Transmission Line Surge Arrester – White Paper

1010233

11178266
11178266
Transmission Line Surge Arrester –
White Paper

1010233

Technical Update, July 2005

EPRI Project Manager

A. Phillips

ELECTRIC POWER RESEARCH INSTITUTE

3420 Hillview Avenue, Palo Alto, California 94304-1395 ▪ PO Box 10412, Palo Alto, California 94303-0813 ▪ USA
800.313.3774 ▪ 650.855.2121 ▪ [email protected] ▪ www.epri.com

11178266
DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES
THIS DOCUMENT WAS PREPARED BY THE ORGANIZATION(S) NAMED BELOW AS AN
ACCOUNT OF WORK SPONSORED OR COSPONSORED BY THE ELECTRIC POWER RESEARCH
INSTITUTE, INC. (EPRI). NEITHER EPRI, ANY MEMBER OF EPRI, ANY COSPONSOR, THE
ORGANIZATION(S) BELOW, NOR ANY PERSON ACTING ON BEHALF OF ANY OF THEM:

(A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (I)

WITH RESPECT TO THE USE OF ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR
SIMILAR ITEM DISCLOSED IN THIS DOCUMENT, INCLUDING MERCHANTABILITY AND FITNESS
FOR A PARTICULAR PURPOSE, OR (II) THAT SUCH USE DOES NOT INFRINGE ON OR
INTERFERE WITH PRIVATELY OWNED RIGHTS, INCLUDING ANY PARTY'S INTELLECTUAL
PROPERTY, OR (III) THAT THIS DOCUMENT IS SUITABLE TO ANY PARTICULAR USER'S
CIRCUMSTANCE; OR

(B) ASSUMES RESPONSIBILITY FOR ANY DAMAGES OR OTHER LIABILITY WHATSOEVER

(INCLUDING ANY CONSEQUENTIAL DAMAGES, EVEN IF EPRI OR ANY EPRI REPRESENTATIVE
HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES) RESULTING FROM YOUR
SELECTION OR USE OF THIS DOCUMENT OR ANY INFORMATION, APPARATUS, METHOD,
PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS DOCUMENT.

ORGANIZATION(S) THAT PREPARED THIS DOCUMENT

K&R Consulting, LLC

ORDERING INFORMATION
Requests for copies of this report should be directed to EPRI Orders and Conferences, 1355 Willow
Way, Suite 278, Concord, CA 94520, (800) 313-3774, press 2 or internally x5379, (925) 609-9169,
(925) 609-1310 (fax).

Electric Power Research Institute and EPRI are registered service marks of the Electric Power
Research Institute, Inc.

Copyright © 2005 Electric Power Research Institute, Inc. All rights reserved.

11178266
CITATIONS

This report was prepared by

K&R Consulting, LLC

64 Sherwood Drive
Lenox, MA 01240

Principal Investigator
K. King

This report describes research sponsored by the Electric Power Research Institute (EPRI).

The report is a corporate document that should be cited in the literature in the following manner:

Transmission Line Surge Arrester – White Paper. EPRI, Palo Alto, CA: 2005. 1010233.

iii
11178266
11178266
PRODUCT DESCRIPTION

Electricity customers are continually demanding improved transmission system reliability and
performance. In order to improve reliability, utilities must focus on mitigating the effects of
lightning, which is often the most frequent cause of transmission outages and service
interruptions. The widespread use of electronic equipment at home and at work makes the
challenge of providing reliable electrical service greater today than ever before. Customers can
be adversely affected by even momentary outages or power quality problems; therefore limiting
the damage caused by lightning strikes must be a key goal for every utility. Improved
transmission line design, grounding, and proper application of transmission line surge arresters
(TLSA) are the best solutions for improving transmission line lightning performance.

Results and Findings

A summary of transmission line surge arrester operation, construction, and application is
provided along with data from all currently available transmission line surge arresters suppliers
in North America. The report also includes the results of an EPRI survey on transmission line
surge arrester use and experience in the United States.

Challenges and Objectives

This document is intended to assist members when selecting and applying transmission line
surge arrestors by providing up to date information on the state of the technology and current
user experience.

EPRI Perspective
This report is the initiating point for the EPRI development of a transmission line surge arrestor
(TLSA) application guide. The intended application guide will assist users in the selection,
application and inspection of TLSA and will address both mechanical and electrical issues.
Development of the guide is intended to start in 2006.

