
to deal with such loads is deterministic and consists of applying overload factors of
1.5 to 2.0 in order to insure such a high reliability. These loads are not usually com-
bined with severe climatic loads, because construction and maintenance operations
are not commonly undertaken during such weather events.
For example, loads during erection of supports are simulated by designing each
support point for twice the static loads at sagging conditions. Under some conditions,
and under controlled construction operations, the factor of 2.0 could be reduced to 1.5.

5.2.4.21  Security Related Loads

As explained earlier, these loads are intended to prevent cascading or uncontrollable
failures. Minimum requirements are specified as follows:

• A broken phase load (torsional load) is applied on any one phase or overhead
ground wire attachment point, and is equivalent to the Residual Static Load
(RSL) calculated with bare conductors at average temperatures.
• A longitudinal load is specified, equivalent to a simulated fictitious ice load equal
to the conductor weight applied on one side of the tower.

For lines that require a higher security level, additional security measures can be
considered such as: Increasing the number of points where the RSL is applied,
Considering the RSL in conjunction with some climatic load, and/or inserting
anti-cascading towers.

5.2.4.22  Limit States of Conductors and Ground Wires

An example of limit states of strength of conductors and ground wires are provided
in Table 5.3.

5.2.4.23  Limit States of Interface Components

Typical strength limit states of interface components are provided in Table 5.4.
204 E. Ghannoum

Table 5.3  Damage and failure limits of conductors and ground wires
Types Damage limit Failure limit
All lowest of: Ultimate tensile stress
types - Vibration limit, or (rupture)
- the infringement of critical clearances defined by
appropriate regulations, or
- 75 % of the characteristic strength or rated tensile
strength (typical range in 70 % to 80 %)

Table 5.4  Damage and failure limit of interface components

Type of interface components Damage limit1 Failure limit
Cable connectors: Dead-end and Unacceptable permanent Rupture
junction fittings and Suspension deformation or slippage
fittings
Insulators (porcelain and glass) 70 % strength rating or broken Rupture of pin, cap,
shed (glass only) cement or shed
Hardware Critical2 permanent Rupture of hardware
deformation or shear of bolts
1
Normally, hardware is designed in a manner to reduce or eliminate wear. Should wear be expected
because of point to point contact, it should be considered in the design. In such case, the damage
limit becomes: exceeding the excepted wear.
2
Defined as the state where the hardware cannot be easily taken apart.

5.2.5 Conclusions

This chapter summarizes advantages of RBD methods over common deterministic

based methods as well as the key features of standard IEC 60826 and its companion
document, Cigré TB 178.
Local weather conditions are taken into account during the design process, and
tools are provided in order to increase reliability and security if warranted either by
the importance of the line or by local conditions.
This RBD method of IEC 60826 Standard should provide for more economical
design for a given target reliability compared to safety factor methods or, inversely,
a higher reliability for given limit loads.
The IEC 60826 has now been integrated in many international standards (ex.
CSA C22.3, 2002, CENELEC EN 50341, IS 802, etc.) and utility practices. It repre-
sents a major contribution to the international trend in migrating toward reliability
based design concepts in overhead line design.

5.2.6 References

[1] IEC 60826 - Ed. 3.0: Design Criteria of Overhead Transmission Lines (2003)
[2] Cigré WG06.: Loading and strength of overhead transmission lines, Electra No.
129, March 1990 (1990)
5  Structural and Mechanical Design 205

[3] Cigré WG22-06.: Probabilistic Design of Overhead Transmission Lines, TB 178

(2001)
[4] Ghannoum, E., Orawski, G.: Reliability Based Design of Transmission Lines
According to Recent Advances by IEC and Cigré. International Symposium of
Probabilistic Design of Transmission Lines, Toronto, June 1986 (1986)
[5] Cigré SCB2.: Probabilistic Design of Overhead Transmission Lines (Cigré
Technical Brochure No. 178, February 2001)
[6] EN 50341–1:2001.: Overhead Electrical Lines exceeding AC 45 kV – Part 1:
General Requirements – Common Specifications (EN or CLC)
[7] ASCE 74.: Guidelines for Transmission Lines Structural Loading: Draft as of
June 2001
[8] NESC C2-2002.: National Electrical Safety Code C2-2002

5.3 Comparison of RBD Methods

5.3.1 Introduction

Since the publication of Technical Report IEC 826 in 1991, a number of other over-
head line design codes based on reliability design methods (RBD) have been pub-
lished. These documents have generally adopted many of the concepts given in IEC
826, but there are also some substantial differences.
The purpose of this Section 5.3 is to compare the methods adopted for the calcu-
lation of mechanical loadings for overhead power lines according to a number of
these design codes with the latest version of the IEC document, IEC 60826 Edition
3 (October 2003). It is noted that more recent versions of the compared codes have
been issued since this study was published by Cigré in 2006 in Cigré TB 289.
The main objective is to assess the level of consistency and the significance of
any major differences between the different codes.

5.3.2 Documents Compared and References

The documents compared in this section are: IEC [1], Cigré [3], EN [6], ASCE [7],
NESC [8].
In addition to the above standards, a comparison of wind loadings calculated in
accordance with the General (or Statistical) Approach of CENELEC (CLC) National
Normative Aspects documents produced by the National Committees of the follo-
wing countries was also included in the study:

EN 50341-3-7 Finland (NNA/FI)

EN 50341-3-9 Great Britain (NNA/UK)
EN 50341-3-16 Norway (NNA/NO)
EN 50341-3-18 Sweden (NNA/SE)
EN 50341-3-19 Czech (NNA/CZ)
206 E. Ghannoum

Note: The symbols adopted in this document are those used in the docu-
ments listed above, but it should be noted that usage varies across the range
of documents. Brief definitions of symbols are included in this paper, but for
full details reference should be made to the appropriate document. It should
also be noted that some parameters which are given similar titles or symbols
in different documents are defined differently.

5.3.3 Basis of Design

In Table 5.5 the basic design requirements of compared standards are provided,

particularly as regard reliability, safety, and security requirements.

5.3.4 Basic Design Equation

The relationships between loads and strength of line components are described in
Table 5.6.

Table 5.5  Summary of design requirements according to different codes and standards
IEC, Cigré EN ASCE NESC
Reliability requirements Yes Yes Yes Yes
– Line reliability levels 3 3 – –
– Line reliability factors – – 4 2
Return period T (years) of 50, 150, 50, 150, 500 50, 100, 200, ?
climatic load(1) (2) 500 400
Security requirements(3) Yes Yes Yes(4) Yes(5)
Safety requirements(6) Yes Yes Yes No
Strength coordination Yes Yes(7) Yes(7) No
Limit state for:
– Reliability (probabilistic) Damage Ultimate Damage Ultimate
Ultimate
– Security (deterministic) Failure Ultimate Failure Ultimate
– Safety (deterministic) Damage Ultimate Damage
Notes related to Table 5.5
(1)
Reference return period is usually 50 years
(2)
Also called limit load in IEC/Cigré and characteristic or extreme load in EN. No specific name
in ASCE.
(7)
General guidance only provided for strength coordination
(3)
To reduce risk of uncontrollable propagation of failures
(4)
General guidance given but the magnitude for security loadings is not defined
(5)
Concept stated only. No specific requirements, or guidance for calculation of security loadings is
given.
(6)
To ensure safe construction and maintenance conditions
(7)
General guidance only provided for strength coordination
5  Structural and Mechanical Design 207

Table 5.6  Basic design equations used in the compared standards

Code Design limit load < Design strength
IEC, Cigré Effect of QT < ΦS × ΦN × ΦQ × ΦC × RC(1)
EN Effect of γF × F50 or effect of FT < RK/γM(2)
ASCE Effect of γ × Q50 < Φ × Rn(3)
NESC OCF × L50(4) < Rn
The nomenclature varies between the various documents.
QT, FT, and LT are defined as climatic loads having a return period T.
γ, γF are load factors applied to loads having a reference return period (refer to Table 5.8 to
Table 5.10)
RC, RK and Rn are defined as the characteristic (IEC/Cigré, EN) or nominal (ASCE, NESC)
strength specified in appropriate standards, also called guaranteed strength or minimum strength
(or minimum failing load for IEC and percentage of an estimated breaking load for ASCE). This
value corresponds usually to an exclusion limit of 2 % to 5 % (i.e. the value being reached with a
98 % to 95 % probability). When not specified or calculated, the exclusion limit is conservatively
taken as 10 % (or with 90 % probability).
Notes related to Table 5.6:
(1)
ΦS, ΦN, and ΦQ are strength factors (IEC/Cigré) depending on strength coordination, number of
components, and quality level respectively. ΦC takes account of variation between the actual exclu-
sion limit of characteristic strength, RC, and the supposed 10 % exclusion limit. Values of Φ are
usually less than 1.0.
(2)
γM is the partial factor (EN) for material property covering unfavourable deviations from charac-
teristic strength, RK, inaccuracies in applied conversion factors and uncertainties in the geometric
properties and the resistance model.
(3)
Φ is a strength factor (ASCE) that takes account of variability of material, dimensions, workm-
anship, strength coordination and variation between nominal strength, Rn, and the (5 %) exclusion
value, Re. Values of Φ are usually less than 1.0.
(4)
OCF is the Overload Capacity Factor (NESC) to be applied to L50.

5.3.5 Combination of Loads

In Table 5.7, structure loading cases and conditions are provided for various stan-
dards and include probabilistic conditions as well as deterministic load cases.

5.3.6 Load Factors for Permanent and Variable Loads

5.3.6.1  Load Factors for Dead Loads

In Table 5.8, load factor for structure dead-loads (self weight of conductors and
structures) are provided for each assessed standard.

5.3.6.2  Load Factors for Wind Loads

The following Table 5.9 describes structure load factor related to wind actions on
overhead line components.

5.3.6.3  Load Factors for Ice Loads

Load factors are applied to ice loads (ice is a generic term use for frozen water that
deposits on structures and conductors such as freezing precipitation or rain, wet
208 E. Ghannoum

Table 5.7  Structure loading conditions

IEC Cigré EN ASCE ANSI
Reliability Based Design Conditions X X X X X
Limit wind load at reference temperature X X X - -
Reduced wind load at low temperature X X X X -
Uniform ice loads on all spans X X X (X) -
Uniform ice loads, transversal bending X X X (X) -
Unbalanced ice loads, longitudinal bending X X X (X) -
Unbalanced ice loads, torsional bending X X X X X
LP ice load and moderate wind load X X X - -
LP wind load and moderate ice load X X X - -
LP drag, moderate wind and ice load X X - (X) -
Deterministic Security Conditions
Torsional loads X - (X) (X) (X)
Longitudinal loads X - (X) (X) (X)
Deterministic Safety Conditions
Construction loads X - (X) (X) (X)
Maintenance loads X - (X) (X) (X)
Notes related to Table 5.7:
LP Low probability (having a high return period T)
X The load combination is specified in the document and bases for the calculation of loadings
given.
(X) The load combination is specified in the document, but only general guidance is given.

Table 5.8  Load factors for structure dead-loads (self weight of conductors and structures)
Code Reliability Based Design Cases Deterministic Load Cases
IEC, Cigré 1.0 -
EN 1.2 -
ASCE 1.0 -
ANSI - Metal Grade B = 1.50; Grade C = 1.90

snow, incloud icing. In Table 5.10, ice overload factors are given for different
standards.

5.3.6.4  Combination Factors for Moderate Loads

Table 5.11.

5.3.6.5  Other Load Factors

In addition to the various overload factors in the previous tables, there are some
additional structure load cases that are specified by each standard as described in
Table 5.12.
5  Structural and Mechanical Design 209

Table 5.9  Load factors for wind loads

Reliability Level 1 2 3
Return period T 50 years 150 years 500 years
(Q50) (Q150 = 1.102 (Q500 = 1.202 Q50)
Q50)
IEC, Cigré(1) 1.00 1.21 1.44
EN(2) 1.00 1.20 1.40
Reliability 1 2 4 8
Factor
Return period T 50 years 100 years 200 years 400 years
ASCE/NESC 1.00 1.15 1.30 1.40
NESC (Grade) Grade B Grade C
Wind 2.50 2.20
(Wire Tension) (1.65) (1.30)
Notes related to Table 5.9:
(1)
Default load factors (IEC, Cigré) based on coefficient of variation (COV) up to 0.30 for wind
speed and derived from the Gumbel distribution function
(2)
Load factor (EN) depending on the selected reliability level and taking account of the possibility
of unfavourable deviations from the characteristic load values, inaccurate modelling and uncertain-
ties in the assessment of the load effects
(3)
Calculated using “super-station methods”

Table 5.10  Load factors applicable to ice thickness or unit weight

Reliability Level
Return 1 2 3
Period T 50 years 150 years 500 years
IEC, Cigré (Q50) (Q150)(1) (Q500)(1)
Ice thickness (2) 1.00 1.15 1.30
Ice weight (2) 1.00 1.20 1.45
EN: Ice weight (3) 1.00 1.25 1.50
Reliability Factor 1 2 4 8
Return period T 50 years 100 years 200 years 400 years
ASCE: Ice weight 1.00 1.15 1.30 1.40
ANSI (Grade) Grade C Grade B
ANSI (metal 2.2(4) 2.5(4)
structures)
Notes:
(1)
If values of Q150, and Q500 are available from statistical analysis, these may be adopted.
(2)
Default load factors based on COV up to 0.30 for ice thickness and COV up to 0.65 for unit ice
weight and derived from the Gumbel distribution function
(3)
Load factor (EN) depending on the selected reliability level and taking account of the possibility
of unfavorable deviations from the characteristic load value, inaccurate modeling and uncertainties
in the assessment of the load effects
(4)
Calculated using “super-station methods”
(5)
Load factors may be applied either to ice thickness or to the weight of ice per unit length of
conductor
210 E. Ghannoum

Table 5.11  Combination of ice and wind structure loads

Code Wind load (with low probability ice) Ice load (with low probability wind)
IEC, Cigré Average of yearly maximum (Q1) Average of yearly maximum (Q1)
EN 0.40 0.35
for 3 year return value (Q3) for 3 year return value (Q3)
ASCE 0.6 (glaze ice) -
0.5 (wet snow) -
ANSI - -

Table 5.12  Additional structure load factors

Wind load at low temperature (Min Security Safety load(1) Construction
Code yearly temperature with return period T) load(1) and maintenance loads
IEC 0.62 = 0.36 1.0 1.5 (if careful works)
Cigré – - -
EN not specified 1.0 1.5
ASCE - 1.0 1.0
NESC - - -
Note: applicable to Table 5.12
Security and Safety loadings are deterministic in nature and their magnitude is independent of
(1)

reliability level.

5.3.7 Wind Loads

5.3.7.1  Terrain Roughness

IEC, Cigré, EN and ASCE define the roughness factor KR, the roughness coefficient α,
or the terrain factor kT and the ground roughness parameter zO for four different terrain
categories. The definitions and values used in each standard are provided in Table 5.13.

5.3.7.2  Reference Wind Speed VR

The reference wind speed (basic wind speed for ASCE) is defined as the wind speed
measured at a height of 10 m above ground level, corresponding to an averaging
period of 10 min (mean wind speed) or 2–3 seconds (gust wind speed) and having a
return period T.
The reference wind speed VR (basic wind speed V for ASCE) measured in
weather stations typical of open flat terrain, such as airports (terrain category B for
IEC and Cigré; II for EN; C for ASCE), is identified as VRB (VR(II) for EN and V for
ASCE). Where available wind data differ from these assumptions, conversion
methods are provided in the Codes (Table 5.14).

5.3.7.3  Unit Action of Wind Speed on any Component of the Line

The following expressions are used:
5  Structural and Mechanical Design 211

Table 5.13  Terrain categories described in each standard

Terrain
Category Roughness characteristic Code Value
KR α
A Large stretch of water upwind, flat IEC, 1.08
coastal areas Cigré
I Rough open sea, lakes with at least EN kT = 0.17 z0 = 0.01
5 km fetch upwind and smooth flat
country without obstacles
B Open country with very few IEC, 1,0
obstacles, airports or cultivated Cigré
fields with few trees and buildings
II Farmland with boundary hedges, EN kT = 0.19 z0 = 0.05
occasional small farm structures,
houses or trees
C Open terrain with scattered ASCE (1.0) 0.14
obstructions (Exposure C as defined (NESC)
in ASCE 7–02)
C Terrain with numerous small IEC, 0.85
obstacles of low height (hedges, Cigré
trees and buildings)
III Suburban or industrial areas and EN kT = 0.22 z0 = 0.30
permanent forests
D Suburban areas or terrain with many IEC, 0.67
tall trees Cigré
IV Urban areas in which at least 15 % EN kT = 0.24 z0 = 1.0
of the surface is covered with
buildings with mean height > 15 m
V Mountainous and more complex EN shall be evaluated separately
terrain where the wind may be possibly by meteorologists
locally strengthened or weakened
Note: the ASCE and NESC documents refer to one terrain category only.

IEC/Cigré  a = ½ μ KR2 VRB2 Cx G

• μ  air density
• KR  roughness factor at the line location
• VRB  reference wind speed at flat and open terrain category B
• CX  drag coefficient for the component X being considered
• G  combined wind factor depending on height above ground level, wind gust,
dynamic response and terrain category

EN  a = ½ ρ KR2 Kh2 (VR(II))2 CX GX Gq

• ρ  air density
• KR  roughness factor at the line location
212 E. Ghannoum

Table 5.14  Reference wind speed VR

Gust wind speed (Average period of Mean wind speed (Average period of
Code 2–3 s) 10 min)
IEC, Cigré - VR
EN VR = Vg VR = Vmean
ASCE V -
ANSI V -

• Kh  takes in account height effect for VR(II)

• VR(II)  reference wind speed at flat and open terrain category II
• CX  drag factor for the component X being considered
• GX  structural resonance factor for the component X being considered
• Gq  gust response factor depending on height above ground level and terrain
category (equals 1 if the gust wind speed is considered)

ASCE  a = ½ μ KZ2 V2 Cf G

• μ  air density
• KZ  takes in account terrain category and height above ground level
• V  basic wind speed at flat and open terrain category C
• Cf  force (i. e. drag) coefficient for the component being considered
• G  gust response factor including span effect, depending on height above
ground level and terrain category (This gust factor is derived from the Davenport
model and ASCE neglects the resonant component of this model.)

5.3.7.4  Comparison of Combined Wind Factors for Conductors

The combined wind factor, G, for conductors which takes account of gust effects
(dependant on span length), conductor height above ground level and terrain cate-
gory is compared in this chapter with the following parameters:

Gc GL of the IEC, Cigré document, where Gc is the combined wind factor for con-
ductors with a span length of 200 m and GL is the span factor (GL = 1.0 for a span
length of 200 m),
Kh2 Gq GXc of the EN code, where Gq is the gust response factor and GXc the reso-
nance factor for the conductor, also termed “span factor”,
KZ Gw of the ASCE code, where Gw is the gust response factor for conductors
(wires) including span effect for wires. To make this value comparable it is mul-
tiplied by (V/VM)2 = 1.432 = 2.04, to take in account the reference to the 10 minute
average wind speed.

The comparison of global wind factor for conductors resulting from span factor,
height effect, and gust response to wind are given in Table 5.15.
The values obtained in Table 5.15 were calculated using effective conductor
heights given in Table 5.16.
5  Structural and Mechanical Design 213

Table 5.15  Comparison of combined wind factors for conductors at terrain category B (IEC) or
equivalent
Height in m Code 200 Span length in m 400 600
IEC, Cigré 1.84 1.73 1.624
10 EN 1.94 1.77 1.72
ASCE 1.51 1.39 1.31
IEC, Cigré 2.25 2.12 1.99
30 EN 2.32 2.17 2.08
ASCE 1.81 1.68 1.61
IEC, Cigré 2.44 2.30 2.16
50 EN 2.60 2.43 2.33
ASCE 1.96 1.84 1.76

Table 5.16  Effective conductor heights used to calculate conductor wind loads
Code Theoretical height Location For support calculation
IEC Centre of gravity Lower third of sag Attachment point of (middle)
conductor
Cigré - - -
EN Centre of wind - -
pressure
ASCE Centre of wind Higher third of sag (for -
pressure no wind)

Table 5.17  Comparison of combined wind factors for towers

Terrain type
Height in m to CG of panel(1) Code A B C D
IEC 1.70 1.95 2.50 3.30
10 Cigré 1.70 1.95 2.55 3.30
EN 1.69 1.94 2.38 3.96
IEC 1.95 2.32 2.95 3.95
30 Cigré 2.00 2.30 3.00 3.90
EN 2.06 2.51 3.01 4.48
IEC 2.15 2.55 3.35 4.40
50 Cigré 2.15 2.50 3.25 4.40
EN 2.29 2.86 3.69 7.50
It is noted that the factors given in Table 5.17 are based on tower panel height above ground level
is measured at the centre of gravity (CG) of the panel.

5.3.7.5  Comparison of Combined Wind Factors for Towers

The combined wind factor, G, for lattice towers (depending on gust, height and
terrain category) as used by IEC, Cigré TB 109, is compared in this chapter with the
following parameters:
214 E. Ghannoum

Gt of the IEC/Cigré document, where Gt is the combined wind factor for towers
Kh2 Gq GXt of the EN code, where GXt is the drag factor for the tower (GXt taken as 1.0)
KZ Gt of the ASCE code, where Gt is the gust response factor for the tower. To make
this value comparable it is multiplied by (V/VM)2 = 1.43 2 = 2.04, to take in
account the reference to the 10 minute averaging wind speed.