Approach
The development of this White Paper on Transmission Line Surge Arresters was based on
engineering standards review, consulting experience, manufactures sales literature, and the EPRI
Survey of 22 Utility Companies on their collective experience with Transmission Line Surge
Arresters. The issues identified in this white paper may be used to better focus future EPRI R&D
activities on the current needs of the member utilities.

v
11178266
Keywords
Lightning Protection
Surge Arrester
Outage Mitigation
Power Quality
Grounding
TLSA

vi
11178266
CONTENTS

1 LINE SURGE ARRESTER – WHITE PAPER ........................................................................1-1

Introduction ...........................................................................................................................1-1
Historical Development .........................................................................................................1-1
Types of Line Surge Arresters...............................................................................................1-3
Arrester Fundamentals..........................................................................................................1-3
TLSA Disconnects ............................................................................................................1-4
Voltage Ratings ................................................................................................................1-6
Protective Levels ..............................................................................................................1-7
Arrester Failure Modes..........................................................................................................1-8
Electrical Failures .............................................................................................................1-8
Mechanical Failures..........................................................................................................1-9
Arrester Application Considerations ....................................................................................1-15
Arrester Design Alternatives................................................................................................1-17
Wrapped Design.............................................................................................................1-17
Tube Design ...................................................................................................................1-19
Cage Design...................................................................................................................1-19
Current Manufacturers Offerings.........................................................................................1-21
Summary of Line Arrester Performance Experience...........................................................1-22

A REFERENCES...................................................................................................................... A-1

vii
11178266
11178266
LIST OF FIGURES

Figure 1-1 Typical line surge arrester construction of fiberglass around metal oxide
blocks with metal end fittings (on left, shown with weathershed removed) ........................1-4
Figure 1-2 Typical line surge arrester and disconnect assembly ...............................................1-5
Figure 1-3 Internal components of a typical disconnect assembly.............................................1-6
Figure 1-4 Voltage across typical 115 kV or 138 kV class line arresters in comparison to
typical insulator flashover levels.........................................................................................1-7
Figure 1-5 Angled arrester installation method that will clear itself from the circuit (via
gravity) in case of arrester failure.....................................................................................1-10
Figure 1-6 Arrester installation hanging from conductor ..........................................................1-11
Figure 1-7 Parallel hung arrester installation method that may cause repeated circuit
breaker operations if the arrester fails and later swings into the energized conductor
during poor weather .........................................................................................................1-12
Figure 1-8 Example of arrester installation method that may cause repeated circuit
breaker operations after arrester fails and later swings into the energized conductor
during light wind ...............................................................................................................1-13
Figure 1-9 Example of arrester mounting arrangement dead-end or post insulators
showing the disconnect mechanism on the energized end of the insulator
attachment point...............................................................................................................1-14
Figure 1-10 Typical gapped line surge arrester installation .....................................................1-15
Figure 1-11 Typical wrapped arrester design (from Ohio Brass sales literature).....................1-18
Figure 1-12 Typical tube arrester design (from NGK sales literature)......................................1-19
Figure 1-13 Typical cage design surge arrester (from Siemens sales literature).....................1-20
Figure 1-14 Area covered by EPRI TLSA survey (survey region shown in color)....................1-23

ix
11178266
11178266
LIST OF TABLES

Table 1-1 North America Line Surge Arrester Suppliers.........................................................1-21

xi
11178266
11178266
1
LINE SURGE ARRESTER – WHITE PAPER

Introduction

High voltage power systems are exposed to overvoltages which may exceed the dielectric
strength of the equipment on the system. These overvoltages may be caused by lightning
strokes, switching operations, unusual load flow conditions, and line-to-ground and phase-
to-phase faults. Depending on the source of the overvoltage, the magnitude and duration can
vary over a wide range. The power system equipment can effectively be protected from the high
magnitude overvoltages through the proper selection and installation of surge arresters.

Transmission Line Surge Arresters (TLSA) are designed to limit voltages between phase
conductors and the tower structure to prevent insulation flashover. TLSA will prevent lightning
related flashovers in both high footing resistance areas (backflash prevention), and poorly
shielded designs (shielding failure prevention) provided they are selected and located properly.
Reducing the ground impedance on a transmission line which is experiencing many shielding
failures will not help improve the lightning performance of the line. Shielding failure flashovers
can only be prevented by improving shielding, or installing TLSA. On transmission lines below
230 kV, TLSA have been applied in the US for many years with excellent results. Taking
corrective action to improve transmission line lightning performance without first determining
the types of flashovers can result in large expense with little improvement in performance.

Historical Development

Spark gaps were one early form of lightning protection equipment used on power systems. Gap
spacing could be sized so that the gap would spark before the line insulation voltage was
exceeded. Spark gap usage provided control over the flashover location and the lightning current
path to ground. Simple spark gaps have several shortcomings. The spark-over voltage is a
function of the length of the gap. Erosion of the electrodes from arcing and mechanical damage
changes the gap spacing, which, in turn, changes the spark-over voltage. Spark-over can also be
affected by the moisture content of the air and by pollution. Gas tubes were developed to solve
these problems on telegraph and telephone systems. The gas tube surge suppressor is simply a
spark gap enclosed in a sealed insulated container filled with an inert gas or air. The dielectric
properties of the gap are maintained by the gas, and the gap spacing is controlled by the
mechanical strength of the container.