5.3.8 Drag Coefficient

For conductors in IEC, Cigré and EN the recommended drag coefficient or factor
CX = 1.0.
ASCE recommends:

1,2 for shield wire

1.23-0.032.(RC)/3 × 104 for conductors
0.9 for snow coated conductor.

For lattice towers in IEC, Cigré and EN the same magnitude for the drag coefficient
CxT is recommended, depending on the solidity ratio of tower panel (ratio between
solid steel area and overall panel surface). In ASCE slightly different values are
given as indicated in Figure 5.2.

5.3.9 Span Factors

The span factor represents the relationship between span length and wind spatial
effect on the span. It is recognized that the longer the span, the less a high wind

4,2

3,8

3,4

3
CxT

2,6 IEC, Cigré,

CLC
2,2

1,8 ASCE

1,4
0 0.2 0.4 0.6 0.8 1
Solidity ratio

Figure 5.2  Drag coefficients for lattice towers.

5  Structural and Mechanical Design 215

event can uniformly encompass all the span. IEC, Cigré and EN have adopted
similar span factors in all terrain categories (for example, in a terrain type B, the
IEC span factor GL gradually drops from 1.0 for a 200 m span to 0.85 for an 800 m
span). Comparatively, the ASCE has adopted lower span factors, where this factor
drops linearly from 1.0 to 0.7 for spans of 350 to 500 m respectively.

5.3.10 Ice Loads

In IEC, Cigré and EN the basic ice load, g (N/m), is referred to a conductor of diameter
30 mm 10 m above ground level. In ASCE reference is not given (refer to Table 5.18).
Many standards also require designing for unbalanced icing conditions as indi-
cated in Table 5.19.

5.3.11 Combined Wind and Ice Loads

Althouth ice accretion on conductors is rarely in the form of a perfect cylinder, an

equivalent round ice diameter is defined in all standards as being the uniform ice
thickness that leads to the same unit weight on the conductor.
In addition to the above, many standards recognize that not only ice thickness is
important for design of conductors and structures, but also the ice density, and its

Table 5.18  Comparison of reference ice loads

Code Ice variable Adjustment factors Diameter (Kd) Height (Kh)
IEC weight X X(1)
Cigré weight X X(1)
thickness X X(1)
EN weight X(2) -
ASCE thickness - -
Notes applicable to Table 5.18:
(1)
only for precipitation icing
(2)
no values specified

Table 5.19  Unbalanced icing condition

Transversal
Code Longitudinal bending bending Torsional bending
IEC α1 = 0.28, α2 = 0.7 α1 = 0.28, α2 = 0.7 α1 = 0.28, α2 = 0.7
Cigré (TB α1 = 0.28, α2 = 0.7 α1 = 0.28, α2 = 0.7 α1 = 0.28, α2 = 0.7
109)
EN α1 = 0.30, α2 = 0.7 α1 = 0.50, α2 = 1.0 α1 = 0.30, α2 = 0.7
ASCE mentioned but no values not specified mentioned but no values
given given
216 E. Ghannoum

Table 5.20  Comparison of drag coefficients and densities

Precipitation In-cloud ice In-cloud ice Precipitation
Code Wet snow Soft rime Hard rime Glace ice
IEC Drag 1.0 1.2 1.1 1.0
Density 600 600 900 900
Cigré Drag 1.0/1.4 1.2/1.7 1.1/1.5 1.0/1.4
Density 600/400 600/400 900/700 900/900
EN Drag 1 1.2 1.1 1.0
Density 500 300 700 900
ASCE Drag 0.9 see [8] see [8] see [8]
Density 300-800 300 900 900

corresponding drag coefficient. Table 5.20 describes various ice densities and drag
coefficients for each of the assessed standard.

5.3.12 Failure and Containment Loads (Security Loads)

IEC, Cigré, EN and ASCE define the torsional and longitudinal structure loads.
The guidelines given in these codes are in principle the same and consist of loads that
attempt to reduce the risk of cascading failures in transmission lines. These loads are
broken conductor loads in different combinations, mostly under every day conditions.
IEC and Cigré provide guidelines for additional security for important lines and
lines subjected to heavy icing conditions.

5.3.13 Construction and Maintenance Loads (Safety Loads)

IEC, Cigré, EN and ASCE provide guidelines for construction and maintenance
loads. These loads are treated as deterministic loads because their purpose is to pro-
tect workers during construction and maintenance activities.

5.3.14 Other Loads

EN provides additional guidelines for loads caused by:

• Short-circuit
• Avalanches
• Earthquakes.

ASCE provides additional guidelines for loads caused by:

• Galloping
• Flooding
• Vibration.
5  Structural and Mechanical Design 217

5.3.15 General Comparative Overview

All considered standards and practices share many common features. All adopt a
reliability based (i.e. semi-probabilistic) method for calculation of climatic loads,
essentially similar to IEC 60826. However, the European Standard EN 50341–1 still
provides a deterministic “Empirical Approach” as an alternative for some countries.
IEC 60826 and ASCE 74 adopt alternative limit states of “damage”, for use with
reliability based and safety loadings, and “failure” for use with security loading
cases. The other documents use “ultimate” limit state for all these loading cases. EN
specifies a “serviceability” limit stage for the checking of clearances (see Table 5.4).
Climatic loadings tend to be the most important in the design of overhead lines,
but all documents also give guidance on security loading cases and construction and
maintenance load cases to be used for design. Neither security nor construction and
maintenance loadings lend themselves completely to a reliability-based approach.
In most documents security loading cases are based on arbitrary assumptions about
hypothetical broken conductors causing unbalanced longitudinal occurring. These
conditions are combined with climatic loadings based on low return periods.
There are minor differences in the formulation of the basic design equations, but
climatic loadings are always based on weather events of a specified return period,
usually 50 years (see Table 5.5).
The range of loading cases considered differs between the various codes, with
IEC offering the most extensive range. IEC also provides the most detailed informa-
tion on strength co-ordination between different elements, with only general gui-
dance being given in EN and ASCE, and no explicit mention of this subject in NESC
(see Table 5.2).
For load factors on wind loadings, IEC applies factors to wind speed. The other
documents apply factors to wind load or pressure factors (Table 5.9). For ice loa-
ding, IEC offers alternative approaches of applying factors either to ice thickness or
weight. The other documents apply factors to ice weight (Table 5.10).
For combinations of weather events occurring simultaneously, for example wind
and ice loading, all documents assume that improbable events will not occur simul-
taneously, i. e. the maximum ice loading based on a 50 year return or higher return
period event would occur simultaneously with a wind loading of much lower return
period (see Tables 5.11 and 5.12).
The terrain categories and defining parameters used in the different codes are
listed in Table 5.13 and the nearest equivalents between the different codes are grou-
ped together. It should be noted that the subsequent wind loading calculations are
based on Terrain Type B to IEC, taken as equivalent to Category II in EN or Category
C in ASCE.
IEC and EN take a reference wind speed based on a 10 minute mean wind speed,
whereas ASCE/NESC adopt a 3 second gust wind speed (Table 5.14).
Formulas for effective wind pressure, including the effects of terrain category,
drag coefficient and gust factors are given in Table 5.15.
Combined wind loading factors including the effects of gust, height, and terrain
are compared in Table 5.15 for conductors and in Table 5.17 for lattice towers.
A comparison of the values of drag coefficient for conductors and lattice towers
are given in Clause 5.3.8. IEC/Cigré and EN all specify a value of 1.0 for conductors.
218 E. Ghannoum

ASCE adopts a formula based on Reynolds Number for un-iced conductors, with a
value of 1.2 specified for shield-wires and 0.9 for snow-covered conductors. For
tower steelwork IEC/Cigré and EN use the same formula. Different, albeit similar
values are given in ASCE.
A comparison of the span factors adopted in the different codes is given in Clause
5.3.9. There are fairly small differences between IEC/Cigré and EN, but ASCE
varies significantly from the other documents.
Table 5.18 compares the use of adjustment factors on ice weight and thickness
depending on conductor height and diameter, and Table 5.19 considers the treatment
of unbalanced ice loading in the various documents.
Table 5.20 highlights the differences in drag coefficient on iced conductor and
the density of ice adopted for different types of icing.
Clause 5.3.12 gives brief notes on the failure containment loads given in the dif-
ferent documents.
Clause 5.3.13 notes that all documents contain guidance on construction and
maintenance loadings.
Clause 5.3.14 looks at other loading conditions not mentioned above. It notes the
ASCE provides guidance on galloping, vibration, and flooding, which are not expli-
citly mentioned in the other documents. Also EN provides guidance on loadings gene-
rated by short-circuits, avalanches and earthquakes which are not addressed in the
other documents.
Appendix A gives a comparison between conductor wind pressures, including
gust, height and span related effects for a range of conductors for wind spans varying
from 200–600 m and mean conductor heights of 10 m, 30 m, and 50 m. Terrain type
B according to IEC or the closest equivalents to the other codes is assumed. This
information is given in tabular form in Tables 5.21, 5.22, and 5.23, and in graphical
form in Figure 5.4. It is shown that IEC/Cigré and EN agree fairly closely, but ASCE
gives significantly lower values than either of the other documents. The main reason
for this is believed to be the use of the simplified Davenport gust response model. It
may be noted that the use of the full Davenport model is allowed as an option in
ASCE, and this gives results much closer to IEC.
Table  5.24 provides a comparison of conductor wind loadings calculated in
accordance with the EN-NNA documents for various European countries, with IEC/
Cigré, ASCE, and the EN Main Body. The countries concerned are Norway, Sweden,
Finland, Czech Republic, and the UK. This information is also provided in graphical
form in Figure 5.4. It appears that the Norwegian, Swedish, Finnish, and UK NNAs
give values similar to ASCE, whilst the Czech NNA follows EN.

5.3.16 Conclusions

The main objective of this Section was to compare the recently published Inter­
national Standard IEC 60826 Ed.3 on “Design Criteria of Overhead Transmission
Lines”, issued October 2003 with the existing codes EN 50341–1, ASCE 74 and
NESC C2, and to assess the consistency. These codes followed the Reliability Based
Table 5.21  Effective wind pressures on conductors for terrain category B to IEC assuming a reference wind speed of 24 m/s, a 10 minute average period and
mean height of conductor of 10 m

Adjusted Air Effective Wind Pressures (N/m2) for given

Terrain Average wind speed Drag density Combined Factors for given span length span length
5  Structural and Mechanical Design

Code Category Time (m/s) factor (kg/m3) 200 m 300 m 400 m 500 m 600 m 200 m 300 m 400 m 500 m 600 m
IEC B 10 min 24.0 1.0 1.225 1.85 1.798 1.743 1.691 1.648 653 634 615 597 582
Cigré B 10 min 24.0 1.0 1.225 1.85 1.798 1.743 1.691 1.648 653 634 615 597 582
EN II 10 min 24.0 1.0 1.225 1.771 1.703 1.654 1.617 1.586 625 601 584 570 560
ASCE C 3 sec 34.3 1.0 1.226 0.738 0.703 0.679 0.662 0.639 532 507 490 478 461
NESC C 3 sec 34.3 1.0 1.226 0.738 0.703 0.679 0.662 0.639 532 507 490 478 461
Notes:
1) The terrain categories quoted are considered to be the nearest equivalents from the various codes.
2) The ASCE and ANSI adjusted wind speed values have been obtained by multiplying the 24 m/s 10 min mean wind speed by 1.43 to convert to 3 sec gust
wind speed, as recommended in Appendix E of ACSE 74.
3) The combined factors and effective wind pressures include the effects of gust, conductor height and span length.
219
 220

Table 5.22  Effective wind pressures on conductors for terrain category B to IEC assuming a reference wind speed of 24 m/s, a 10 minute average period and
mean height of conductor of 30 m

Adjusted Air Effective Wind Pressures (N/m2) for given

Terrain Average wind speed Drag density Combined Factors for given span length span length
Code Category Time (m/s) factor (kg/m3) 200 m 300 m 400 m 500 m 600 m 200 m 300 m 400 m 500 m 600 m
IEC B 10 min 24.0 1.0 1.225 2.23 2.168 2.101 2.038 1.987 787 765 741 719 701
Cigré B 10 min 24.0 1.0 1.225 2.23 2.168 2.101 2.038 1.987 787 765 741 719 701
EN II 10 min 24.0 1.0 1.225 2.321 2.232 2.169 2.120 2.080 819 788 765 748 734
ASCE C 3 sec 34.3 1.0 1.226 0.884 0.821 0.679 0.803 0.789 638 610 592 579 569
ANSI C 3 sec 34.3 1.0 1.226 0.884 0.821 0.679 0.803 0.789 638 610 592 579 569
Notes:
1) The terrain categories quoted are considered to be the nearest equivalents from the various codes.
2) The ASCE and ANSI adjusted wind speed values have been obtained by multiplying the 24 m/s 10 min mean wind speed by 1.43 to convert to 3 sec gust
wind speed, as recommended in ACSE 74 Appendix E.
3) The combined factors and effective wind pressures include the effects of gust, conductor height and span length.
E. Ghannoum
Table 5.23  Effective wind pressures on conductors for terrain category B to IEC assuming a reference wind speed of 24 m/s, a 10 minute average period and
mean height of conductor of 50 m

Adjusted Air Effective Wind Pressures (N/m2) for given

Terrain Average wind speed Drag density Combined Factors for given span length span length
5  Structural and Mechanical Design

Code Category Time (m/s) factor (kg/m3) 200 m 300 m 400 m 500 m 600 m 200 m 300 m 400 m 500 m 600 m
IEC B 10 min 24.0 1.0 1.225 2.43 2.362 2.289 2.221 2.165 857 833 808 784 764
Cigré B 10 min 24.0 1.0 1.225 2.43 2.362 2.289 2.221 2.165 857 833 808 784 764
EN II 10 min 24.0 1.0 1.225 2.60 2.50 2.43 2.38 2.33 918 883 858 839 823
ASCE C 3 sec 34.3 1.0 1.226 0.96 0.92 0.90 0.88 0.86 695 667 648 634 624
ANSI C 3 sec 34.3 1.0 1.226 1.01 0.97 0.94 0.90 0.90 727 697 676 646 646
Notes:
1) The terrain categories quoted are considered to be the nearest equivalents from the various codes.
2) The ASCE and ANSI adjusted wind speed values have been obtained by multiplying the 24 m/s 10 min mean wind speed by 1.43 to convert to 3 sec gust
wind speed, as recommended in ACSE 74 Appendix E.
3) The combined factors and effective wind pressures include the effects of gust, conductor height and span length.
221
222 E. Ghannoum

Table 5.24  Effective wind pressures on conductors under extreme wind loadings having been
calculated in accordance with IEC, EN, ASCE and the EN-NNAs of Finland, Sweden, Norway and
the UK
Comparison of Conductor Wind Loads derived from different Standards
Basic Data Symbol Unit Value
Reliability level 1
Return period of loads T years 50
Conductor diameter d mm 30
Conductor height (average) h m 30
Span length L m 400
Terrain category II
Altitude (from sea level) H0 m 0
Air temperature Ta °C 15
Reference wind speed (terrain II) VR(II) m/s 24.0
IEC ASCE EN EN - NNA
Load Symbol Unit 60826 74 MB FI SE NO UK
Parameters (EN)
(EN)
Terrain B C II II II II II
category
Terrain factor kT - KR = 1.00 0.19 0.19 0.19
Ground z0 - 0.05 0.05 0.05
roughness
parameter
Reference VR m/s 24.0 34.3 24.2 24.2 24.0 21.4
wind speed
at site
Span factor GXc - GL = 0.94 0.81 0.65 0.50 0.70 0.51
Height Kh - KZ = 1.26 1.21 1.21 1.21 1.20
correction
factor
Gust factor kg - 1.36 1.36 1.36 1.06
for wind
speed
Gust factor - Gw = 0.65 1.84 1.84 1.84
for wind G q = k 2g
pressure
Combined GC - 2.23
wind factor
for L = 200  m
(continued)
5  Structural and Mechanical Design 223

Table 5.24 (continued)
IEC ASCE EN EN - NNA
Combined G - 2.10 1.03 2.17 1.74 1.88 2.21
wind factor
(See § 7.4)
Mean wind Vh = Kh m/s 29.2 29.2 29.0
speed at VR
conductor
height
Gust wind kg Vh m/s 39.6 39.6 39.3
speed at
conductor
height
Altitude - 1.00 1.00 1.00 1.00
factor for
density
Temperature - 1.00 1.00 1.00 1.00
factor for air
density
Air density ρ or μ kg/m3 1.225 1.226 1.23 1.23 1.23 1.22
Gust (μVh2/2) N/m2 788 959 959 920 946
pressure at Gq
conductor
height
Drag factor CXc - 1.00 1.00 1.00 1.00 1.00 1.00
Load factor γw - 1.00 1.00 1.00 1.00 1.30 1.00 1.00
for wind load
(or Speed)
Design wind Qd N/m2 788 959 959 1196 946
pressure
Wind load of Qc = Qd N/m 22.3 23.3 18.6 17.9 19.9
conductor GXc d
per unit
length
Effective Fc = QcL N 8911 9305 7444 7176 7948
wind load on
wind span
Effective Qd GXc N/m2 743 593 775 620 598 662 615
wind
pressure on
conductor
224 E. Ghannoum

Terrain Category B to IEC Terrain Category B to IEC

100 100
IEC/Cigré
IEC/Cigré
800 ASCE/NESC
800 ASCE/NESC
CLC
600 CLC
600
400
400
200
200
0
0
CLC CLC
200 200
300 ASCE/NESC
300 ASCE/NESC
400 400
500 IEC/Cigré
Span (m) 600 Span (m) 500 IEC/Cigré
600

Terrain Category B to IEC

100
IEC/Cigré
800 ASCE/NESC
CLC
600

400
200
0
CLC
200
300 ASCE/NESC
400
Span (m) 500 IEC/Cigré
600

Figure 5.3  Comparison of Effective Wind Pressure on conductors for EN (CLC), ASCE/NESC,
and IEC/Cigré for various span lengths and heights above ground level for Terrain Category B to
IEC or equivalent.

Height=30 mm, Span=400 mm

1000

800

600

400

200

0
C

CE

FI

SE

NO

UK

CZ
CL
IE

A/

A/
AS

A/
A/
A/
NN

NN

NN
NN
NN

Figure 5.4  Effective wind pressure calculated according to various standards.

5  Structural and Mechanical Design 225

Design method developed in the Technical Report IEC 826 issued in 1991, a previ-
ous version of IEC 60826. Cigré WG B2.06 already contributed to the review of this
Technical Report in Cigré TB 109 and with the Companion Document TB 178. The
detection of the most significant differences between IEC 60826 and the other Codes
is a new step in the improvement of RBD Codes.
This Section III concludes with the comparison of results of the numerical calcu-
lations of conductor wind loadings between the different codes, including the Nati­
onal Normative Annexes of EN 50341–3.

Appendix A - Sample Comparison of Wind Pressure for Different Codes

Calculations carried out by Chris Thorn and Pekka Riisiö in Tables 5.21, 5.22,
5.23, and 5.24.
The effective wind pressures, taking account of gust, span and height effects
have been calculated in Tables 5.21, 5.22, 5.23, and 5.24 for wind span lengths bet-
ween 200 m to 600 m:
Table 5.21: mean conductor height: 10 m
Table 5.22: mean conductor height: 30 m
Table 5.23: mean conductor height: 50 m
Figure 5.3 for mean conductor heights 10, 30 and 50 m
Table 5.24 provides a comparison between conductor wind loadings, calculated
in accordance with the EN-NNA documents for various European countries, with
IEC/Cigré, ASCE, and the EN Main Body:
Table 5.24: mean conductor height: 30 m; span length: 400 m.
Figure 5.4: mean conductor height: 30 m; span length: 400 m.
General Assumptions
Wind Speed: 24 m/s (10 minute mean wind speed, 10 m height, terrain type B)
Conductor diameter: 30 mm
Air temperature 15 °C
Notes: Terrain Categories are not entirely compatible between the different codes,
but the closest equivalent has been chosen.

5.4 Tower Top Geometry and Mid-span Clearances

5.4.1 Introduction

The aim of this section of Chapter 5 based on Cigré TB 348 is to give guidance on
the calculation of the geometric dimensions of the tower top to ensure that the inter-
nal electrical clearances necessary between live parts and earthed structures as well
as between conductors in mid span are achieved.
The available clearances depend on the conductor and insulator set positions which
vary due to wind action. The wind speed causes swinging of conductors and reduces the
clearances. Therefore, wind loads are significant for defining the tower top geometry.
Cigré TB 48 on “Tower Top Geometry” [1], issued in 1996, already provided guide­
lines to check the internal clearances at towers. This section reviews those guide-
lines and extends them to the clearances between conductors within the span.
226 E. Ghannoum

As the design of mid span clearances is more complex due to the additional influ-
ence of conductor sag and possible asynchronous oscillation of adjacent conductors
due to differential wind, a questionnaire was issued by Cigré WG B2.06 to collect
and compare existing national design methods.
The approach developed in Cigré TB 348 represents an improvement on the vari-
ous empirical methods in use up to now.

• Section 5.4.2, Part, 1 analyses the existing practices on internal clearances, for

which information was gathered by the questionnaire and completed with infor-
mation from European Standards, including the empirical equations;
• Section 5.4.3, Part 2, deals with the determination of swing angles of conductors
and insulator sets under wind load. It is assumed that wind speed is randomly dis­
tributed. Especially “still air” and “high wind” conditions are considered;
• Section 5.4.4, Part 3, compares available information on electrical clearances pha-
se-to-earth and phase-to-phase required for the various conditions of lightning or
switching surge impulse and power frequency voltages;
• Section 5.4.5, Part 4, gives proposals for the coordination of conductor positions
and electrical clearances;
• Section 5.4.6, Part 5, demonstrates the approach developed by means of an appli-
cation example.

The study presented is valid for all self-supporting and guyed overhead line supports
and for all line configurations (vertical, horizontal, triangle).
The design guidance is limited to the influence of wind actions, but the approach
philosophy can easily be extended to other issues such as:

• Differential conductor sags caused by differential ice accretion

• Combined ice and wind loading
• Ice drop
• Galloping
• Live line maintenance.

No design guidance is provided for those issues, but they are considered in 5.4.2.,
Part 1.

5.4.2 Part 1: Existing National Practices

In most countries existing practices for the definition of tower top geometry refer to
National Standards or Codes, which quote appropriate formulae and meteorological
conditions (temperature, wind, ice) to determine swing angles of:

• insulator sets at tower top

• conductor at mid span,

to associate them with corresponding internal electrical clearances.

5  Structural and Mechanical Design 227

Part 1 of Cigré TB 348 reviews the information referring to the different aspects
of loading cases considered for the determination of internal clearances.
The information received from the answers to the questionnaire was comple-
ted with the design rules for clearances available in the National Normative
Aspects (NNAs) EN 50341–3 to the European Standard EN 50341–1 [2]. Since
the publication of Cigré TB 48, national practices have been changed in some
countries.
The following countries responded to the WG B2.06 questionnaire:
Australia, Belgium, Brazil, Czech Republic, Finland, France, Germany, Great
Britain, Italy, Japan, Lithuania, the Netherlands, New Zealand, Norway, Serbia, South
Africa, Ukraine and USA.
The practices of the following countries have also been considered, as they are
detailed in the NNAs of EN 50341–3:
Austria, Denmark, Estonia, Greece, Iceland, Ireland, Portugal, Spain and Sweden.
The internal clearances according to EN 50341–1 are summarized in Table 5.25:

• Dpe is the phase-to-earth clearance, that refers to the internal clearance between
phase conductors and objects at earth potential, including the tower structure and
the earth wire;
• Dpp is the phase-to-phase clearance, that refers to the internal clearance between
phase conductors.

Dpe and Dpp are required to withstand lightning (fast front) or switching (slow front)
impulse voltages. Dpe_pf and Dpp_pf are required to withstand power frequency voltages.
EN 50341–1 distinguishes three load cases for the internal clearances. They are
summarized in Table 5.26.
All information gathered from the questionnaire and EN 50341 has been summa-
rized in Cigré TB in various synoptic tables. All wind and/or ice load cases are
classified per country with the corresponding type of clearance.
According to the answers to the questionnaire the most used load cases corres-
pond to the recommendations of EN 50341–1. They have been ranked in Table 5.27
with the preferred clearance type according to Table 5.26. The rate of preference is
given in %.

Table 5.25  Internal clearances according to the European Standard EN 50341-1

Clearance for lightning or Clearance for power
Voltage level switching impulse voltage frequency voltage
Nominal Highest system Phase-to-­ Phase-to-­ Phase-to-­ Phase-to-­
voltage voltage earth phase earth phase
UR US Dpe Dpp Dpe_pf Dpp_pf
kV kV m m m m
110 123 1.00 1.15 0.23 0.37
220 245 1.70 2.00 0.43 0.69
380 420 2.80 3.20 0.70 1.17
480 525 3.50 4.00 0.86 1.47
228 E. Ghannoum

Table 5.26  Type of internal clearance per load case (EN 50341–1)
Clearance Phase-to-­ Phase-to-­
type earth phase Load case
1 Dpe Dpp Still air conditions with maximum conductor
temperature or ice load
2 k1 · Dpe(1) k1 · Dpp (1) Reduced wind load (T = 3 years)
3 Dpe_pf Dpp_pf Extreme wind load (T = 50 years)
(1) Reduction factor k1 is defined by the National Normative Aspects of EN 50341–3

Table 5.27  Ranking of load cases for the internal clearances

Ranking Preferred clearance type (Table 5.25)
No. % Load case No. Clearance %
1 76 % Maximum temperature 1 Dpe, Dpp 72 %
2 68 % High wind load 3 Dpe_pf, Dpp_pf 52 %
3 64 % Reduced wind load 2 k1 · Dpe, k1 · Dpp 60 %
4 60 % Extreme ice load 1 Dpe, Dpp 44 %

Galloping is mentioned by 32 % of the countries as a determining factor for mid

span clearances. Other load cases are applied by only 24 % or less of the countries.
Some countries consider different wind or ice loads on adjacent conductors.
Many countries use empirical equations for the mid span clearance between
adjacent conductors if they are made of the same material and having the same
cross-section and sag. These equations include three terms:

• The first term indirectly takes into account the differing wind speeds on the con-
ductors. This term depends on the sag, the mean length of the suspension insula-
tor set, the swing angle and the relative position of both conductors;
• The second term takes into account the system voltage;
• Finally the third term considers the outside diameter of conductor bundles, if any.

A synoptic table in Cigré TB 348 summarizes the parameters used by each country
for their empirical equation.

5.4.3 Part 2: Swing Angles

Part 2 of Cigré TB 348 compiles the material on swinging of conductors due to wind
loads, which is available to the WG B2.06. The wind action varies with time and space
and can be described as randomly distributed using statistical approaches [1; 2; 3; 4].
Due to the boundary effect, the wind velocities increase with the height above
ground. The longer the wind span the more the wind velocity will vary along this span.
Relevant standards, take care of this effect by introducing a span factor. Additionally,
the swing angles depend on line parameters such as ratio of wind span to weight span,
conductor type, etc.
5  Structural and Mechanical Design 229

For the design of the tower top geometry the same meteorological data should be
used as for the Reliability Based Design (RBD) [3; 4] methods.
Various experimental investigations were carried out on test stands in the past to
study the relation between wind velocities and swing angles. The basic results are:

• The swing angle observed for a measured wind velocity, whether instantaneous
or average over a certain period of time, scatter considerably;
• The majority of observed swing angles stay well below those determined theo-
retically using observed peak wind velocities and generally agreed formulae.

To determine the distribution of swing angles it is necessary to establish the time

distribution of wind velocities.
For large wind velocities having a return period of more than two years the Gumbel
distribution can be adopted, which is based on the measured annual maximum wind
velocities.
For relatively small wind velocities the most appropriate approach to establish a
time distribution would be to evaluate the wind statistics from weather stations over
the year. Alternatively, this time distribution can be established based on a Weibull
distribution, the parameters of which can be derived from wind velocities having a
probability of occurrence of 1 or 2 years.
Often wind statistics refer only to the occurrence of maximum wind values without
referring to wind directions. However, only winds acting perpendicularly to the span
will cause the maximum swing angle. In order to take care of the variation of wind
intensity with its direction the time probability of the swing angles can be assumed as
being half of the corresponding wind velocities when the swing angle is more than 2°.

5.4.4 Part 3: Required Clearances

The required clearances should be derived from the electrical characteristics of the
transmission system. There are various approaches available to determine the clea-
rances necessary to withstand given electrical stresses.
According to IEC 60071–1 the standard highest system voltages, US, for equip-
ment are divided into two ranges:

• Range I: above 1 kV to 245 kV included,

• Range II: above 245 kV.

For lines within range I lightning (fast front) impulses, Uff, decide on the insulation
level and for lines within range II switching (slow front) impulses, Usf.
Standards IEC 60071–1 [5] and IEC 60071–2 [6] deal only with phase-to-earth
clearances Dpe. Since the insulation is self-restoring the lightning withstand voltages,
U90%_ff, are specified for the probability of 90 % of being withstood. The coefficients
of variation are assumed to be 3 % for lightning impulses and 6 % for switching
impulses.
230 E. Ghannoum

Cigré TB 72 [7] deals with the dielectric strength of external insulation and con-
ductors, lightning and switching impulses. The lightning impulses to be used are
those which can propagate beyond a few towers from the point of the lightning strike.
For the purpose of determining clearances this is to be taken as U50%_ff. The switching
impulse voltage to be used, Usf, is the highest that can occur on the lines (Table 5.28).
Table 5.29 lists the clearances under impulse voltages for some of the often used
voltage levels according to the IEC, Cigré and EN formulae given in detail in Cigré
TB 348.
In Table 5.29 the power frequency clearances, Dpe_pf and Dpp_pf, are given. The
comparison reveals that the procedure according to EN 50341–1 results in less con-
servative data than TB 72. No information is given in IEC 60071–2.

5.4.5 P
 art 4: Coordination of Conductor Positions and Electrical
Stresses

The study resulted in a proposal to consider at least two cases of designing tower top
geometry and mid span clearances, taken into account the varying position of con-
ductors and insulator sets under the wind action and the electrical stress of the air

Table 5.28  Clearances under impulse voltages

System voltage Impulse voltages IEC 60071-2 Cigré TB 72 EN 50341-1
US U90%_ff U50%_ff Usf Dpe Dpe Dpp Dpe Dpp
kV kV kV kV m m m m m
123 450 490 - 0.9 0.94 0.96 0.89 1.00
550 590 - 1.1 1.13 1.17 1.07 1.20
245 850 920 - 1.7 1.76 1.83 1.67 1.86
950 1030 - 1.9 1.97 2.05 1.87 2.07
1050 1140 - 2.4 2.18 2.26 2.05 2.29
420 1175 1270 850 2.4 2.40 2.70 2.25 2.67
1300 1410 950 2.9 2.80 3.20 2.64 3.15
1425 1540 1050 3.4 3.25 3.73 3.02 3.68
525 1300 1410 950 2.9 2.80 3.20 2.64 3.15
1425 1540 1050 3.4 3.25 3.73 3.02 3.68
1550 1680 1175 4.1 3.90 4.48 3.60 4.40

Table 5.29  Clearances under power frequency voltage

System voltage Cigré TB 72 EN 50341-1
US Dpe_pf Dpp_pf Dpe_pf Dpp_pf
kV m m m m
123 0.32 0.42 0.23 0.37
245 0.64 0.85 0.43 0.69
420 1.16 1.52 0.70 1.17
525 1.68 2.20 0.86 1.47
5  Structural and Mechanical Design 231

Table 5.30  Coordination of conductor positions and electrical stresses

Conductor and insulator positions
Electrical stresses of the air Still air or moderate High wind velocity
gap wind during 99 % of with T = 50 years
time
Probability High Low
Power frequency voltage High Covered by cases 1 & 2 Case 1
Lightning/switching impulse Low Case 2 Not considered
voltage

gap by power frequency voltage and overvoltages caused by lightning strokes or

switching operations (Table 5.30).

• Case 1 is described by conductor and insulator positions which would occur under
the action of the design wind velocities having a return period T of say 50 years.
These positions are combined with clearances required to withstand power fre-
quency voltage. The probability of flashovers depends on the probability of occur-
rence of the corresponding design wind velocity. For a wind velocity having a return
period of 50 years the probability of flashover will be 1 % per year at maximum.
• Case 2 is described by conductor and insulator positions which are assumed as
not being exceeded during 99 % of the time. These positions are combined with
clearances required to withstand the lightning or switching overvoltages with a
probability of more than 90 %. The flashover probability is that by which the
design overvoltage conditions will be exceeded.

5.4.6 Part 5: Application Example

The approach proposed to determine tower top geometry and mid span clearances is
demonstrated by means of an example for a suspension tower equipped with a twin
conductor bundle ACSR 564/72. The mean value of the yearly maximum wind velo-
cities is 20 m/s (10 min mean value, height 10 m), the coefficient of variation is 14 %.
The highest system voltage US is 420 kV and the switching impulse voltage Usf is
1050 kV. The mean suspension insulator length is 5,0 m, the sag 20 m and the dis-
tance between sub-conductors 0,4 m.
The design of the tower top is performed in 5 steps. The detailed calculation is
given in Cigré TB 348.

• The wind velocity with a return period of 50 years is calculated according to the
Gumbel distribution function for extreme values. The moderate wind velocity is
calculated using the Weibull distribution function;
• The wind velocities are adjusted to the time period of measurement and the
height above ground according to the IEC 60826 equations. In Tables 5.31 and
5.32 the results are given for a 5 minute wind speed 20 m above ground level;
232 E. Ghannoum

Table 5.31  Clearances between insulator set and earthed structure

Wind Insulator swing angle Clearance
Pos. Wind Wind Mean Standard Max. Impulse Power Distance
condition speed swing deviation swing voltage freq. to earth
angle angle IEC voltage
TB 72
V20 m ϕ σϕ ϕ + 2σϕ Dpe Dpe_pf L0
m/s ° ° ° m m m
1 T = 50 32.0 42.2 2.2 46.6 - 1.16 4.99
years
2 Swing 1 % 16.1 12.9 1.5 15.9 3.40 - 4.97
of year

Table 5.32  Clearances between adjacent conductors in mid span

Wind Conductor swing angle Clearance TB 72
Pos. Wind Wind Mean Standard Min. Max. Impulse Power Dist.
condition speed swing deviation swing swing voltage freq. cond.
angle angle angle volt.
V20 m ϕ σϕ ϕ − 2σϕ ϕ + 2σϕ Dpe Dpe_pf L1
m/s ° ° ° ° m m m
1 T = 50 32.0 43.7 2.2 39.3 48.1 - 1.52 -
years
2 Swing 16.1 13.5 1.5 10.5 16.5 3.73 - 6.70
1 % of
year

• The results for the swing angle of the suspension insulator set at the tower and
the swing angle of conductors at mid span are somewhat different;
• Further the scattering of the swing angle is determined:
• The maximum swing angle at the tower is obtained by adding two standard
deviations to the mean swing angle of the insulator. The maximum angles for
the two positions of the insulator set are depicted in Table 5.31, and Figure 5.6;
• For the adjacent conductors both the minimum and maximum swing angles
are obtained respectively by deducting and adding two standard deviations.
The most unfavourable positions of the swinging conductors are depicted in
Figure 5.6;
• Finally the minimum electrical clearances taken from Tables 5.31 and 5.32 are
used to determine the horizontal distance between:
• The attachment point of the first conductor and the tower structure (5.0 m);
• The attachment points of two adjacent conductors in a horizontal configura-
tion (6.7 m).

The numerical results of the calculations are summarized in Figure 5.5 and

Table 5.31 for the clearances between insulator set and earthed structure at tower
and in Table 5.32 and Figure 5.6 for clearances between adjacent conductors in mid
span. Cigré TB also compares these results for other alternatives.
5  Structural and Mechanical Design 233

Figure 5.5  Example of 5.00 m

clearances between
insulator set and earthed
structure. Earthed structure

Pos.1

5.00 m
(46.6°)
1.1
6

Pos.2
(15.9°)
3.4
0

0.4 m

5.00 m 6.70 m
earthed
structure
5.0 m

°
48.1 °
39.3

16.5°
20.0 m

10.5°
1.
52
°

3.73°

0.4 m 0.4 m

Figure 5.6  Example of clearances between adjacent conductors in mid span.

234 E. Ghannoum

5.4.7 Conclusion

On the basis of the review of national practices, the study results in a proposal to
consider two cases when designing the tower top geometry and mid span clearances
in view of conductor and insulator swing under wind action. Reference wind speeds
and ice loads as well as parameters, such as the number of standard deviations, have
to be selected accordingly based on studies and experience.

5.4.8 References

[1] Cigré SC22 WG06: Tower top geometry. TB 48 (1996)

[2] EN 50341–1: Overhead electrical lines exceeding AC 45 kV – Part 1: General requirements –
Common specifications (2002)
[3] IEC 60826: Design criteria of overhead transmission lines, Edition 3.0 (2003)
[4] Cigré SC22 WG06: Probabilistic design of overhead transmission lines. Cigré TB 178 (2001)
[5] IEC 60071–1: Insulation coordination – Part 1: Definitions, principles and rules
[6] IEC 60071–2: Insulation coordination – Part 2: Application guide
[7] Cigré SC C4: TB 72: Guidelines for the evaluation of the dielectric strength of external insula-
tion. Cigré (1992)
[8] Kiessling, F., Nefzger, P., Nolasco, J.F., Kaintzyk, U.: Overhead Power Lines – Planning, Design,
Construction. Springer, Berlin/Heidelberg/New York (2003)
[9] Cigré SC22 WG06: Tower top geometry and mid-span clearances. Technical Brochure No. 348,
June 2008

5.5 Load Control Devices

5.5.1 Summary

Electricity utilities need to design overhead lines to withstand potential cascade fai-
lures of towers, which could occur as the result of exceptional icing or rupture of
conductors. Some utilities simply design the lines with section towers at frequent
intervals and other utilities use special devices called «load control devices» (LCD),
the function of which is to limit the stresses applied to the towers in the case of the
application of exceptional longitudinal loads. These devices are usually based on
one of the following principles:

• Releasing of the conductor from the tower,

• Sliding of the conductor through suspension clamps so as to balance the load,
• Elongation of the conductor fitting, so as to increase the apparent length of the
conductor and to decrease its tension,
• Elastic deformation of the crossarms on the towers.

This Section based on Cigré study [1] describes the characteristics and performance
of the elongating fittings occasionally used in Japan and the controlled sliding clamps
used in France.
5  Structural and Mechanical Design 235

Some Japanese utilities have replaced conventional suspension sets by Suspension-

Tension sets with elongating fittings, on existing lines under which buildings appea-
red after construction. These devices have been designed by EPDC (Electric Power
Development Co, the research institute of Japanese Utilities) and make it possible to
raise the height of conductors (in normal service) without replacing or modifying the
existing suspension towers.
The first type of device is based on the unfolding of the fitting, triggered by the
rupture of calibrated shear pins; the loads are damped by an impact-absorbing ele-
ment. The second type of device is based on sliding of the shackle in a groove desi-
gned into the yoke plate.
In France, EDF systematically installs controlled sliding clamps on suspension
towers. Each tower of a line is designed for a conventional asymmetric ice load.
The sliding load of the clamp is chosen in a range of standard values such that if the
actual load from the asymmetric ice load reaches the calculated load then the con-
ductor will slide through the clamp. The suspension tower is thus protected from
failure from exceptional longitudinal loads and at the same time the tower cost is
not increased.
The results of a survey show very varied utilization of the LCDs in different coun-
tries: either not used at all, occasionally used or systematically used. The choice of
device is based on many reasons and are linked to different local practices and para­
meters:

• The countries which systematically use LCDs, such as France, Belgium and
Romania, give one or more of the following arguments:
• The global cost of a line is reduced,
• The efficiency of these devices has been checked in test stations and on seve-
ral occasions in the field.
• The countries which use these devices in a non systematic way, such as Japan, do
it in specific cases, like on existing lines to avoid the replacement of the towers
when buildings are built under them,
• The countries, which never use LCD, give one or several of the following argu­
ments:
–– Towers are systematically designed for specified longitudinal loads,
–– Icing is unknown or very rare in the country,
–– The efficiency of the devices is uncertain,
–– The increase of the conductor sag, generated by an LCD, is unacceptable in
the view of public safety,
–– The use of controlled sliding clamps causes severe damage to ACSR conductors.

In conclusion, the utilities that want to go further in the field of load control devices,
should concentrate on the following aspects: reliability, safety, and estimation of the
cost effectiveness. With respect to this last aspect, the experience in some countries
show that, for the future, the modifications and upgrading of existing lines are a
potential field of application for LCDs.
236 E. Ghannoum

5.5.2 Introduction

The most serious overhead line failure, which could have severe consequences affec-
ting the grid operation, is the cascade failure of several transmission towers.
This cascade failure is usually initiated by the failure of a single tower or by the
breaking of conductors and is the result of insufficient design strength of towers to
withstand high longitudinal loads.
The choice of solution to avoid such damage will depend upon a compromise
between the risk and consequences of a cascade failure and the extra cost of const-
ruction specified by the designer.
The most frequent solutions to limit or avoid series tower failures are, either to
use section supports at regular intervals (also called anti-cascading towers), or to
provide enough longitudinal strength to the suspension towers. In the second solu-
tion, the use of LCD can make it possible to limit considerably the magnitude of the
longitudinal loads and thus the cost of the towers.
After a general presentation of the different categories of LCD, this Section
describes some devices which are currently available. It then summarizes the experi-
ence of the countries that took part in the survey conducted as part of SC22 WG 06
of Cigré.