The two primary design problems with lightning arresters were to control the voltage at which
the device operated and to limit the 60 Hz current subsequent to the flow of lightning currents.
The principal problem with spark gaps was the fact that the circuit had to be de-energized to
clear the gap. One attempt at limiting the circuit breaker operations to clear spark gaps was to
1-1
11178266
Line Surge Arrester – White Paper

use a fuse in series with the gap to clear the fault power current after the gap operated. This
approach was technically successful, but the fuse had to be replaced after each operation.

Building an effective lightning arrester requires that the operating voltage (spark-over) be
precisely controlled to insure that the arrester operate below the insulation level of protected
equipment. To effectively block the 60 Hz power frequency current after the lightning impulse
discharge, the arresters have a highly non-linear Voltage – Current relationship.

Continued research to develop a better lightning arrester for large power transformers led to the
development of the silicon-carbide valve arrester. The valve arrester uses a spark gap to control
spark-over voltage and a series valve element for current limitation. The valve element is a
nonlinear resistor whose resistance decreases as current increases. The silicon-carbide arresters
were quickly scaled down and applied to distribution systems. At first, the arresters used an
external spark gap. Soon, the gap was enclosed within the arrester housing, and the spark-over
characteristic was better controlled. In 1953, the first distribution surge arrester operating on the
magnetic gap principle and using silicon-carbide valve elements was introduced. Silicon-carbide
arresters continued to develop into the early 1970's with emphasis on improving spark-over and
discharge characteristics.

Most of the complexity in a silicon-carbide lightning arrester is in the gap structure used to
determine spark-over. The complexity of the gap structure impacts cost to manufacture and
reliability. To solve the steep wavefront problem, simplify the design, and improve reliability,
the lightning arrester gaps needed to be eliminated. The first gapless metal oxide varistor
(MOV) lightning arresters were built for substation protection of transformers.

The metal oxide varistor (MOV) lightning arrester eliminated the lightning arrester spark gap. A
characteristic of the MOV material is that it essentially does not conduct at normal line voltage.
At the surge overvoltage level, the MOV material goes smoothly into conduction and returns to a
non-conductive state when the voltage returns to normal levels. The volt-time characteristic of
the MOV arrester is only moderately affected by the rate of rise of the surge impulse. As with
the earlier arresters, the MOV arrester was scaled down and became available for distribution
protection.

Metal-oxide surge arresters, first developed in 1968, were introduced in the USA in 1977.
Because of concern for the stability and life of the metal oxide, these first station class arresters
contained gaps to reduce the normal power frequency voltage placed on the blocks.
Subsequently, the gaps were eliminated with improved block formulations, and the present
gapless arrester evolved.

Without a gap, the normal power-frequency voltage continuously appears across the metal oxide,
producing a few milliamperes of current. This low-magnitude current is not harmful. However,
higher currents resulting from power-frequency voltage excursions, or temporary overvoltages
(TOV) during faults or ferroresonance, will produce heating in the metal oxide. If the TOVs are
sufficiently large in magnitude and long in duration, temperatures may increase enough to cause
thermal run-away and an arrester failure.

1-2
11178266
 Line Surge Arrester – White Paper

Types of Line Surge Arresters

The most common type of line surge arresters in North America is a gapless design where the
line voltage is continuously applied to the arrester blocks. The arrester discharge voltage for this
arrangement is directly proportional to the height of the arrester block stack. Arrester discharge
voltage during a lightning transient event is also affected by the rate of rise of the current surge.
A faster wave front will have a slightly higher associated peak voltage while a slower wave front
will have a slightly lower associated peak voltage, as compared to a standard test wave. Typical
gapless line arrester currents are in the range of hundreds of micro amperes at normal power
system voltages.

Another type of line surge arrester includes an air gap in series with the metal oxide arrester
block stack. This type of gapped arrester design was common with older Silicon Carbide
arresters, which could not be continuously exposed to line voltage. A typical gapped line
arrester available today still uses metal oxide arrester blocks, but the air gap in series with the
arrester allows for the use of fewer blocks in series. The overall protective characteristics of a
gapped arrester are slightly different than a gapless design. An incoming voltage surge will have
to exceed a certain minimum value before the arrester gap sparks over and the arrester starts to
conduct surge current. Once the gap sparks over, the surge currents then flows through few
arrester blocks in series and the corresponding discharge voltage is less than the same discharge
current through a gapless design.