5.5.3 Classification of LCD

The task assigned to an LCD is to avoid major and costly line failures, which could
occur in the case of certain exceptional loads, by releasing or limiting the stresses
applied to a main line-component.
Practically, LCD are generally designed to protect the towers from exceptional
stresses applied by the conductor in the case of unusual icing, cascading, etc.
Considering the existing devices, two categories of Load Control Device can be
defined:

5.5.3.1  Load Release Devices

In this case, the conductor load that could cause the failure of towers is totally relea-
sed from the tower so that the conductor system is not controlled (or there is a very
limited control). Different types of devices exist:

• Devices in which the conductor grip within the tower attachment fitting is relea-
sed, such as release type suspension clamps (there is no more control of the lon-
gitudinal tension),
• Devices in which the conductor or the fitting itself become detached from the
tower, by means of mechanical fuses such as shear pins or safety bolts, associated
with a system (such as a loose jumper) to prevent the conductor from falling down
to the ground.
5  Structural and Mechanical Design 237

5.5.3.2  Load Reduction Devices

In this case, the conductor load that could cause tower failure is reduced to an accep-
table value and the system maintains control of the conductors. Different types of
devices exist:

• Devices acting on the fastening of the conductor within the fitting, such as cont-
rolled sliding clamps, in which the conductor slides when the longitudinal load
exceeds a specified value,
• Devices which act on the length of the fittings (unfolding or sliding elements) so
as to increase the apparent length of the conductor.
• Devices which enable elastic deformation or swinging of the crossarms of the
tower to absorb longitudinal load.

Dynamic damping devices constitute another category of control equipment. These

act on the origin of the stress and are designed to stop or reduce dynamic phenomena
such as conductor vibration and galloping: damping jumpers, Stockbridge dampers,
spacer-dampers etc. These devices cannot really be considered as LCD and these
will not be addressed further.

5.5.4 Technical Data Related to Available LCD

This chapter gives the principles and characteristics of some important load reduc-
tion devices of two types:

• Special elongating fittings

• Controlled sliding clamps.

5.5.4.1  Special Elongating Fittings: Japanese Experience

These devices have been designed by EPDC and are mainly used in Japan. This is
mainly due to the increasing restrictions to the building of new lines and the increa-
sing urbanisation around the existing lines. In Japan new buildings are often built
under existing lines and Japanese utilities must find solutions to compensate for the
lack of height of the conductors from the ground and buildings, if possible without
replacing or raising towers.
One solution is to replace the conventional suspension systems by Suspension-­
Tension Insulator Assemblies. These devices can also be used to improve the insu-
lation level of the lines, by increasing the number of insulators without lowering the
height of conductors. In these cases, the existing suspension insulator set is replaced
by two tension sets suspended under a special suspension fitting, which in normal
service is shorter than the conventional suspension set.
In order not to degrade the mechanical behavior of the existing towers in the case
of conductor breakage in one span or unbalanced tension (unusual ice accretion or
238 E. Ghannoum

ice falling off), these devices have been designed as load reduction devices: an
excessive differential tension across the tower elongates the fitting and increases the
apparent conductor length so as to reduce its tension.
Two types of these fittings have been developed in Japan.

Type I
Principle
The elongation of the TYPE I device is triggered by the breaking of shear pins
and is damped by the deformation of an impact load absorber.
The following figures illustrate the functioning of the system in the case of con-
ductor breaking in one span. 1) Situation in normal service (Figure 5.7). In this nor-
mal situation, no loads are applied to the shear pins or the impact load absorber.
2) Intermediate situation (Figure 5.8).
The metal suspension fitting moves to the right, due to the breaking of the con-
ductor in the span to the left. If the dynamic load is too high for the shear pins, they
break, which leads to the following situation.
3) Final situation (Figure 5.9).
The dynamic effects of conductor breaking are damped by deformation of the
impact load absorber. In the final situation, the conductor tension is reduced by
increasing the length of the span.

Design Characteristics

• The rupture strength of the shear pins is 40 to 60 % of the maximum working
tension of the conductor.
• The permissible load of the shear pins against unbalanced conductor tension (ice
drop, strong wind) is 40 % of the ultimate tensile strength of the conductor.
• Rupture of the conductor generates a sudden extension of the fitting, which cre-
ates a large impact load. The impact load absorber is designed to reduce this load
to the same level as that existing in conventional suspension sets.
• The impact load absorber is designed to control loads up to 1.5 times the maxi-
mum working tension of the conductor.

Mechanical Performance
The performance of the device was checked on an actual scale test line.

• Conductor breakage:
–– The final tension of the conductor in the opposite (intact) span is less than
60 % of the maximum working tension (with the worst atmospheric loads).
–– The impact load is less than with the conventional suspension sets, because of
the impact load absorber.
• Unbalanced conductor tension (ice drop, unusual ice accretion, strong wind):
–– The unbalanced tension applied to the tower is less than 25 % of the maximum
tension of the conductor in the case of ice drop or strong wind.
–– The special suspension hardware of the device is shorter than the conventio-
nal suspension set, so that the dynamic unbalanced tension and the swing
5  Structural and Mechanical Design 239

Impact load absorber

shear pins

Figure 5.7  Initial geometry of an LCD type.

Figure 5.8 Intermediate
position: after the
conductor breaks but
before the rupture of the
shear pins.

Figure 5.9 Final
equilibrium position of the
LCD after trigger.

angle are larger. The tests showed the need to consider wear design aspects
considering the longitudinal swing of the device.
• Wear performance
–– The wear calculations, assuming an expected life of 50 years, showed no
problem.
Electrical Performance
The tests showed that the device could withstand a power arc equal to or higher
than that withstood with conventional tension sets.
240 E. Ghannoum

Service Experience
At the present time, there are around 1245 Suspension-Tension Insulator Assemblies
of this type installed, of which:

• 170 for 66 kV lines

• 875 for 154 kV lines
• 200 for 275 kV lines.

No data was given about the efficiency of these devices in real faults.

Type II
The elongation of the so-called “tension balance type” device occurs by sliding of
the shackle in a groove designed into the yoke-plate; the overall elongation is shorter
than with the TYPE I device, and consequently the increase in conductor sag is less.
Figure 5.10 illustrates the system functioning in the case of a conductor system
breakage in one span, compared with a conventional device. The principle is worse
for unbalanced tension due to ice drop or unusual ice accretion.

Note: After failure of the conductor in one span, the device enables the
conductor tension in the opposite(intact) span to be reduced to 60 % of the
maximum working tension in the worst atmospheric loads.

Figure 5.10  Type II Tension Balance LCD.

5  Structural and Mechanical Design 241

There are limitations for the use of such devices; they are not suitable in the case
of high tension differences between spans (which might cause inopportune sliding
or no sliding at all) and their efficiency in reducing the tension decreases with the
number of successive suspension spans in a section.
It is worth mentioning the following elongating device, which is currently under
consideration for use on tension towers in Japan.

5.5.4.2  Controlled Sliding Clamps

Controlled sliding clamps are used on a large scale in some countries, such as Belgium,
Romania and France.

French Experience
In France, controlled sliding clamps have been used systematically for more than 45
years on all transmission lines, both for the phase conductors and earthwires. During
this time, the technology of these devices has continuously evolved, in order to
improve their performance, and to optimise their use and adapt them for new con-
ductors (Fig. 5.11).

Principle of the Controlled Sliding Clamps

A controlled sliding clamp is made of a body in which rests the conductor and of a
cover (see Figure 5.12). This cover is installed over the conductor by means of a system
the pressure of which can be adjusted to make it possible to carry out the calibration of

Figure 5.11  Elongating LCD type.

Figure 5.12 Controlled
sliding clamp.
242 E. Ghannoum

the load that causes the conductor to slide through the clamp. The first types of clamps
used torsion bars and were called “elastic tightening clamps”. EDF finally abandoned
the use of torsion bars in 1980, in favour of clamps exclusively based on the controlled
tightening of the bolts; these clamps are simply called “controlled tightening clamps”;
they enable a more precise adjustment and allow better performance.

Calibration
The conductor must slide within the clamp at a specified longitudinal load, FS,
that may occur as a result either of a conductor rupture or an exceptional unbalanced
ice load.
FS = longitudinal load to be applied to the conductor for sliding
= Sliding load due to the tightening of the conductor in the clamp
+ Sliding load due to the vertical load applied to the clamp by the conductor
= (f1 + f2). FC + f1 .FV
f1 = friction coefficient between clamp body and conductor
f2 = friction coefficient between clamp cover and conductor
FC = compressive load of the clamp cover
FV = vertical load applied to the body (in the general case, this load is taken
= V 2 + H 2 , to take into account the transverse load H)
FS = longitudinal load at which sliding occurs
Each tower of a line is designed to resist asymmetric icing. EDF conventionally
considers an ice sleeve of n cm (radial thickness) on all spans on one side of the tower
and (n-2) cm on all spans on the other side; the value of n is 2,4 or 6, depending on the
area, the ice density is taken 0.6. The loads are calculated while taking into account
the deflection of the insulator sets. Table 5.33 gives some examples of application.
The sliding calibration of the suspension clamp is chosen from the following
range of standard support loads: 8, 14, 20, 30 kN. For each tower, it is chosen as the
value immediately higher to L-0.3 V 2 + H 2 , where 0.3 is the assumed value of the
friction coefficients, V, H and L are the vertical, transverse and longitudinal loads
applied to the conductor in the clamp, in the asymmetric icing assumption. Support
loads greater than 30 kN require other solutions such as the use of clamps in series,
the use of suspension- tension sets or the use of a tension tower.
Design of a Controlled Sliding Clamp
In order to make sure that the system operates correctly, the profile of the clamp
body must be determined in a precise manner.
It is necessary that the conductor pressure be equally spread out over the useful
length of the body to ensure a regular sliding on the basis of a longitudinal load deter-
mined by a precise adjustment. The length of the body is determined experimentally
in such a way that the compression stresses do not exceed the maximum pressure
allowable at the point of contact among the strands.
It is also necessary to calculate the radius of curvature of the clamp body so that
the bending stresses within a strand remain acceptable.
A computer model, presented in Figure 5.13, was developed to calculate the opti-
mal longitudinal profile of a clamp body.
Table 5.33  Order of magnitude of the loads applied to the supports with unequal ice loading
Mechanical tension in
the conductor (daN) Longitudinal stress (daN)
Tangent
Conductors Ice overload tower
Type External Weight (daN/m) Ultimate load (cm) End End With Section 225 kV (1) 400 kV (2)
diameter (mm) (daN) with ice ice (n-2) Tower
n cm cm
5  Structural and Mechanical Design

Conductor (4)
ASTER 570 31 1.57 18 400 2 7 800 4 420 3 380 1 200 900
PETUNIA 612 32 2.24 32 700 4 14 000 9 500 4 500 2 000 1 600
Earthwires (5)
PHLOX 116.2 14 0.63 10 800 2 4 150 1 900 2 250 1 600
(3)
PHLOX 228 20 1.24 21 200 4 10 700 6 800 3 900 2 000
Notes applicable to Table 5.33
(1) Length of the suspension set: 2.30 m for 225 kV
(2) Length of the suspension set: 3.45 m for 400 kV
(3) Conventional value (span < 1 000 m)
(4) All aluminum alloy conductors
(5) Aluminum alloy steel reinforced earth wires
Catenary constant: 2 200 m at 45 ° C
Length of the adjacent spans: 550 m
243
244 E. Ghannoum

Ultimate pressure between two

layers by point of contact
The layers slide Specialization of the
Self pressing tensions
sliding clamp
Reaction of the clamp body
Pressing force of the cover

Static strain induced in Minimal length of the clamp Ultimate load on the
wire near clamp body suspension clamp

Calculation:
Permanent and extreme
- of the curvature of the cable loads on suspension clamps
- of the strains induced in the wires Low temperature and
storm assumption

Drawing of the longitudinal

Radius of curvature
of the clamp body
Induced strains in the wires for
Length of the clamp
the various loading assumptions

Figure 5.13  Research of the optimal longitudinal profile of a clamp body.

Test
The accuracy of prediction of the load to cause conductor sliding depends on seve-
ral factors: frictional coefficients, clamp and conductor manufacturing tolerances, etc.
It is necessary to reduce the dispersion as much as possible. Nevertheless, the
precision on predicting loads that cause sliding, with respect to the clamp calibra-
tion load, cannot be better than 15 %.
To verify the sliding characteristics of the suspension clamps, tests are carried
out with each new type of suspension clamp:
Measurement of Friction Coefficient Test Conditions
This test is performed with an adjustable sagging machine. The conductor rests
on the body of the suspension clamp without the cover. The winding angle is bet-
ween 20 and 30 degrees and the vertical load is given by Table 5.34.
The measurements of the sliding efforts are performed over a length of one
meter.
Test Requirements
The friction coefficient shall not exceed 0.35.
Measurement of the Sliding Threshold Forces Test Conditions

• Measurements of sliding threshold values are made according to the cover pressure.
• The conductor is loaded at 20 % of its ultimate tensile strength in a suspension
clamp with a winding angle of 15 degrees (in order to settle each wire in position).
• The cover is tightened to the value given by the manufacturer and the conductor
is released.
5  Structural and Mechanical Design 245

Table 5.34  Vertical loads for sliding tests according to the type of suspension clamp
Suspension clamp class Conductor Vertical Load (kN)
A ASTER 75 8,5
B PHLOX 75,5 17
C PHLOX 147 24
D ASTER 570 40
E PETUNIA 612 45
F POLYGONUM 1185 45
G POLYGONUM 1185 90

Sliding load
(kN)

NON AUTHORIZED ZONE

A
AUTHORIZED ZONE I

B AUTHORIZED ZONE II

NON AUTHORIZED ZONE

d (length of sliding)
MAXI 1000
10

Progressive
loading

Figure 5.14  Design data for sliding clamps.

• The suspension clamp is then pulled over a length of one meter.

• The sliding force is the maximum tension applied to the conductor on the stres-
sed side (the other being released).

Test Requirements
The value of the sliding force relative to the sliding length must remain within the
allowable areas such as defined in Figure 5.14; the force must range between the
values A and B following a displacement not exceeding 1 cm. The actual curve must
246 E. Ghannoum

be shifted by a value C so as to take account of the friction due to the vertical load
according to the formula C = 0.05× UTS.
The breaking load of the conductor following sliding must remain higher than
95 % of the ultimate tensile stress (UTS) of new conductor for homogeneous con-
ductors, and 90 % of the UTS for bimetal conductors.

Device Efficiency
Sliding Limitation
When sliding occurs it is stopped whenever the tension difference applied to the
clamp on either side of the clamp falls below the calibration value.
Thus the sliding length depends mainly on the initial tension difference applied
to the clamp, on the relaxation due to the deflection angle of the insulator set and
hence on the length of this set.
The conductor sliding range observed in the overhead line network, in several cases
of unequal loading, ranged from 1 to 2 m. For earthwires this value is much higher.
In case of a breakage of a conductor or a support, considerable sliding, of ten
meters and more, were noted on towers located on either side of the defect.
On the other hand, the movement declines very rapidly on subsequent towers.
Finally, in some cases, in particular when sliding is considerable, the conductor
can be damaged (rupture of strands, bundling and bird cage formation); this pheno-
menon however is very rare on conductors having an outside layer made of alumi-
nium alloy (AAAC and AACSR conductors).
However, this inconvenience which requires the replacement of a length of conduc-
tor, is in all situations less serious than the destruction or damage of several support
structures. The line downtime is reduced and the repair costs are considerably lower.
Action of Clamps During Real Faults
Among the incidents, whose consequences were limited by the action of sliding
clamps, we can give the following examples (Table 5.35):

• In 1949, a 40 degree angle tower fitted with suspension sets and located in the
mountains, with a high level difference, was carried away by an avalanche. Two
of the three conductors were broken; the conductors slid in the clamps of several
suspension towers. Only one crossarm was distorted.
• In 1964, after the rupture of the top of a 50 degree angle tower of a 400 kV line
within the Paris area, damage was limited to 4 towers, as a result of sliding of the
6 conductors within their suspension clamps.

Table 5.35  Example of values used for the design of sliding clamp
RANGE (kN) A (kN) B (kN) C (kN)
8 8 6.5 3.6
14 14 11 6.3
20 20 15.5 8.9
30 30 23 13.3
NOTE: C = (A + B)/4
5  Structural and Mechanical Design 247

• In 1971, an overall and persisting icing phenomenon, unusual within the area of
Le Creusot, gave rise to an ice sleeve formation on the phase wires of a 400 kV
line, with linear weight up to 15 kg. This overload was approximately 4 times
higher than that considered in the line design calculation. The complete failure of
the conductor bundle was prevented by means of the sliding of the conductors
within the clamps of 5 successive towers. Only 2 of them were damaged.
• In 1974, an aircraft hit 2 of the 6 twin conductor bundles made of AAAC, of a
400 kV line. The 4 conductors slid in their clamps. The 2 suspension towers
supporting the damaged span, had to be replaced, as well as some elements of the
two adjacent tangent towers.
• In 1988, an aircraft hit the 3 twin bundles of a 400 kV line near a substation. The
broken conductors slid in the clamps over more than 3 km. 2 suspension towers
had to be replaced.

Belgian Experience
Since 1971, controlled sliding clamps have been systematically installed on the
conductors of 380 kV lines.
The first aim was to reduce the weight and cost of tangent towers, by reducing
the accidental stresses considered in the design of the towers with the assumption of
conductor breakage (the effect of breaking all the conductors of the bundle).
The torsion moments have traditionally been calculated by application of the
maximum tension of the conductors of the phase conductor on one side of the cros-
sarm, without taking into account the dynamic and releasing effect due to the rup-
ture of the phase conductors and the swinging of the insulator sets. This method, if
roughly applied on 380 kV lines (equipped with twin bundles), would have led to
un-economic tangent towers. The use of controlled sliding clamps was estimated to
offer a reduction in the weight of tangent towers of 6 %.
These arguments do not apply to earthwires, because of the lower stresses and
the lower potential load reduction and consequently sliding clamps are not installed
on earthwires.
Two types of controlled sliding clamps are used. One type uses a torsion bar bol-
ted in the body of the clamp and calibrated shear pins in the cover. The second type
is based on the controlled tightening of the bolts.
In the lines, the conductors are systematically protected by armour-rods; the spe-
cified sliding load is determined as a function of the conductor type and is taken as
a relatively high value to avoid inopportune sliding. No sliding in clamps has been
seen in Belgium.

Romanian Experience
For the past 30 years, controlled sliding clamps have been used for the conductors
of all transmission lines in Romania. These clamps are based on the controlled tigh-
tening of the bolts by a limited torque wrench. Their actual efficiency has not been
measured on the network but no failure due to cascading has occurred, although most
supports are guyed poles. RENEL is still working on the optimisation of the control-
led sliding clamps.
248 E. Ghannoum

Release type suspension clamps have been used, but very occasionally and only
on lines without problems of conductor galloping. Their benefit has been demons-
trated on several occasions.

5.5.4.3  Service Experience of LCD in Different Countries

The summary of the answers of the countries replying to the survey is given in
Table 5.36.

5.5.5 Specification for an Ideal LCD

Load Control Devices (LCD) are devices designed and used to limit the damage to
structures in case of a failure that occurs in a transmission line.
In a previous WG B2.06 report, published in Electra No. 193 [1] and Cigré TB
174 [4], a survey of existing LCD’s installed on overhead lines was performed,
including comments on their use in each country. This report is summarized in the
chapter above. In the Electra report, a distinction was made between two types of
devices: those intended to reduce loads applicable to structures and those comple-
tely releasing the conductors from structures. However, in this report the wording
LCD is used to describe both types of devices indiscriminately.
The catastrophic line failures in 1998 (Canada) and 1999 (Europe) raised new
interest about LCD’s as well as questions about the performance of existing ones.
Consequently, WG B2.06 decided to prepare a basic generic specification or a
list of properties of an ideal device that would be helpful to operators and designers
of overhead lines. This work should help current and potential users of LCD’s to
check the performance of the devices they currently use or intend to use, and to
promote the development of better devices by manufacturers that have a minimum
number of non-compliances to the properties of the ideal device.
This report does not intend to compare and assess the performances of LCD’s
with other possibilities to limit failure containment, such as anti-cascading towers,
strength coordination, etc.

5.5.5.1  Criteria for the Evaluation of a Load Control Device

An ideal Load Control Device should be judged by many design and performance
criteria. Tentatively, the following criteria are considered and will be further discus-
sed in this report.

• Compliance with the primary functions of Load Control Devices

• Tolerance of triggering action
• Design flexibility to fit the as-built support strength
• Safety issues on clearances
• Impact on construction and maintenance.

It can be argued that these criteria can be more theoretical than practical, but one
purpose of this report is to alert developers and users of LCD’s on the importance of
5  Structural and Mechanical Design 249

Table 5.36  Replies of various countries to the questionnaire on LCDs

COUNTRY/COMPANY USE OF LCD/COMMENTS
ARGENTINA Not used
AUSTRALIA Not used
Conductors are generally ACSR
AUSTRIA/TKW Sporadic use in the past: only on a few towers on
lines built before 1940
BELGIUM/TRACTEBEL Systematic use for conductors of lines equipped
with bundles;
see paragraph Belgian Experience
BRAZIL/CEMIG Sporadic use
CANADA/ONTARIO-HYDRO Not used
HYDRO-QUEBEC
Sporadic use:
1) Controlled sliding clamps were installed for the
reconstruction of 735 kV lines, after heavy icing
events. No other incident has occurred since then,
so the efficiency of this equipment has not been
verified.
2) After erection, safety bolts for the attachment of
the construction cable are used on « chainette
towers ». A loose jumper is installed to retain the
construction cable from falling.
CZECH REPUBLIC/ENERGOVOD Not used
FINLAND/IVO Not used
FRANCE/EDF Systematic use on all transmission lines;
see paragraph French Experience
GERMANY/BAYERNWERK Sporadic use in the past:
Now, the German utilities consider that
1) the reliability of LCD and their effectiveness has
not been adequately demonstrated
2) the increase of the conductor sag is
unacceptable, in view of public safety, because of
the high density of population in Germany.
GREAT BRITAIN/NGC Not used
INDIA Sporadic use
ITALY/ENEL Not used
1) All towers are designed to withstand a specified
longitudinal load.
2) ENEL utilises ACSR conductors and argues that
the wires are considerably damaged after sliding in
the clamp.
3) ENEL is not convinced of the LCD efficiency,
especially in mountainous areas.
JAPAN/EPDC Sporadic use; development in progress;
Raising the conductors on existing lines under
which buildings appeared
KOREA/KEPCO Not used
(continued)
250 E. Ghannoum

Table 5.36 (continued)
COUNTRY/COMPANY USE OF LCD/COMMENTS
NETHERLANDS/KEMA Not used
NEW ZEALAND Not used
NORWAY/STATNETT Sporadic use
When the sagging of the earthwire is considered
less serious than a mechanical damage, sliding
clamps are used for earthwires. The clamps are
basically adjusted for a 30 kN sliding load. The
tower tops are designed accordingly.
PORTUGAL/EDP Not used
1) all towers (suspension or tension towers) are
designed to withstand, in the worst conditions and
with a certain safety coefficient, the rupture of a
conductor or an earthwire
2) zones with potential icing are very rare in
Portugal and avoided whenever possible.
ROMANIA/RENEL Systematic use for conductors; development in
progress;
see paragraph Romanian Experience
SPAIN/RED ELECTRICA Not used
SWITZERLAND/EOS Sporadic use;
In the case of double tension sets used for twin
bundles, EOS uses special yoke plates to damp the
dynamic effect of the rupture of the set.
USA/LINDSEY MANUFACTING Sporadic use in the past;
COMPANY since early 1970’s, two LCD systems have been
used in conjunction with porcelain line post
insulators throughout North America : shear pins
and elastic–plastic base. These devices were a type
of mechanical fuses.

these criteria in any new development and promote their use as a rating method and/
or a checklist to compare the potential performance of these devices.

5.5.5.2  Primary Function of Load Control Devices

This issue is one of the most important problems preventing or limiting the use of
Load Control Devices. Some Load Control Devices reviewed may cause more harm
when they trigger than if they had not. In order to prevent the consequential damage
due to the trigger of the Load Control Device, the following guidelines are establis-
hed and explained:

• The release of an LCD in a given support should not lead to unbalanced forces
which could cause the failure of the support itself or neighboring supports

This is an obvious requirement, otherwise the purpose of introducing the LCD in the
line would be defeated.
5  Structural and Mechanical Design 251

LCD’s are likely to trigger in random sequences5 on individual phases depending

on the load level and the mechanical characteristics of these devices. There is no
doubt that in such cases, there will be some imbalance in the loads applied to struc-
tures and these should be kept to a strict minimum.
Let us consider an LCD set to release a suspension or a dead-end string once a
given vertical or longitudinal load is reached. If this LCD triggers, the loads on the
structure itself and adjacent structures will become unbalanced and may cause
structural failures. For example, a flat configuration suspension tower designed to
support 100 % of a vertical ice load in all three phases, may break under a combina-
tion of loads such as 90 %-0 %-90 % (central phase is released) of vertical loads or
0 %-90 %-90 % (left phase released). Furthermore, these releases will impose a sig-
nificant increase in the longitudinal and vertical6 loads on the adjacent structures at
the released phase and could cause their failure.

• The trigger of the LCD should cause the least amounts of dynamic loads

The trigger of an LCD can cause sudden changes in loads applied to structures and
inevitable dynamic load effects. Designers of such devices should try to minimize
dynamic loads that could have serious consequences on adjacent structures. This
consideration is particularly important for all devices that completely release con-
ductor loads from structures. Devices of the slip clamp type and dead-end string
extensions create less dynamic loads than the previous devices.

• Load Control Devices Installed on a Line must be the least exposed to a cascade
of Load Control Devices

In some cases, LCD’s can fail in a cascading mode if one triggers at high load levels.
Users and manufacturers of LCD’s should develop some mitigation measures to pre-
vent such events from occurring. Such measures may include the use of anti-casca-
ding structures or limiting the number of LCD’s in a series of suspension towers.

• The static forces following the release of a Load Control Device should be
minimized

Users and developers of LCD’s should analyze the structural behavior of the trans-
mission line in the eventuality of their trigger action. Consequently, there is a need
to develop special structure loading cases when LCD’s are used in order to check
the adequacy of tower strength to withstand the static loads resulting from a trigge-
ring of an LCD.
This report does not include requirements regarding the deterioration of the device
and/or the conductor and regarding the performance under a variety of climatic con-
ditions, e.g. if the device is frozen solid with ice.

 In practice, LCD’s will likely not trigger at the same load even if they are of the same type.
5

 In the case of separation of a suspension insulator string.

6
252 E. Ghannoum

5.5.5.3  Tolerance (Variability) of Triggering Action

Many devices are set to trigger at a certain load or stress level fixed in relation to the
line component this device is supposed to protect. Unfortunately, the triggering load
in almost all devices reviewed is quite variable and time dependent, thus creating
difficulties when the user specifies the range of loads that would trigger their action.
For example one of the most used Load Control Device’s is the slip clamp where
the conductor is supposed to slip in the clamp if the resulting longitudinal load
exceeds a certain threshold. Tests performed on slip clamps show not only a large
dispersion in slip clamp trigger, but also a variability of this threshold with time.
For example, let us assume that the longitudinal strength of one tower crossarm is
constant and equal to 20 kN and this crossarm is equipped with a slip clamp designed
to slip before any longitudinal load causes the tower or its crossarm to fail. If the
clamp threshold is set at 19.5 kN, and assuming a coefficient of variation (COV) of
20 % of this LCD, which is not uncommon, about 50 % of the clamps will not trigger
prior to the tower failure. If we set the threshold at 15 kN, we secure that about 95 %
of the clamps will trigger below 20 kN, some of them at a very low level of about 10
kN. In such circumstances, the tower would have an extra longitudinal capacity, paid
for dearly, but unused because of the slip clamp. There could also be cases where the
tower would have been able to withstand the longitudinal load, but the trigger of the
slip clamp at low load levels would have already created a permanent fault.
The dispersion problem of the LCD is further compounded by the unavoidable
dispersion of the strength of the component it intends to protect. For example the
above strength of the crossarm, assumed constant at 20 kN, does not represent rea-
lity. Based on data from tower tests, the dispersion of strength of lattice towers is
accepted to be in the range of 5 to 10 %. The combination of both dispersions of the
component and the LCD increases problems of specifying a trigger load of the LCD.
Consequently the following guidelines are established as regards the tolerance of
the triggering actions:

• The mechanical Load Control Device should have small strength dispersion
(COV ≈ 2-3 %). In order to grasp the importance of this requirement, let us consi-
der the same example above and assume an average strength of the crossarm of 20
kN and a COV of 10 %. The lower 10 % exclusion limit of the crossarm strength
is equal to (1–1.28 *10 %) * 20 kN = 17.4 kN. If it is required to ensure that in 90 %
of the cases the LCD triggers before the tower, the trigger threshold will vary with
the strength dispersion of this device. The strength factor of Table 5.23 in IEC
60826 [2] and in Cigré TB 178 [4] can be used to calculate the required strength
of the LCD as a function of the strength dispersion of both the LCD and the com-
ponent it needs to protect. If the strength dispersion of the LCD is 5 %, then we
should set its average trigger threshold to about 20/1.15 = 17.3 kN. If the LCD
dispersion is 20 %, the average trigger will be about 20/1.37 = 14.6 kN. This value
is further reduced to 12.2 kN for a dispersion of 30 %. These numerical examples
confirm that when the LCD dispersion increases, the trigger of the LCD has to be
set a low value compared to the strength of the tower thus preventing in some
cases the use of an available and expensive tower strength.
• The threshold of release of the Load Control Device should vary very little in
time or with the mechanical loads to which the line is subjected.
5  Structural and Mechanical Design 253

5.5.5.4  Design Flexibility

During the design of a Load Control Device, it is important that the strength of the
component that this device is protecting is known. For example, if a Load Control
Device is supposed to disconnect the conductor from the support, the vertical capa-
city of any support of the line needs to be known. This capacity is not constant, but
very dependent on the actual weight span. It would not be appropriate to specify the
threshold of the Load Control Device in relation to the maximum spans; because very
few supports are normally used with the maximum design spans.
For example, assume a tangent support is designed for a weight span of 900 m
supporting a 20 N/m conductor with a limit ice load of 80 N/m. The support should
be designed to support [900 m * (20 + 80) N/m] = 90 kN of vertical load if the insu-
lator string weight is neglected. If this support is used with a 450 m weight span in
the line, it will be able to withstand an ice load of [(90.000 N/450 m) − 20 N/m] = 1
80 N/m. It is obvious that the Load Control Device should trigger before the ice load
reaches about 80–90 N/m otherwise, the tension in the conductor can increase sub-
stantially and cause other problems such as to angle towers.
The above examples raise the issue of the necessity of modeling the Load Control
Device to fit the as-built support strength, as long as the latter is known.
Consequently, the threshold of the release of the Load Control Device should be
adjustable in order to harmonize it with the strength of the line component to be
protected, if variable.

5.5.5.5  Safety Issues on Clearances

In some cases, successful triggering of a Load Control Device might cause a safety
problem and potential injury to people. For example, after the slippage of the conduc-
tor in the suspension clamp or a disconnection in the suspension clamp of the suspen-
sion insulator string, the conductor sag will increase and may pose a safety hazard,
especially if the line is still in operation. This impact on safety would have been the
same as a tower failure if the conductor slips in the clamp at a load level close to the
tower strength (i.e., if the trigger threshold is equal to the tower strength). However,
since perfect fine tuning between the trigger of the LCD and tower strength, due
mostly to the dispersion in both values, is not currently possible, the early (prema-
ture) trigger of an LCD compared to the strength of the tower should be of concern
to the users.7
Besides, the Load Control Device should have some kind of indicator that shows,
at a distance, that this Load Control Device has triggered and needs a human
intervention.
Consequently the safety guidelines are as follows:

• The release of a Load Control Device should not permit clearances to ground and
objects to infringe minimum safety distances when the line is still live or when
such reduced clearance can lead to risk of injury to people, except when specifi-
cally allowed by applicable standards and safety codes,

7
 In the example above, imagine an LCD set to trigger at 12–15 kN while the tower has a strength
capacity of 20 kN. If triggered at such low percentage of the tower strength, the LCD could cause
safety hazards.
254 E. Ghannoum

• The Load Control Device should have visual indicators that confirm the trigge-
ring of the device, unless such action is obvious to the observer.

5.5.5.6  Impact on Construction and Maintenance

Construction and maintenance personnel will not likely work on lines equipped
with release mechanism devices (except may be in the case of slip clamps) unless
the release mechanism is disabled. Furthermore, the replacement of Load Control
Devices should be made at the least possible cost.

5.5.5.7  Conclusion
At the time of writing all above generic specifications and properties of an ideal LCD
are not currently fully met by any device currently used or available in the market.
Despite this situation, owners of transmission systems may decide, and have deci-
ded, to adopt some devices because they feel or they have found through experience
that their advantages outweigh their inconveniences.
At a minimum, the specifications given in this document can be used to compare
and select the best device, or the one that has the minimum number of non-­compliance
to these specifications.
In addition to the above, developers of new devices will find guidance in this docu-
ment if they want to develop and validate better performing LCD’s.

5.5.5.8  References
[1] Load Control Devices on Overhead Transmission Lines – Electra No. 193, December 2000 –
Cigré WG B2.06
[2] Design Criteria of Overhead Transmission Lines – IEC 60826, Ed. 3.0, October 2003
[3] Probabilistic Design of Overhead Transmission Lines – Cigré TB 178, February 2001 – Cigré
WG B2.06
[4] Load Control Devices on Overhead Transmission Lines, Cigré Brochure 174, December 2000

5.6  echanical Security of Overhead Lines Containing

M
Cascading Failures and Mitigating their Effects

5.6.1 General

Overhead line cascades started to occur soon after the first lines were erected in the
first part of the 20th century, and they continue to occur. Why?
The last 30 years have brought many improvements in overhead line design
practice, more specifically with the introduction of the reliability-based design appro-
ach. But even with careful consideration of the various sources of uncertainty and
variability in the loads and resistances, failures will still occur. The issue is no longer
to assess whether “IF” it will happen but rather to address the “WHAT IF” and assess
and limit the impacts of the consequences of cascading failures. With the difficulty in
the prediction of extreme climatic loads, especially those loads associated with big
storm events of all kinds, the utilities are searching for effective ways to improve the
continuity of service of their power transmission grid.
Containing line cascades is a high priority for most utilities, especially for those
that had been hardly hit by huge accidents in the past. The main objective of Cigré
5  Structural and Mechanical Design 255

TB 515 [11] is to help line designers and planners to achieve this challenging goal
through a better understanding of line cascading phenomena. This understanding is
essential for effective and sustainable mitigation, which must be accounted for in
the general context of power grid planning and operations rather than be restricted
to the mechanical line design process.
TB 515 [11] summarizes the most up-to-date knowledge gained by Industry on
what causes cascading failures of overhead line supports and how to increase the
mechanical robustness and security of transmission lines. Based on scientific rese-
arch and the experience of several Cigré utility members with catastrophic cascading
failures during big storm events over the last 15 years, the document outlines various
cascade mitigation strategies and techniques that have been proven successful. In
conclusion, the report suggests that the costs of preventive and mitigation measures
and the associated gains in risk reduction of service disruptions be duly accounted
for in the overall power grid reliability analysis in order to yield sustainable line
security measures.

5.6.2 E
 xceptional Loads, Accidental Loads and Mechanical
Security of Overhead Lines

Addressing the mechanical security of overhead lines requires careful consideration

of the many loading conditions that may eventually trigger various types of failures.
A distinction is made between exceptional loads, which are defined as climatic or
natural loads in excess of design loads, and accidental loads. Exceptional loads are
related to extreme natural events which can be characterized by a probability of
occurrence and therefore be specified using a reliability analysis approach. Accidental
loads result from events that are not specifically accounted for in design; they are
considered deterministic as it is generally difficult to assign reliable values of proba-
bility of occurrence to various accidents.
From a line security perspective, both the exceptional and accidental loads have to
be considered together. Clearly, accidental loads involving storms of large footprints
are likely to trigger multiple failures of weak components while more localized con-
sequences are expected when accidents happen in normal operational conditions. In
traditional design practice, accidental loads on transmission lines have been represen-
ted by conductor failure (conductor breakage load or unbalanced residual static load
(RSL)) occurring in normal operational conditions. Although conductor breakages or
failures of conductor attachment hardware are rather uncommon in normal conditions,
their likelihood is increased in severe weather conditions and their consequences will
obviously differ. This highlights the importance of these considerations as they will
influence the choice of security measures in the specific context of each utility.

5.6.3 Line Cascade or Multiple Support Failures?

Cascades of power line supports are (unfortunately) classical examples of failures

by progressive collapse, and Figures 5.15, 5.16, 5.17, and 5.18 illustrate recent acci-
dents. It is noteworthy that the study of progressive collapse of structures has
Figure 5.15 Longitudinal
cascade on a 735 kV line
in Canada during the Great
Ice Storm of January 1998
(Photo: B. Breault, La
Presse, Montréal).

Figure 5.16  Close-up of

tower top failure in a
longitudinal cascade on 11
supports of a 500 kV line
in Oregon (USA) (Photo:
L. Kempner, Jr.).
5  Structural and Mechanical Design 257

Figure 5.17  Transverse cascade on the 400 kV D/C Rubí-Vandellós-Pierola line in Spain during
a 2008 a High Intensity Wind (HIW) storm (Photo: J. Santana López).

Figure 5.18  Transverse cascade of five 400-kV towers due to a May 2006 a High Intensity Wind
(HIW) storm in Czech Republic (Photo: P. Fronek).
258 E. Ghannoum

emerged as a specific research domain in structural engineering only in the last

decade [1, 2]. While the successive failures of several towers are not always the
result of a domino-effect, they often are. The definition of cascades used in the inter-
national standard IEC 60826 [3] is adopted here: “… uncontrolled progression of
failure (cascades) which might otherwise extend well beyond the failed section,
whatever the extent of the initial failure.” It is useful to distinguish between cascades
triggered by vertical and longitudinal effects (Figures 5.16 and 5.17), and transverse
effects (Figures 5.17 and 5.18) to understand how they happen, while cascade miti-
gation has to encompass all of them.

5.6.4 Learning from Recent Major Tower Cascading Failures

Chapter 2 of Cigré TB 515 [11] reviews the salient features of the most important
cascading failures of the overhead line history, with more emphasis on the catastro-
phic line cascades that occurred in the last 20 years. This review points to triggering
events (rupture of hardware conductor fittings, shield wires and conductors, and sup-
port failures) systematically occurring during extreme weather conditions such as
high intensity wind storms (downbursts and tornadoes), hurricanes, large scale ext-
reme synoptic winds, and ice storms. It includes the salient features of Cigré TB 344
Big Storm Events What We Have Learned (2008). These large scale weather events
are incentives to intensify maintenance programs. Poor condition of the weakest
elements sensitive to primary failures has to be registered systematically and mitiga-
ted. Vegetation management includes clearing vegetation and widening of corridors
to avoid trees falling on conductors and supports, especially when fast-­growing trees
threaten the conductors.
One of the main conclusions of Cigré TB 344 is that an emergency response plan
includes not only the emergency restoration systems and the spare parts, but also the
complete organization, reciprocal assistance agreements and regular exercises.
Crews must be very familiar with the restoration systems and their erection and use.
A good balance between preventive and corrective measures is highlighted. To pro-
vide a practical example in some detail, the experience with emergency restoration
systems in South Africa is presented in more details in Chapter 2 of Cigré TB 515
[11] Since 2001, Eskom’s Transmission has been investing significant effort into
developing and expanding their emergency restoration response capability. The uti-
lity has developed an emergency response strategy which involves a number of dif-
ferent restoration systems, each with different capabilities and strengths, and suited
for use under different emergency conditions.

5.6.5 Current Understanding of Dynamic Line Cascading

Chapters 3 and 4 of Cigré TB 515 [11] present the current understanding on the
response of OHL to accidental loads, describing (with some examples) the structural
dynamics of a cascading line section. This understanding has been greatly enhanced
5  Structural and Mechanical Design 259

by research advances in OHL computational dynamic analysis in the last 20 years.

Careful numerical simulations make it possible to assess the cascading vulnerability
of existing lines and to simulate the effects of various techniques or approaches to
limit cascades. They are also instrumental in deriving and validating simplified pro-
cedures more amenable to design practice. Slower cascades such as those triggered
by damage to towers or foundations are essentially quasi-static phenomena, while
dynamic amplification effects of unbalanced loads need to be considered to describe
faster cascades triggered by a rupture in the conductor system (cable rupture or hard-
ware fitting). It should be kept in mind that static analysis of towers under various
unbalanced load scenarios (including residual static loads (RSL) calculated assu-
ming a conductor rupture) is only representative for “slow” cascades.
The magnitude of the peak dynamic load (PDL) alone may not necessarily provide
a realistic indicator of the ability of the damaged system to contain the failure event.
These aspects will be discussed in more details in upcoming Cigré TB. Generally, the
impact of the event on the tower and foundations is affected by:

• The natural frequency of the tower and intact span(s)

• The response of tower members to transient loads of short duration
• The dispersion of the failure shock from the conductor support point into the
superstructure
• Soil-structure interactions.

Current understanding of the response of OHL systems to accidental loads stems

from a number of published experimental studies [4, 5], at both the reduced-scale
and the full-scale, and several references are listed in Chapter 9 of Cigré TB 515
[11]. It is only recently that numerical simulations [6–8], have allowed the expansion
of this understanding (see Table 5.37), with the possibility of simulating previous
physical tests for validation purposes, and then as predictive tools for various failure
scenarios for which experimental results are not available. One should keep in mind
that each modeling scenario and physical test is unique with its own assumptions and
physical characteristics. In particular, one should be careful in extrapolating specific
results to other situations. This is an important caveat of small-scale tests results as
it is practically impossible to replicate realistic problem parameters related to inertia
effects and internal damping.

5.6.6 Recent Developments in OHL Cascading Mitigation

Prevention of cascades by effective failure containment measures is a critical aspect

of line design. Chapter 5 of Cigré TB 515 [11] reviews several systems and strate-
gies that have been developed and implemented in the last few decades, with varia-
ble degrees of success as, unfortunately, OHL systems continue to fail during big
storm events. Older lines constructed before strict longitudinal load requirements
started to appear in design guidelines are particularly at risk, considering the additi-
onal uncertainty in component resistance that comes from aging.
260 E. Ghannoum

Table 5.37  Line parameters affecting the peak dynamic load (PDL) and the residual static load
(RSL) due to conductor breakage
Line parameter Effect on PDL and RSL
Conductor tension The strain energy stored in the conductor in the intact profile
is the primary source of kinetic energy following the broken
wire event. Larger energy releases induce larger
accelerations, and consequently larger conductor tensions in
the intact profile will produce greater inertia forces and
motions, and larger PDL.
Heavy conductors, or conductors experiencing sustained
overload will also result in a higher RSL.
Assembly Type (Suspension/ Suspension I- or V-string insulators swing out, thus
Strain) increasing the available slack and decreasing the RSL.
However suspended spans will experience a second peak
dynamic load as the healthy span “bottoms out” following
conductor failure, which will increase the PDL (as it will be
governed by the second peak).
The assembly type also affects the load application rate.
Towers fitted with suspension assemblies will initially see no
load until the suspension assembly swings out, however strain
assemblies will transfer an instantaneous step change or
shock load propagated by the failure.
Length of suspension insulator Longer suspension strings, and specifically the ratio of
string suspension string length to slack, decrease the RSL.
The same ratio may have the effect of increasing the PDL as
the injection of more slack into the healthy section will allow
further acceleration before the healthy section “bottoms out”.
Number of suspension spans The number of suspension spans to the next strain point will
from the assessed tower to the tend to increase the RSL, since successive suspension sets
next dead-end structure also swing out, thus reducing the net slack length injection
into the healthy span.
However, the energy dissipated by successive suspension sets
also moving will decrease the PDL. These effects typically
saturate after about four spans.
Load application rate The failure mode has a dominant effect on the PDL. This may
vary from an instantaneous shock load (e.g. from severed
conductor bundles) to a more gradual quasi-static application
(e.g. from the longitudinal failure of an adjacent support).
Span/Sag ratio Spans with lower span to sag ratios tend to produce a higher
PDL and RSL.
Span length PDL decreases with a decrease in span length (and more for
flexible supports). A shorter span adjacent to the breakage
point promotes dynamic interactions with successive spans,
especially the second one.
Flexibility of adjacent supports Adjacent supports may be deflected by the step change in
conductor tension and thus damp the movement of the
healthy span, thus lowering the PDL. Slower transverse wave
propagation, increased duration of peak loads, increased peak
tower displacements and increased interaction with intact
cables also result.
The effect is minimal on RSL.
(continued)
5  Structural and Mechanical Design 261

Table 5.37 (continued)
Line parameter Effect on PDL and RSL
Flexibility of affected support Flexible supports may absorb some strain energy when
failing in a ductile failure mode not involving instability.
However, this ductile collapse occurs after the failure shock
load has been transmitted to the adjacent spans and their
supports. Tower strength is a more important parameter than
tower stiffness and ductile failure modes are beneficial for all
types of supports, especially when the failure event does not
involve cable breakages.
Supporting effect of ground Where the ground wire remains intact during a failure event,
wires it acts as a longitudinal guy that significantly reduces the
longitudinal ground line moment. This will reduce both the
PDL and RSL and induce strong dynamic interaction at the
tower next to breakage point.
Intact phases Decrease in PDL and wave propagation speed. Decrease in
amplitude of tower displacements. Decrease in maximum
tension in guy wires. No (or very little) dynamic interaction
for stiff supports, more with guyed supports.
Plastic hinging or ductile Filters high frequency peaks of response. Reduces PDL at
failure of cross arm next to first support but increases dynamic interactions with next
breakage point support. No or little effect on displacements of second
support and on speed of transverse stress wave.
Local failure of supports involving large displacement will
significantly reduce RSL.

The first consideration in OHL cascade mitigation is to assess whether or to what

extent the post-elastic reserve of the supports can be relied upon. Cascade mitigation
devices such as control sliding clamps and load-limiting cross arms protect the sup-
ports from excessive loads by providing additional slack and localized dissipation of
energy. Another approach is to combine the use of clusters of a few stronger suspen-
sion towers at regular intervals with regular suspension towers of nominal strength.
Anti-cascading supports are intended to confine longitudinal failures to line sections
within their boundaries: while stronger suspension supports may suffice to confine
slow cascading line sections, anti-cascading strain supports are necessary to stop
dynamic cascades that are induced by ruptures in the conductor-­tension system. On
power grids located in regions prone to atmospheric icing, line deicing provides a
non structural solution, keeping in mind that other failure scenarios involving no
icing will require proper mitigation as well. Finally, a few special load control
devices are discussed, which have been considered by some utilities but have not
been fully validated or implemented.

5.6.7 Security Design Criteria to Prevent OHL Cascades

Chapter 6 of Cigré TB 515 [11] reviews the line security design criteria for new
lines as stipulated in the following documents: IEC 60826 [3], CENELEC EN
50341–1 [9], and ASCE Manual 74 [10]. While a comprehensive review of the
262 E. Ghannoum

criteria used by all Cigré utility members was beyond the scope of the work, some
particularities specific to several utilities and countries have also been collected
from two surveys, the first was conducted in 2001 by Ghannoum under a CEATI
(Centre for Energy Advancement through Technological Innovation http://www.
ceati.com/) project [12] and the second is an informal working group Cigré B2.22
survey conducted in 2010.
In 2001, at the initiative of its special interest group on Wind and Ice Storm
Mitigation (WISMIG), CEATI sponsored a survey on the design approaches used by
its utility members for overhead transmission lines under severe climatic loads due
to wind and atmospheric icing effects. The survey questionnaire and collected res-
ponses, reproduced with CEATI’s permission in Appendix B of Cigré TB 515 [11],
pertained to new lines as well as lines to be upgraded. The respondents to the ques-
tionnaire represented 13 utilities from North America, Japan and some European
countries. It is noteworthy that 12 of the 13 respondents stated that they make cont-
ainment provisions to limit damage in case of weather exceeding the design loads.
Eight utilities reported that they consider a broken iced conductor load (any conduc-
tor or ground wire, one at a time) for suspension tower design case, although none
of the respondents has reported to consider any dynamic load effect on the tower
either under bare or partial iced conditions.
To complement the results of the CEATI survey with more participation from
other utilities members of Cigré, WGB2.22 has conducted an internal survey (among
the WG members and their affiliates) specifically on current design practices used for
preventing OHL cascade failures. Both the survey questionnaire and a summary table
reporting the salient results are reproduced in Appendix C of Cigré TB 515 [11].
The results of the internal WGB2.22 survey indicate that most National Trans­
mission Line Standards used in European and North American countries (Belgium,
Czech Republic, Poland, Serbia, Spain, France, United States and Canada) and in
other countries around the world (Australia, Brazil, Japan, South Africa) prescribe
some design load criteria for preventing or mitigating the effects of OHL tower
cascading failures. It is also usual practice in those countries with severe icing con-
ditions, to install reinforcing stop structures or anti-cascading supports with an aver-
age interval of 10 spans (4 to 5 km). For most utilities, a stop structure is a strain
tower, whereas for some others anti-cascading towers are suspension supports desi-
gned to resist large quasi-static unbalanced longitudinal loads.
Brazil has reported the successful use of an alternative containment load case that
consists of the simultaneous application of 50 % of the everyday longitudinal load on
all phase attachment points and full everyday tension at ground-wire on all suspen-
sion type supports. This practice provides a more continuous distribution of mecha-
nical robustness along the line and therefore does not require a limitation on the
number of spans in a line section between anchor towers. None of the transmission
lines designed with this failure containment criterion has suffered any cascading
failure to date.
The failure containment philosophy of Bonneville Power Administration in the
United States (presented in some detail in Section 5.5 of Cigré TB 515 [11] is an
intermediate design approach between the discrete robustness provided by stop
5  Structural and Mechanical Design 263

structures or anti-cascading suspension towers, and the continuous robustness con-

cept adopted by Brazil: it has proved quite successful as well as BPA has experien-
ced only one line cascade to date, where damage was limited to the tower tops of 11
structures in a single-circuit 500 kV line in Oregon (see Figure 5.16).

5.6.8 F
 ramework for Successful Design to Limit Overhead Line
Cascades

In the light of the information presented before and with a clear understanding of the
various possible cascading mechanisms, a combination of reliability-based design
(RBD) and deterministic approaches seems inevitable, a fact that has been recogni-
zed by most utilities: i.e.

• Use RBD (also called Load & Resistance Factor Design LRFD) in the intact line
system;
• Use pragmatic engineering based on judgment, imagination and experience for
line security design. This combines deterministic considerations, careful analysis
and decision-making about the acceptable extent of collapse and the ­acceptable
other damage. Again, although the focus of Cigré TB 515 [11] is on technical
aspects, economical considerations are crucial in the decision-making.

The question still remains open as to whether “Preventing local failure” or “Assuming
local failure and mitigating its consequences” is the best way to go, but both approa-
ches may be applied depending on the load level. A conceptual model is illustrated in
Figure 5.19 (from Chapter 7 of Cigré TB 515 [11]) which relates environmental loa-
ding conditions in the line section and various robustness enhancement measures of
diverse efficiency and cost to yield a range of extent of collapse or damage that leaves
way for utilities to make decisions that best fit their specific contexts. For example,
one could define acceptable ranges of failures (Nf) of 1 to 3 towers for accidents under
every day operational conditions, 3 to 5 towers for accidents under severe environ-
mental load conditions (ex. Intermediate combinations of ice and wind if applicable,
or wind pressure with small probability of exceedance), and 5 to 9 towers for extreme
(design-level) load conditions.

5.6.9 Conclusions and Recommendations for Future Action

The global grid reliability approach is now the ruling design method in most electric
utilities. In this global context, security measures are specifically encompassed
inside the analysis framework. A concern for OHL design engineers is that cur-
rently, global grid reliability models used by most utilities are not properly accoun-
ting for mechanical security aspects of lines. As a matter of fact, in the global grid
context, what matters is the availability of any OHL line section to transit electri-
city: if any specific section is not available while the electric power can safely be
264 E. Ghannoum

Environmental
loading and cost of
conditions robustness
in the intact enhancement
system measures
Every Low
day

Severe Moderate

Extent of collapse or damage, N f

Extreme High

Figure 5.19  Schematic of incremental OHL cascade mitigation.

delivered to clients by an alternative route, the consequences of its “failure” are not
as costly as if strategic “back bone” routes are affected. Since the global-grid analy-
sis framework does not yet allow introducing different failure-containment options
for different OHL routes, it is very difficult for OHL engineers to introduce innova-
tive methods in their structural design and to develop failure containment strategies
that can be both technically effective and cost-effective. Interesting mitigation
options can therefore be rejected on the basis of lack of validation lack of understan-
ding, or rigid traditional decisions. Since research and development funds for novel
mitigation options are often linked to major capital investment projects, rejection of
options due to lack of validation is a great concern.
In conclusion, it is suggested that the costs of preventive and mitigation measures
and the associated gains in risk reduction of service disruptions be duly accounted
for in the overall power grid reliability analysis in order to yield sustainable line
security measures. For this to really happen, experts in mechanical/structural over-
head line design will have to work together with grid reliability analysis experts.
Cigrè SC B2 and its related normative bodies have the opportunity to take the lea-
dership in this collaborative effort.
5  Structural and Mechanical Design 265

5.6.10 References

[1] Knoll, F., Vogel, T.: Design for Robustness. International Association for Bridge and Structural
Engineering (IABSE), Structural Engineering Documents Series, No. 11, ISBN 978-3-85748-
120-8, Zurich, 99 p (2009)
[2] Starossek, U.: Progressive Collapse of Structures. Thomas Telford, London, 153 p (2009)
[3] International Standard IEC 60826, 3rd Ed. 2003–10. Design criteria of overhead transmission lines
[4] Vincent, P., Huet, C., Charbonneau, M., Guilbault, P., Lapointe, M., McClure, G.: Testing and
numerical simulation of overhead transmission lines dynamics under component failure condi-
tions. In: 40th General Session of Cigré, Paris, 29 August- 3 Sept., Paper No. B2-308, 8 p (2004)
[5] Clark, M., Richards, D.J., Clutterbuck, D.: Measured Dynamic Performance of Electricity
Transmission Towers Following Controlled Broken Wire Events. In: Proceedings of the 41st
General Session of Cigré, Paris, Paper No. B2-313 (2006)
[6] McClure, G., Lapointe, M.: Modeling the structural dynamic response of overhead transmis-
sion lines. Comput. Struct. 81, 825–834 (2003)
[7] Tucker, K., Haldar, A.: Numerical model validation and sensitivity study of a transmission line
insulator failure using full scale test data. IEEE Trans. Power Deliv. October 2007, PWD-22(4),
2439–2444 (2007)
[8] Peabody, A.B., McClure, G.: Modeling the EPRI-Wisconsin power and light broken wire tests.
IEEE Trans. Power Deliv. 25(3), 1826–1833 (2010)
[9] EN 50341–1: Overhead Electrical Lines exceeding AC 45 kV –Part 1: General Requirements –
Common specifications, CENELEC European Committee for Electrotechnical Standardization,
Brussels (2001)
[10] American Society of Civil Engineers (ASCE).: ASCE Manual 74, Guidelines for Electrical
Transmission Line Structural Loading, Manuals and Reports on Engineering Practice,
American Society of Civil Engineers, Reston (2010)
[11] Technical Brochure 515, Mechanical Security of Overhead Lines Containing Cascading
Failures and Mitigating Their Effects, Cigré October 2012
[12] Ghannoum, E.: Assessment of the impacts of increasing structural reliability and security by
designing lines for longitudinal broken conductor and unbalanced icing loads, Cigré paper
B2-204 (2008)
[13] Ghannoum, E.: Comparison Between Unbalanced Ice Loads and Security Loads and Assessment
of their Impact on Line/Structure Designs. CEATI Publication T063700 3335 (2008)

5.7 Influence of Design Parameters on Line Security

5.7.1 Introduction

This chapter is based on a CEATI research study performed by Ghannoum (Ghannoum

2008) and a Cigré paper by the same author (Ghannoum 2008).
All transmission lines are designed for vertical and transverse weather loads such
as wind as well as for combined wind and ice if in icing areas. Besides these loads,
there are other types of loads that apply important longitudinal forces on transmis-
sion structures, the first are unbalanced ice loads (UIL) and the second broken con-
ductor loads (BCL).
The requirements for longitudinal loads vary between design practices. IEC
60826 (Probabilistic Design of Overhead Transmission Lines 2001) and EN 50341
266 E. Ghannoum

(Ghannoum and Orawski 1986) standards cover both UIL and BCL requirements
for designing towers, while ASCE 74 (2006 draft) (EN 50341–1 2001) specifies
only BCL requirements. The above loading conditions are discussed in this chapter
and the results obtained provide some answers about the severity of UIL and BCL
compared to weather loads such as wind and combined wind and ice loads as well
as in order to assess their relative severity and impact on line and tower design.
Different types of structures (lattice single and double circuits, steel poles and
wood H frames) with spans from 150 m to 475 m were used in documents (Ghannoum
2008; Ghannoum 2008). A parametric sensitivity study was also performed to pro-
vide a broader picture of the impact of many line parameters on UIL and BCL such
as varying ice thickness, flexibility of structures, number of spans to next dead-end
structure, suspension insulator length, and the use of a dynamic impact factor applied
to the static residual load resulting from a broken conductor.

5.7.2 The Need for Unbalanced Longitudinal Loads

In the past 50 years, many utilities have suffered cascading structure failures that
extended from about a dozen structures to over hundreds of structures. As a conse-
quence of such types of failures, the design of lines for broken conductor loads (BCL)
has been introduced in many practices and standards and recognized as one of the
main methods that would prevent or reduce the risks of cascading tower failures. In
addition to BCL, longitudinal unbalanced ice loads (UIL) can occur on overhead
lines as a natural consequence of either ice accretion or ice shedding on conductors
and structures and has to be dealt with in the design process.

5.7.3 R
 equirements of Standards, Design Codes, and Utility
Practices

Standards and practices such as IEC 60826, ASCE 74, NESC, and EN 500341, as
well as WISMIG8 utility practices regarding unbalanced ice loads (UIL) and broken
conductor loads (BCL) have been investigated. The most common security loads
found in these references consist of broken conductor/wire loads,9 calculated at
every day tension (EDT). The major difference between IEC-EN (Probabilistic
Design of Overhead Transmission Lines 2001; Ghannoum and Orawski 1986) and
ASCE (EN 50341–1 2001) relates to the number of broken conductor points for a
double circuit line. The ASCE proposes to double the number of broken conductor
points from one broken phase to two. There is also a requirement in both IEC and
EN (Probabilistic Design of Overhead Transmission Lines 2001; Ghannoum and
Orawski 1986) to design for an unbalanced longitudinal load produced by the ten-

8
 Utilities participating to the Overhead Line Design Issues & Wind and Ice Storm Mitigation
Interest Group of the CEA Technologies Inc. (CEATI).
9
 In this paper, conductor is used as a general term for ground wires and phase conductors.
5  Structural and Mechanical Design 267

sion of bare conductors and GW in all spans in one direction from the structure and
with a fictitious overload equal to the weight, w, of the conductors and ground wire
(GW) in all spans in the other direction.
As regards UIL, IEC (Probabilistic Design of Overhead Transmission Lines 2001)
and EN (Ghannoum and Orawski 1986) specify three cases of unbalanced ice loads,
longitudinal bending, transverse bending, and torsion where the heavily loaded spans
are subjected to 70 % of design ice weight,10 while the lightly loaded spans are sub-
jected to 40 % of the heavily loaded spans. These ice loads are applied in various
schemes that maximize either torsional or bending moments on the structure.
Based on the above requirements, the following longitudinal structure loading
cases were used in this study, in addition to conventional weather loads (wind and
combined wind and ice):

• Any one broken conductor or GW at every day tension (IEC and EN standards)
• Any two broken conductors (ASCE requirements for double torsion)
• Various ice unbalanced (transverse, bending and torsion) schemes as per IEC and
EN
• Longitudinal unit conductor load (IEC and EN)
• All broken conductors and GW at EDT (optional in IEC)
• Any one broken conductor with a dynamic factor of 1.8 applied to the RSL (opti-
onal in IEC)
• Any one broken conductor or GW with maximum ice of 30 mm thickness (opti-
onal in IEC).

5.7.4 Comparison of BCL and UIL with Weather Load Cases

5.7.4.1  Methodology
In order to perform the necessary calculations of this study, five different lines, and
structures were modeled in PLS-CADD, TOWER, and PLS-POLE11 as indicated in
Table 5.38. The three 230 kV structures (wood H frame, lattice steel, and steel pole)
were first optimized for basic weather loads (transverse) and then used to assess the
impact of longitudinal unbalanced loads on their design. It is noted that all ­longitudinal
loads were calculated taking into account the flexibility matrix of structures as well
the movement of insulator strings12 in a section of about 15 spans or more.
The severity of longitudinal unbalanced loads was assessed in relation to trans-
verse weather loads in many fashions, i.e. by comparing: a) Loads applied to a single
GW or conductor point due to each load case, b) Net (total) longitudinal or trans-
verse structure loads due to each load case, and c) The detailed design of structures
(stresses in members, moments, foundations reactions, etc.).

10
 In case the design ice load is specified in radial thickness, this should be converted to unit weight
of ice on the conductor prior to applying the factors 70 % and 28 %.
11
 All the software are trademarks of Power Line Systems Inc.
12
 This option is called level 3, Finite Element analysis in PLS-CADD software.
268 E. Ghannoum

Table 5.38  Modeled lines and structures

Span Height Vertical insulator Ratio of slack to
(m) (m) length (m) insulator length (%)
161 kV single circuit wood 150 15.9 1.4 8 %
H frame
•230 kV single circuit wood 300 22.7 2.2 27 %
H frame
•230 kV double circuit steel 400 55 2.2 55 %
pole structure
•230 kV double circuit 330 37.1 2.2 34 %
lattice steel structure
•735 kV single circuit lattice 475 40.5 6.0 32 %
steel structure

A parametric study was also performed in order to assess the variation of the
resulting loads with the:
a) Flexibility of structures, b) Length of insulator strings, c) Number of spans to
the next dead-end structure, d) Magnitude of the design ice thickness, and e) Dynamic
Impact factor.
Furthermore, clearances between conductors and overhead ground wires were
calculated for various unbalanced ice conditions.

5.7.4.2  Modeled Structures

The main parameters of the modeled lines and structures are given in Table 5.38. It
is noted that all conductors were strung using two limits: a) The initial bare conduc-
tor catenary parameter (horizontal tension/unit weight) should not exceed 1800 m at
minimum temperature, and b) The maximum conductor tension should not exceed
75 % of the UTS under the most critical climatic condition.
The weather base cases used for the initial design of structures consisted of a
wind speed of 30 m/s (10 min. average, at 10 m of height, in a terrain type B) as per
IEC 60826 and an ice thickness of 30 mm (relative density of 0.9).

5.7.4.3  R esults of Comparison Between UIL and BCL Loads

and Weather Case Loads
These results given in Table 5.39 can be summarized as follows:

Transverse (IEC) and Longitudinal Loads on a Single Ground Wire (GW) or

Conductor Attachment Point
Broken GW or conductor loads (longitudinal) are significantly larger (1.5 to 2.5
times) than wind loads (transverse) at the same point.
The residual static load (RSL) of a broken GW is more critical than a broken
conductor load in proportion to their respective wind loads. This result is due to
effect of the length of the insulator string in reducing the conductor RSL compared
to the GW that does not have a suspension string.
The RSL of a broken conductor is about 40 to 70 % of the initial conductor ten-
sion. This percentage decreases with the ratio of the insulator string length to the
slack in the conductor.
5  Structural and Mechanical Design 269

Table 5.39  Longitudinal loads (% of IEC wind) for various load cases and lines
161 kV 230 kV 230 kV 230 kV 735 kV
wood Wood steel pole lattice lattice
Longitudinal load on a single GW 254 % 196 % 153 % 192 % 167 %
point due to broken GW 1
Longitudinal load on a single 160 % 153 % 133 % 157 % 106 %
conductor point due to broken
phase 3
Net (total) longitudinal loads due 6 % 4 % 4 % 10 % 5 %
to broken GW 1
Net (total) longitudinal loads due 12 % 10 % 10 % 21 % 32 %
to broken phase 3
Net (total) longitudinal bending 28 % 27 % 45 % 68 % 54 %
loads due to unbalanced ice
(70 %-28 %)

Net (total)13 Transverse and Longitudinal Conductor and GW Loads on the

Structure
Net or total structure loads resulting from one or two broken conductors are signifi-
cantly smaller than the net wind loads. In the case of one broken GW, total loads are
in the range of 4 to 10 % of total wind loads. In the case of a broken conductor, total
longitudinal loads range from 10 % to 32 % of transverse wind loads.
Total unbalanced ice loads are significantly higher (1.7 to 4.5 times) than total
loads due to one broken conductor, but still less than the total transverse wind loads.

5.7.4.4  S  ummary of Results Related to the Variation

of the Suspension String Lengths (Table 5.38)
This study confirmed that for flexible structures, intact ground wires act as longitu-
dinal guys that oppose broken conductor cases. This behavior significantly reduces
total unbalanced loads acting on the H frame wood structures due to the breakage of
one phase.
The drop of the RSL in % of the conductor tension prior to breakage (initial ten-
sion) is almost linear with the ratio of insulator length (Li) to the conductor slack
(difference between conductor and chord lengths) for all lines, except for the 161 kV
line with much shorter spans.
In the case of net (total) longitudinal loads acting on structures, all of them
decrease with increasing length of insulators, except for broken GW cases where the
opposite scenario occurs. It is noted that net (total) unbalanced ice loads are reduced
by about 7 to 14 % in the range of increasing insulator lengths. The drop is lower for
flexible structures than for the rigid lattice steel structures.
The linear variation of the RSL with the ratio of Li/slack shown in Figure 5.20 is
an important result of the study. According to the author, it is the first time this fin-
ding it is being published in the technical literature.

13
 Net or total loads are the algebraic sum of all forces on the structure either in the longitudinal or
transverse direction, depending on the load case.
270 E. Ghannoum

80%

75%

70%

65%

60%

55%
1.0' 1.5' 2.0' 2.5' 3.0' 3.5' 4.0' 4.5' 5.0' 5.5'

Ratio Li/slack

Figure 5.20  Residual static tension of phase 3 in % of initial tension (excluding the 161 kV).

Table 5.40  Global summary of the effect of insulator length (Li) on longitudinal loads
Line Options of insulator lengths
Li (m) 1.1 1.4 1.7 2
161 kV wood Li/Slack 9.7 12.4 15.1 17.7
Broken ph-3 48 % 45 % 41 % 39 %
Li (m) 1.8 2.2 2.6 3
230 kV Wood Li/Slack 3.0 3.7 4.3 5.0
Broken ph-3 61 % 58 % 56 % 54 %
Li (m) 1.8 2.2 2.6 3
230 kV lattice Li/Slack 2.4 2.9 3.4 4.0
Broken ph-3 65 % 62 % 59 % 57 %
Li (m) 1.8 2.2 2.6 3
230 kV pole Li/Slack 1.5 1.8 2.1 2.5
Broken ph-3 76 % 73 % 71 % 69 %
Li (m) 5.5 6 6.6 7.1
735 kV steel Li/Slack 2.8 3.1 3.4 3.7
Broken ph-3 63 % 61 % 60 % 59 %
Note: Broken conductor tensions are given in % of GW or conductor tensions prior to breakage

5.7.4.5  S
 ummary Results as Regards the Impact of Structure
Flexibility
The common assumption of rigidity of structures increases the majority of loads
compared to the real flexible behavior of the same structures, most noticeably with
H frame wood poles (Table 5.40).
5  Structural and Mechanical Design 271

In the wood H frame structures (161 kV and 230 kV), the rigidity assumption
leads to an increase of about 3 to 6 times in net (total) loads compared to the flexible
calculations.
In the case of 230 kV steel pole, the ratio between rigid and flexible calculated
longitudinal loads can reach a value of 2.35.
In the case of the 230 kV lattice steel tower, the ratio between rigid and flexible
longitudinal load calculations can reach a value of 1.21.
For the 735 kV lattice steel tower, the most rigid structure in this study, the ratio
between rigid and flexible longitudinal loads can still reach a value of 1.06.
Based on the above results, neglecting flexibility of structure when calculating
unbalanced longitudinal loads will lead to unnecessary increased loads and there-
fore, an increase in the structure costs. This practice should be abandoned, particu-
larly when there is reliable software that can perform such calculations easily.

5.7.4.6  S  ummary Results of the Effect of Number of Suspension

Spans to Next Dead-end
In the majority of cases, the magnitude of unbalanced loads decreases when the
number of spans between the studied structure and the next dead-end structure
decreases (Table 5.41).
The reduction of total unbalanced ice loads (70 % and 28 %) is very sensitive to
the number of spans to the dead-end (DE) structure, particularly when this number
drops to 1 to 3 spans. The net (total) unbalanced ice loads drop from 100 % to (27 %
to 59 %) for all structures, when the number of spans to the next DE structure drops
from 7 to one span.
Net (total) loads on the structure are almost unaffected by the number of spans to
the next DE structure for broken GW cases.
Net (total) loads due to broken conductors are affected by the number of spans to
the next DE structure, but to a lesser degree than unbalanced ice loads. The lines that
have the highest value of Li/Slack (161 kV wood H frame) showed the highest reduc-
tion in total loads when the number of spans was reduced.

Table 5.41  Variation of loads on a single point due to number of spans to the next dead-end
7 5 3 2
spans spans spans spans 1 span
161 kV wood H frame Broken GW 1 100 % 100 % 99 % 96 % 87 %
Broken ph-3 100 % 94 % 85 % 77 % 62 %
230 kV wood H frame Broken GW 1 100 % 100 % 100 % 99 % 93 %
Broken ph-3 100 % 98 % 94 % 88 % 72 %
230 kV steel pole Broken GW 1 100 % 100 % 100 % 100 % 98 %
Broken ph-3 100 % 100 % 98 % 95 % 83 %
230 kV lattice Steel Broken GW 1 100 % 100 % 100 % 100 % 99 %
Broken ph-3 100 % 100 % 97 % 92 % 78 %
735 kV lattice Steel Broken GW 1 100 % 100 % 100 % 100 % 100 %
Broken ph-3 100 % 100 % 98 % 94 % 81 %
272 E. Ghannoum

If the number of spans to the next DE structure increases to more than 5 to 7

spans, there is little variation in the results.

5.7.4.7  S  ummary of Results Related to the Variation of the Design

Ice Thickness
The main conclusions of structure loads evaluation using different design ice thick-
nesses of 25, 30, 35, 40, and 45 mm (Table 5.42) are: Longitudinal broken conductor
loads and transverse wind loads are not usually affected by the ice thickness (all
minimum broken conductor cases are considered at EDT).
Net (total) unbalanced ice loads increase at more than twice the rate of the increase
of combined wind + ice loads when the ice thickness is increased. For example, in the
case of the 735 kV line, total unbalanced ice loads increased 2.7 times between
25 mm and 45 mm of ice, while the combined wind + ice loads increased only 1.5
times in same range of ice thicknesses.
Unbalanced ice loads increase at higher rates with long spans and increasing ice
thickness.
Lines using lattice towers with long spans and designed with 40 mm or more of
ice thicknesses are likely to be vulnerable to unbalanced ice loads because these will
control the design of many tower members.

5.7.4.8  I mpact of Unbalanced Loads on Structures

and Foundations
The three 230 kV structures were checked for the combinations of loading cases of
Table 5.43.
Base on the above calculations, the following main results can be concluded:

230 kV wood H frame

Minimum14 broken conductor loads and unbalanced ice loads (based on 30 mm of
ice thickness) mostly affect the crossarm dimensions of this H frame structure.
However, if additional security measures are used such as designing for broken
conductors, either with a dynamic impact factor of 1.8 or under ice conditions, the
class of the pole and size of crossarms will both be affected.

230 kV lattice steel structure

UIL and BCL were not found critical for the complete tower, but they were critical
for some members directly affected by these loads (mainly crossarms and GW peaks).
Depending on the combinations of broken conductor loads, the tower weight increa-
sed up to 45 % for the most conservative combination (case 6 of Table VI) compared
to a tower designed only for weather loads. Similarly, the 230 kV square based lattice

 In this paper, minimum requirements are those specified by IEC, EN or ASCE, i.e. RSL no wind
14

or ice (EDT)
5  Structural and Mechanical Design 273

Table 5.42  Total longitudinal or transverse loads (kN) for different ice thicknesses
Radial ice thickness in mm
Line type Load case 25 30 35 40 45
161 kV wood H Unbalanced ice (bending) 3.6 4.7 5.8 6.9 7.9
frame 70-28 %
IEC combined wind and ice load 15.8 17.9 19.9 22.0 24.1
230 kV wood H Unbalanced ice (bending) 9.8 12.3 14.7 16.8 18.8
frame 70-28 %
IEC combined wind and ice load 107 120 134 147 160
230 kV steel pole Unbalanced ice (bending) 43 54 64 74 91
70-28 %
IEC combined wind and ice load 107 120 134 147 161
230 kV lattice steel Unbalanced ice (bending) 47 61 76 89 104
70-28 %
IEC combined wind and ice load 79 89 99 109 119
735 kV lattice steel Unbalanced ice (bending) 107 140 176 231 293
70-28 %
IEC combined wind and ice load 204 227 250 273 297

Table 5.43  Combinations of load cases for assessing the impact of UIL and BCL on structures
Case Case Case Case Case Case
Load combinations 1 2 3 4 5 6
IEC wind and ice (30 mm) x X X x x x
IEC broken conductor x x x x
IEC unbalanced ice x x x
ASCE broken conductors X
All broken conductors @ EDT x x
Broken conductor with dynamic impact factor x
of 1.8 on RSL
One broken conductor with 30 mm ice x
thickness

tower, once designed for weather loads, will require a weight increase by 4 % for IEC
and 10 % for ASCE15 to sustain the minimum prescribed broken conductor loads.

230 kV steel pole

The maximum ground line moment of the steel pole is dictated by the wind load case,
thus the pole foundation is not affected by BCL and UIL, except when conductors are
broken under full ice case (30 mm). The pole arms are controlled by the broken

 The larger percentage for ASCE requirements is due to the fact that ASCE requires two broken con-
15

ductors for double circuit towers, while IEC requires, as a minimum, only one broken conductor.
274 E. Ghannoum

conductor loads either with the dynamic impact factor of 1.8 or with design ice thick-
ness. The weight of the complete steel pole increases by about 9 % if the most severe
security loads were adopted. In summary, the steel pole of this study (regular 12 sided
polygon) designed for weather wind and ice loads is capable of withstanding import-
ant longitudinal poles without impact on its weight. This is mainly due to the regular
12 sided polygon used for this structure, and the fact that the weather load cases
control the design of the steel shaft.

5.7.4.9  C  learances Between Conductors due to Unbalanced Ice

Loads
Two cases of unbalanced loads cases were assessed for clearance purposes: the first
where the conductor or the GW above is uniformly loaded with ice throughout the
section, while the conductor below is bare at −5 °C, and the second calculations,
where it was assumed that the upper single span of GW or conductor in the section
is loaded with ice, while the lower conductor is bare at −5 °C.

• The results confirmed that unbalanced ice conditions can lead to reduced clea-
rances, hence possible flashover either between vertical phase conductors or bet-
ween the GW and the phase below it.
• Double circuit lower voltage lines were found more susceptible to clearance pro-
blems under unbalanced ice loads than higher voltages, particularly if the former
are designed with long spans.
• Single circuit structures with horizontal configurations were less affected by
unbalanced icing clearances between GW and conductors.
• In the case of the double circuit steel pole structure where the GW is above the
conductor, about 20 mm of unbalanced ice can cause flashover between the GW
and the lower phase. Comparatively, the 230 kV double circuit tower can handle
up to 30 mm of ice unbalance.
• Partial icing applied on one span on an upper phase of a double circuit structure
will sag due to ice overload as well as due to the longitudinal movement of suspen-
sion insulators. In some vertical tower configurations, a partial icing of 15–20 mm
on one conductor span is sufficient to sag the ice covered conductor under the bare
lower conductor. Thus partial icing in one span should be checked.

5.7.5 Recommendations

Based on the studies performed, the following recommendations are proposed:

• Standards and codes of practice should include specific requirements for unbalan-
ced ice loads. IEC and EN standards cover this requirement but ASCE manual 74
only discusses them, but does not provide any values to be considered in the design.
5  Structural and Mechanical Design 275

• The contribution of the intact GW to the static equilibrium of the structure when
subjected to UIL or BCL is real and very important, particularly in flexible struc-
tures: It should always be taken into account in the calculations of structure loads
and their design.
• Clearances between conductors and ground wires under unbalanced ice loads con-
ditions can lead to reduced clearances and possibly flashovers. This issue should be
addressed by standards committees for possible inclusion of some requirements.
• The simplified methods of calculating the residual static loads (RSL) such as
using Govers (NESC C2-2002 2999) are not needed anymore. The exact RSL
can now be calculated using appropriate software.
• The cost penalty for increasing the security levels (broken conductor loads) is not
very high for structures of the double circuit steel pole type having a regular
polygonal section. The fact that these structures are located mostly in urban
areas, may justify such increase in security.
• The transverse horizontal offset of the middle phase of a double circuit structure
helps clearances in the case of unbalanced icing between phases. In heavy icing
areas, it may be justified to design structures with an appropriate transverse hori-
zontal offset. This also might reduce galloping problems
• In the case of lattice towers designed for ≥40 mm of ice and having long spans,
UIL can be critical to structure design and will reflect in increasing structure
weight, hence costs.
• It is strongly recommended to design structures with symmetrical square bases
and bracing, particularly in icing areas. Lattice towers with rectangular bases
are much more susceptible to failures from broken conductors and unbalanced
ice loads.

5.7.6 Conclusions

This study confirmed that, for symmetrical cross-section structures, broken conduc-
tor loads can be critical locally for the design of crossarms and GW peaks. Similarly,
unbalanced ice loads generate important longitudinal bending and torsional moments
that could dictate the design of some structure parts. Furthermore, in flexible struc-
tures, the ground wire (GW) was found to act as a longitudinal guy in the case of
longitudinal unbalanced loads.
Many important findings resulted from the parametric sensitivity study:
The natural flexibility of suspension structures will reduce unbalanced loads,
even in self supporting lattice structures, and should be taken into account in all
cases.
The number of spans to the next dead-end structure, and the length of the insula-
tor strings were also found to significantly affect unbalanced loads.
Finally, it was confirmed that the importance of unbalanced ice loads relative to
wind loads increase with an increasing design ice thickness. Lines designed for an
276 E. Ghannoum

ice thickness of 40 mm or more will be more affected by UIL than those designed for
25 mm of ice or less.
The difference in the design of symmetrical section steel structures with and
without unbalanced loads was assessed and found to cause an increase in the lattice
steel 230 kV tower weight from 4 to 10 % if minimum code requirements for unba-
lanced loads were used. This percentage increases to about 45 % if broken conduc-
tors with ice and all broken conductors at EDT are used. The steel pole structure was
not noticeably affected by unbalanced loads because of its geometrical properties.
Finally the wood pole H frame needed a change in its crossarm to resist minimum
UIL and BCL.

References
ASCE 74, Guidelines for Transmission Lines Structural Loading: Draft as of June 2001
Cigré WG06.: Loading and strength of overhead transmission lines. Electra No. 129 (1990)
Design Criteria of Overhead Transmission Lines – IEC 60826, Ed. 3.0 (2003)
EN 50341–1: Overhead Electrical Lines exceeding AC 45 kV – Part 1: General Requirements -
Common Specifications (EN or CLC) (2001)
Ghannoum, E.: Comparison Between Unbalanced Ice Loads and Security Loads and Assessment
of their Impact on Line/Structure Designs. CEATI Publication T063700 3335 (2008)
Ghannoum, E.: Assessment of the impacts of increasing structural reliability and security by desig-
ning lines for longitudinal broken conductor and unbalanced icing loads, Cigré Paper B2-204
(2008)
Ghannoum, E., Orawski, G.: Reliability Based Design of Transmission Lines According to Recent
Advances by IEC and Cigré. In: International Symposium of Probabilistic Design of Transmission
Lines, Toronto (1986)
Govers, A.: On the Impact of Unidirectional F on High-Voltage Towers Following Conductor Breakage.
In: International Conference on Large Electric Systems (Cigré), Paris, Paper 22–03 (1970)
IEC 60826 – Ed. 3.0.: Design Criteria of Overhead Transmission Lines (2003)
NESC C2-2002, National Electrical Safety Code C2-2002
Probabilistic Design of Overhead Transmission Lines – Cigré Technical Brochure No. 178, February
2001 – Cigré WG B2.06

Elias Ghannoum  is an internationally renowned expert having

43 years of experience in overhead transmission line design. He
worked 28 years for Hydro-Quebec where he held the position of
chief transmission engineer for 20 years. Since 1998, Elias started
his own consultancy practice and provided expertise to many inter-
national clients such as The World Bank, Electricité de France,
ESKOM, Power Grid Corporation of India, Transelec (Chile),
Manitoba Hydro, etc. He also lectured during 15 years a graduate
course on transmission line design at the Université Polytechnique
of Montreal. Elias is Fellow of the Institute of Electrical and
Electronics Engineers (IEEE) since 1994, and has received Awards
from Cigré, the IEEE and the Canadian Standards Association for
outstanding contributions to technical work on transmission lines.
He is currently Chairman of IEC/TC7 on Overhead Conductors,
Chairman of Working Group MT1 responsible for maintenance of
all IEC/TC11 standards and Chairman of the Canadian Standards Subcommittee C22.3 responsible
for overhead line design standard based on reliability principles. Elias has authored a very large
number of technical papers and publications in Cigré, IEEE, ASCE, CEA.
Environmental Issues
6
Cathal Ó Luain, Lionel Figueroa, and Paul Penserini

Contents
6.1 Introduction ......................................................................................................................... 278
6.2 Environmental Procedures and Assessment - Guidelines ................................................... 280
6.2.1 Strategic Environmental Assessment (SEA)........................................................... 280
6.2.2 Permit Procedures and Environmental Impact Assessment .................................... 284
6.3 Environmental Impacts and Mitigations - Guidelines ........................................................ 286
6.3.1 Visual Impact ......................................................................................................... 286
6.3.2 Impact on Land Use ................................................................................................ 290
6.3.3 Impact on Ecological Systems ................................................................................ 291
6.3.4 Impact of Construction and Maintenance ............................................................... 293
6.3.5 Environmental Management Plans ......................................................................... 295
6.4 Fields, Corona and other Phenomena, Impacts and Mitigations ........................................ 295
6.4.1 Electric and Magnetic Fields at Extremely Low Frequency (ELF-EMFs) ............. 296
6.4.2 Electric Field at Extremely Low Frequency ELF- EF ............................................ 296
6.4.3 Magnetic Field at Extremely Low Frequency ELF-MF ......................................... 298
6.4.4 Assessment of the Exposure to Magnetic Field for Epidemiological Studies ........ 301
6.4.5 DC-EF and Ion Current Phenomena ....................................................................... 304
6.4.6 Corona ..................................................................................................................... 306
6.4.7 Radio and Television Interferences ......................................................................... 309
6.4.8 Atmospheric Chemistry (Ions and Ozone) ............................................................. 312
6.4.9 Aeolian Noise.......................................................................................................... 313
6.4.10 Conclusions/Guidelines .......................................................................................... 314

Originally published by Cigré, 2014, under the ISBN 978-2-85873-284-5. Republished by

Springer International Publishing Switzerland with kind permission.

C.Ó. Luain (*)

Dublin, Ireland
e-mail: [email protected]
L. Figueroa • P. Pensrrini

© Springer International Publishing Switzerland 2017 277

K.O. Papailiou (ed.), Overhead Lines, CIGRE Green Books,
DOI 10.1007/978-3-319-31747-2_6
278 C.Ó. Luain et al.

6.5 Concerns and Issues, Consultation Models for OHL Projects and
Stakeholder Engagement Strategies .................................................................................... 315
6.5.1 Introduction ............................................................................................................. 315
6.5.2 Concerns and Issues Facing Utilities -Guidelines .................................................. 315
6.5.3 Consultation Models for OHL Projects .................................................................. 316
6.5.4 Stakeholder Engagement Strategies ........................................................................ 321
6.6 Life Cycle Assessment (LCA) for OHLs............................................................................ 326
6.6.1 Introduction ............................................................................................................. 326
6.6.2 LCA Development and Early Applications ............................................................ 327
6.6.3 Power System and Overhead Line LCA in Scandinavia ........................................ 328
6.6.4 Comparison of LCA Software ................................................................................ 329
6.6.5 LCA, Overview for OHL Components, Construction and Maintenance ................ 329
6.6.6 LCA, Studies on OHL............................................................................................. 330
6.6.7 Conclusions/Recommendations .............................................................................. 331
6.7 OHL and Sustainable Development .................................................................................... 335
6.8 Highlights............................................................................................................................ 336
6.9 Outlook ............................................................................................................................... 337
References ................................................................................................................................... 338

6.1 Introduction

Overhead Lines and Environmental Issues and their interaction have been under
consideration within Cigré for many years. The issues covered have ranged from
permit procedures, environmental impact assessments and consultation methodolo-
gies for Overhead Line projects to mitigation of environmental impacts be they
visual, ecological, on land use or of construction and maintenance. The develop-
ment of reduced visual impact designs and aesthetic designs has been traced and all
issues associated with field effects inclusive of the debates on EMF and mitigation
measures have been investigated. Life Cycle Assessment for Overhead Lines has
been reported on. Many utilities and TSOs have addressed and provided informa-
tion on all these issues.
Cigré Study Committee B2 Overhead Lines (formerly SC22) investigated the
question of Overhead Lines and Environmental Issues initially in the eighties and a
report was produced. It included material on EMFs and alleged health affects but
because of sensitivities in that debate at that time it was never published as a Cigré
TB. Matters moved on and attitudes changed. Overhead Lines and Environmental
Issues were later dealt with through a number of dedicated working groups from the
mid-nineties into the mid two thousands. These were WG 14, Environmental
Concerns and Regulatory Controls, and WG 15, Life Cycle Assessment and
Environmental Concerns.
A number of Cigré TBs were produced as listed below:

• Cigré TB 147 High Voltage Overhead Lines - Environmental Concerns,

Procedures, Impacts and Mitigations (Oct 1999) (Cigre 1999).
• Cigré TB 265 Life Cycle Assessment for Overhead Lines (Dec. 2004) (Cigre 2004).
• Cigré TB 274 Consultations Models for Overhead Line Projects (June 2005)
(Cigre 2005a).
6 Environmental Issues 279

An Electra article on Environmental Management Plans (EMPs) for Activities

Associated with OHLs (Fitzgerald 2004) was also published.
Reports on Overhead Lines and Environmental Issues were produced for prefe-
rential subject (PS) 2 for the Cigré Paris Session in 1996 (Reports of Cigre 1996),
for preferential subject 3 for the Paris Session in 2002 (Reports of Cigre 2002),
preferential subject 2 for the Paris Session in 2004 (Reports of Cigre 2004a), prefe-
rential subject 3 for the Paris Session in 2004 (Reports of Cigre 2004b) and for
preferential subject 1 for the Paris Session in 2010 (Reports of Cigre 2010a). These
issues will be returned to in the 2014 B2 Paris Session in Preferential Subject 1
Minimising the Impact of new Overhead Lines.
With the reorganisation (and relabeling) of Cigré Study Committees in 2003 a
new Study Committee, C3, System Environmental Performance, was set up. Hence-
forth all environmental matters would dealt with by working groups in this SC. and
various TBs were produced by their working groups which are relevant to Overhead
Lines as part of broader power systems developments. The C3 TBs relevant to
Overhead Lines are.

• Cigré TB 487 Strategic Environmental Assessment for Power Developments

(Feb 2012) (Cigre 2012a).
• Cigré TB 548 Stakeholder Engagement Strategies in Sustainable Development -
Electricity Industry Overview (Aug. 2013) (Cigre 2013).

Cigré C3 Paris session reports from 2004 Preferential Subject (PS) 2 (Reports of
Cigre 2004c), 2006 PS 2 (Reports of Cigre 2006), 2008 PS 2 (Cigre 2008), 2010 PS
1 (Reports of Cigre 2010b), 2010 PS 2 (Reports of Cigre 2010c), 2012 PS 1 (Reports
of Cigre 2012) and 2012 PS2 (Cigre 2012b) contain many reports of interest in the
area of Overhead Lines and Environmental Issues. These issues will be returned to
in the 2014 C3 Paris Session in Preferential Subject 2 Integrated Sustainable
Approaches for T & D Development and Preferential Subject 3 Acceptance of High
Voltage Transmission Assets near Urban Areas.
A number of other C3 working groups also dealt with Environmental Issues and
Overhead Lines. One was C3.04 Corridor Management. This working group was to
review environmental issues relating to Overhead Lines corridor management for
the whole of life for high voltage networks. Cigré TB, “Corridor Management: An
overview of international trends for high voltage networks” Is expected to be pub-
lished sometime in 2014. Another C3 WG, 08 Internalising the External Environ-
mental Costs of OHLs, has essentially completed its work. A final mature draft is
awaiting approval and a Cigré TB should be published in 2014.
Other working groups in B2 and some in other study committees also produced
Cigré TBs relevant to Overhead Lines and Environmental Issues and impacts. The
B2 ones are:

• Cigré TB 278 Influence of Line Configuration on Environment Impacts of

Electrical Origin (Aug. 2005) (Cigre 2005b).
• Cigré TB 416 Innovative Solutions for OHL Supports (June 2010) (Cigre 2005c).
280 C.Ó. Luain et al.

Other Study Committee’s TBs from WGs or JWGs of interest are:

• Cigré TB 373 Mitigation Techniques For Power Frequency Magnetic Fields

(Feb. 2009) (Cigre 2009).
• Cigré TB 473 Electric Field and Ion Current Environment of HVDC Overhead
Transmission Lines (Aug. 2011) (Cigre 2011).

A Joint Working Group C3/B1(Cables)/B2 was set up in 2010. The terms of refe-
rence covered reviewing and updating of many of the issues dealt with in B2 TB 147
High Voltage Overhead Lines - Environmental Concerns, Procedures, Impacts and
Mitigations (Oct 1999) (Cigre 1999). The JWG would address issues which relate
to the processes, procedures and environmental impact assessment to obtain permits
for transmission lines including the cases of proximity of electricity transmission
lines and built development. It would also cover how transmission companies and
organisations plan the routes of new transmission lines and design the lines to
reduce environmental and visual impact in rural areas and near residential and com-
mercial buildings. It would investigate how transmission companies and organisa-
tions deal with requests to underground proposed new overhead lines and with
requests to relocate (or underground) existing lines when residential or commercial
development is planned, or is newly built, near existing lines (It may be noted that
in this Cigré B2 Green Book this latter topic of underground cables or overhead
lines is dealt with in a separate Chapter 19). However a Cigré TB is unlikely to
emerge from this JWG before 2015.
This chapter, in line with the general direction given, is based on summaries of
the above Cigré TB with updates in some key areas, including Field Effects, Corona
and other phenomena.

6.2 Environmental Procedures and Assessment - Guidelines

6.2.1 Strategic Environmental Assessment (SEA)

Strategic Environmental Assessment (SEA) is a form of environmental assessment

that is meant to evaluate the effects of major new policies or planned developments
on the environment, sustainability and stakeholder needs of large areas. Being a
relatively new kind of assessment, it has emerged in several different forms and
applications, such as:

• Environmental assessment for policies, plans and programs

• Regional assessments, referring to the environmental assessment of plans drawn
up for a specific geographical area and
• Sector environmental assessment, designed for specific economic activities, such
as energy. The latter is a common type, mainly due to the demands of multilateral
financing agencies, who have required the use of this type of instrument for pro-
ject approvals.
6 Environmental Issues 281

Develpment

Environmental Impact Strategic Environmental

Assessment (EIA) Assessment (SEA)

Environment

Economic

Social

Biophysical

Figure 6.1 Illustration of the main difference between EIA and SEA.

SEA is often assumed to be simply an Environmental Impact Assessment (EIA)

for large areas. One possible reason for this is that SEA shares its origins and com-
mon principles with EIA. However, there are significant differences between these
two assessment methodologies. As illustrated in Figure 6.1 the fundamental diffe-
rence between the two is that EIA most often focuses on the effects of a proposed
development on the environment, while SEA aims to anticipate the effects of new
developments on the environmental resources, sustainability and stakeholder needs
in an area at a more conceptual stage. The SEA will help confirm the viability of the
development concept and may contribute to the design or formulation of the
development itself.
As a result of the similarities between EIA and SEA, the approach to SEA assess-
ments has taken some common forms. One is the EIA-based approach, which invol-
ves applying the same kind of assessment procedures as in a traditional EIA, except
on a broader scope and usually at an earlier stage of planning. The second is the
sustainability-based SEA, in which an attempt is made to study the carrying capaci-
ties of the environment and the collective desire of the stakeholders for a preferred
level and type of regional development.

6.2.1.1 SEA General Structure

The SEA legislation developed in 1969 in the US (with NEPA) stated the necessity
to provide analyses of programmes, activities and regulations brought in by federal,
state and local governments with reference to the effects these activities may have
on the environment and the conservation and use of natural resources. In Europe,
each country, within the EU, has enacted or legislated its own version of the SEA
Directive 2001/42/CE. Each country has also created regulations and guidelines as
to how the assessment will be administered, and how often it will be revised.
282 C.Ó. Luain et al.

On March 21, 2003, during the Convention of Assessment of Environmental Impact,

better known as the Kiev Protocol, fellow-members of the United Nations Economic
Commission for Europe signed the Protocol of Strategic Environmental Assessment.
Only in Spain, Portugal, Italy, Belgium and the Netherlands is an SEA mandatory
for grid development planning or other plans in the power sector.
There are some countries, such as Brazil, South Africa, China and Australia,
where SEA studies are not regulated by legislation, and, therefore, are not man-
datory. Some countries have published guideline documents for undertaking SEA
(e.g. South Africa and Canada). In other countries, SEA studies are being developed
voluntarily or through informal agreements. The following generic stages are
usually part of the process for conducting an SEA study

• Screening: identifying the need for SEA

• Scoping: setting targets and boundaries for the study
• Identification and assessment of alternative scenarios
• Report: analysis and report preparation and review
• Decision: consultation and decision making
• Monitoring: measure report, monitoring and follow-up.

The first step – Screening - is to determine if an SEA is actually needed. The entity
responsible for developing the plan or programme (promoter) consults the authori-
ties, with specific environmental responsibilities, to confirm if the plan or pro-
gramme needs an SEA procedure. This is followed by a scoping process, which
consists of identifying the issues to be addressed and the targets and boundaries that
are pertinent to these issues. Alternatives for the plan or development are then
considered.
For each alternative under consideration, the expected impacts are predicted, as
is the significance of each impact on the environment, social conditions and sustain-
ability of the area. Cumulative effects are also investigated for the areas involved. A
preliminary report is prepared. The choice of a preferred plan or development option
then takes into account the findings and suggestions of the SEA report. This process
is accompanied by consultations with the authorities and with the public. Once
approved, a monitoring process is put in place. The preliminary report is then made
available to the public, and to those authorities with specific environmental respon-
sibilities, in order to ensure an adequate review. Once the review period is conclu-
ded, the promoter must then take into account these results and draft the final version
of the plan or programme.
SEA for transmission has been applied to plans and programmes, such as
National Development Plans. It has also been applied as the first step of the plan-
ning process for regional and international power system interconnections. In these
cases, the planning process provides the assessment of a region with the intent to
select a “corridor” for the transmission line. A “corridor” is understood as the sec-
tion of territory where technical, environmental and territorial conditions meet the
requirements for routing power transmission lines and related plants. Alternatives
are analysed based on indicators, and the best alternative is determined, as shown in
Figure 6.2. When plans are implemented, monitoring of significant environmental
6 Environmental Issues 283

Site 1

Legend Site 2
Cities
State Capital
Federal Capital
Highway
Study Area
Boundary State

Land Use
Tropical Rainforest
Tropical Seasonal Forest
Seasonal Wetland Forest
Wetland Forest
Savanna
Farming Area
Urban Area Site 3

Figure 6.2 Transmission Line Corridor.

effects must be carried out to ensure that during the course of project development
unanticipated effects are identified and the appropriate measures are taken.
One of the main steps in the SEA methodology is the development of indicators
relevant to the selected objectives. Although many studies have used indicators,
those used for SEAs are often interrelated in order to evaluate the effects of the plan
or programme on the carrying capacities and sustainabilities of the resources found
in the specific area. Therefore, the list of indicators may present significant varia-
tions depending on the country or the type of plan or programme being studied.
Indicators can differ from country to country and may be different for generation
and transmission.

6.2.1.2 Description of the Cigré TB 487 (2012a), SEA for Power

Developments
Cigré TB 487, produced by WG C3.06 is divided into an Introduction and seven
sections. An Executive Summary synthesises the TB. The Introduction presents the
general concepts of SEA, the main characteristics of this type of environmental
study and highlights its strategic level.

Section 1: Presents the main environmental and SEA concepts and definitions.
Section 2: Presents a specific comparison of SEA and Environmental Impact
Assessment (EIA) studies, indicating the main differences between them.
284 C.Ó. Luain et al.

Section 3: Describes general examples of SEA application and examines its major
objectives, the reasons for developing SEA studies, existing legislation and the
main responsibilities and accountabilities.
Section 4: Summarises the specific application of SEA studies for generation and
transmission systems.
Section 5: Presents an overview of the indicators and other measurement techniques.
Section 6: Provides guidelines and recommendations for SEA in the power sector.
Section 7: Closes the TB with a brief conclusion.
Appendix 1: Presents a summary of SEA studies from each country participating in
the Working Group.
Appendix 2: Presents a list of acronyms and abbreviations.

6.2.1.3 Guidelines
• It is imperative to conduct an SEA for major Plans and Programmes
• In all cases, SEAs should be strategic and comprehensive studies
• SEA should be included in national and regional long-term plans both for gene-
ration and transmission
• Define specific objectives and the scope of the study on which the assessment
will be based and choose an appropriate method of measurement in accordance
with the environmental authorities
• Identify different alternatives and analyze their effects
• Include analysis of environmental, social, economic, sustainability, accumula-
tive effects and stakeholder needs
• Identify the stakeholders in the earlier stages of the process
• Take into account the results of public consultation
• SEA studies should result in meaningful deliverables, including proposals for
mitigation, compensation and monitoring measures
• Make the results of the study and public consultation available in a Final Report
• Conduct a monitoring phase to provide feedback and to identify the value of
lessons learned for the next SEA process.

6.2.1.4 Recommendations
• An appropriate communication plan should be established.
• Modelling tools should be developed to support the analysis of policies and
plans, as well as decision making.
• Geographical Information Systems (GIS) should be used, particularly to analyse
the cumulative and synergetic effects of the plan or programme.
• Build capacity and training on the SEA skills, which should be implemented.

6.2.2 Permit Procedures and Environmental Impact Assessment

6.2.2.1 Introduction
A survey was carried out by WG 14 on these issues (including others on OHLs
and Environmental Issues, see sections below) and was reported on in Cigré TB
6 Environmental Issues 285

147 High Voltage Overhead Lines - Environmental Concerns, Procedures,

Impacts and Mitigations (Oct 1999) (Cigre 1999). The relevant Section 2 of the
Cigré TB gave information on the regulations, procedures, appeal processes, ele-
ments of EIA and outline of EIA processes in a range of countries at that time.
Nevertheless it should be emphasised that national conditions vary considerably,
as may the regulatory controls and environmental impact assessment processes in
each country. Which procedures or practices to adopt therefore will depend to
some extent on prevailing regulatory regimes in each country and on their unique
cultural and environmental climates and it may depend also on the particular fea-
tures of each project.
It was noted that it was evident that regulatory controls and associated EIA requi-
rements were undergoing regular review and changing frequently. This is continu-
ing and the position for any particular country needs to be confirmed directly or by
way of more recent Cigré publications. Guidelines appropriate to the then situation
are given below and many are still applicable.

6.2.2.2 Guidelines
• Efficient management of the licensing process is vital. Unnecessary reiteration in
the licensing process should be avoided.
• It is desirable to develop a strategy which, while providing full justification for
the project and meeting all environmental and consultation requirements, also
anticipates that all decisions and procedures will be challenged.
• In the phase of recognition of the necessity and public interest of overhead line
projects it is not desirable to have an open ended process. Co-ordination of licen-
sing procedures should be undertaken to minimise delays in the whole process
• In the case of Overhead Lines with a national function a special regulatory pro-
cess may be beneficial.
• Appeals procedures should have fixed time spans. While processes vary consi-
derably and may involve operating through a number of levels it is essential for
utilities to take all precautions and perform all necessary checks to ensure that
any risk of losing the permit or licence is absolutely minimised.
• The use of environmental or community compensation schemes for communities
affected by Overhead Lines should be considered, particularly where communi-
ties perceive only impacts and no specific benefits to them from a project. Various
models exist which may prove appropriate depending on local circumstances.
• Models exist of changes to regulatory systems which involve earlier consulta-
tion, environmental compensation, commitments to further efforts to minimise
overhead line impacts - particularly in the context of overall network lengths and
development strategies - and design reviews. These may have relevance to evol-
ving situations elsewhere depending on the stage of development of networks.
• Environmental requirements should be made clear in advance (e.g. in the
Environmental Impact Assessment legislation) and confirmed in consultation
with authorities. Nevertheless new requirements may arise in relation to the
environmental impact studies or their methods of evaluation and it is desirable to
try to anticipate these in so far as possible.
286 C.Ó. Luain et al.

• Environmental specialists should be used to provide expert inputs on the over-

head line impacts relating to their particular areas of expertise for inclusion in
environmental impact statements. Various successful organisational frameworks
exist for assimilating these.
• Comprehensive Environmental Impact Assessments should be carried out inclu-
sive of all consultation requirements and environmental impact statements
should cover the full range of necessary subjects, description of impacts and
mitigation measures.

6.3 Environmental Impacts and Mitigations - Guidelines

TB 147 High Voltage Overhead Lines - Environmental Concerns, Procedures,

Impacts and Mitigations (Oct 1999). Cigre (1999) reviewed the different types of
impacts of Overhead Lines on the Environment under the heading listed below and
in Section 6.4 and reported on them and on mitigations measures.

6.3.1 Visual Impact

Assessment and routing including different methodologies for routing for minimal
visual impact were described. Routing guidelines were given and methods and tools
for visual impact assessment described. Overhead line designs for minimal visual
impact were dealt with including structures, conductors, insulators and fittings and
methods of camouflage of full overhead lines or the principle elements.
The use of compact lines and pole structures to minimise right of way and
impacts was outlined. The development of the “first generation” of aesthetic line
designs in a number of countries such as France and Finland, and other solutions to
reduce visual impact such as in Sweden (T Tower, see Figure 6.3) and in Hydro
Quebec and South Africa (Chainette Cross Rope) were described and examples
shown. Some of the aesthetic designs were developed through public competition.
Such aesthetic or “landscape tower” designs have continued to multiply.
TB 416 Innovative Solutions for OHL Supports (June 2010) (Cigre 2005c) deals
with the further development of such designs, be they intended for full lines routes,
segments of line routes or used as special structures in particular locations. It gives
designs from seventeen countries. Monopole supports are covered and more infor-
mation is given on the Finnish and French (see Figure 6.4) aesthetic designs menti-
oned in Cigré TB 147 and developments since in those countries. Quite a few
examples of the design of towers as artworks or transformation of some towers into
artworks are given. A new Danish 400 kV single pole design with a stainless steel
tube head framework supporting the insulator V assemblies in a delta formation is
shown and Icelandic studies and competition results outlined.
TB 416 concludes “Demand for electricity has grown dramatically and the need
for electricity will continue to grow. As a consequence, different regions of the
world will face different challenges, concerning the environmental impacts, to
6 Environmental Issues 287

Figure 6.3 Swedish

Reduced Visual Impact T
Tower, which also reduces
EMF.

supply more power. As far as transmission lines are concerned, some Utilities will
have to construct long (up to 2500 km) UHV OHLs in the near future. For such
lines, as reported in Cigré TB 416, the so called “aesthetic solutions” will be desi-
gned based on simplicity, invisibility, slenderness, compactness, all together driven
by costs. On the other hand, in other regions of the world, the construction of new
lines arouses more environmental and aesthetic concerns. With the growing
demand, there will be more requests for alternative design solutions, i.e., for
visually attractive landscape towers. In the majority of the cases, the desirable
solution is to hide the structures, making them invisible or camouflaged. When, for
any reason, this is not feasible, it is always possible to make them more aesthetic,
more beautiful”.

6.3.1.1 Guidelines on Visual Impact (Cigré TB 147, 1999)

• With regard to the visual impact of Overhead Lines the different approaches of
trying to integrate an overhead line within the landscape or to assert the line wit-
hin it may be used depending on the type of landscape, see Figure 6.5. The essen-
tial purpose of visual impact assessment is to describe the change in visual
quality associated with the landscape setting and this assessment may be carried
out by qualitative or quantitative methods.