The air gap length for a gapped arrester is typically selected long enough to withstand more than
nominal line voltage. The gap allows the associated circuit breaker to operate momentarily and
re-close if there is an electrical failure of the arrester blocks. The gap also prevents the arrester
from conducting any current under normal operating conditions.

Arrester Fundamentals

A surge arrester impedance is extremely high under normal line voltage conditions and only a
small amount (tens to hundreds of micro amperes) of capacitive current will flow through an
arrester under normal conditions. When the voltage across an arrester is increased to 30% to
50% above nominal, the arrester impedance quickly drops to a very low value and large resistive
current will flow through the arrester. During this process the arrester effectively limits the
voltage across any object in parallel with the arrester (a line insulator or transformer winding)
and prevents insulator flashover. During the conduction process the arrester may absorb many
kilojoules of energy that must be dissipated through the arrester housing as excess heat.

The core of the arrester contains a large number of individual metal oxide blocks. The metal
oxide material is mainly composed of zinc oxide with additives to control arrester properties and
voltage – current characteristics. The overall voltage rating of the arrester and the voltage
discharge characteristics are directly proportional to the height of metal oxide disks in the core.
The diameter of the individual disks determines the energy handling characteristics of the overall
arrester. The diameter of line arrester disks is typically 40 mm to 60 mm. The diameter of
typical station class surge arrester disks is approximately 80mm to 90 mm. Surrounding the
metal oxide core is either a wrap of fiberglass cloth or a series of vertical fiberglass rods used to
provide mechanical strength to the disk stack. Surrounding the metal oxide and fiberglass core is

1-3
11178266
Line Surge Arrester – White Paper

a weather shed system that is very similar to non ceramic insulators (NCI). The weathershed
housing and end fitting seals are critical arresters components. The weathershed is essential for
maintaining the electrical insulating properties of the arrester at normal operating voltages. Just
as is the case with an NCI, if moisture penetrates the housing or the end fitting seals, arrester
failure is likely. The weathershed material is typically Silicone rubber, Ethylene–Propylene (EP)
rubber, or blend of the two. The metal end fittings and studs are used to connect external
hardware to the arrester for mounting and to contain the arrester materials. Figure 1-1 shows
typical line surge arrester components.

Figure 1-1
Typical line surge arrester construction of fiberglass around metal oxide blocks with metal
end fittings (on left, shown with weathershed removed)

TLSA Disconnects

When an arrester absorbs more energy than its ultimate capability, the arrester fails in a short
circuit mode. If the arrester was permanently attached to the line and failed in a short circuit, the
line circuit breaker would not be able to momentarily operate and then reclose to restore power
flow. To prevent line lock out after arrester failure, most arrester manufactures include a
frangible link, called a “disconnect”, inline with the arrester. The disconnect is designed to break
the arrester from the line if the arrester becomes electrically shorted. The disconnect does not

1-4
11178266
 Line Surge Arrester – White Paper

clear the fault current during the operation. The system circuit breaker must momentarily
operate to clear the fault and allow the arrester to swing away from the line. Then the circuit
breaker can reclose the line. On a line arrester with no disconnect, an external air gap is included
in the design. The air gap is sufficient to withstand normal power frequency voltage, and only
sparks over under overvoltage conditions. If a gapped arrester fails, the circuit breaker operates
each time an overvoltage sufficient to spark over the air gap occurs. Figures 1-1 and 1-2 show
typical TLSA disconnects.

Figure 1-2
Typical line surge arrester and disconnect assembly

1-5
11178266
Line Surge Arrester – White Paper

Figure 1-3
Internal components of a typical disconnect assembly

Voltage Ratings

The main voltage rating parameter for TLSA is Maximum Continuous Operating Voltage
(MCOV). As the name implies, the MCOV is the maximum line-to-ground, power-frequency
voltage (RMS) that can be continuously applied across the arrester. Voltages above the MCOV
will cause the arrester to change impedance and absorb excessive energy from the system.
Depending on the over voltage and length of time the voltage is applied, arrester life may be
shortened or the arrester may be completely destroyed. Arrester manufacturers specify both the
MCOV and the length of time voltages in excess of the MCOV may be applied without
damaging the arrester. Typical TLSA can withstand 150% of the MCOV for 5 seconds and
110% of the MCOV for 2000 hours with no loss of arrester life.

1-6
11178266
 Line Surge Arrester – White Paper

Protective Levels

The protective level of an arrester is the maximum crest voltage that appears across the arrester
terminals under specified test conditions. For gapless metal oxide arresters, the protective level
is the arrester voltage for a specified discharge current. IEEE Standard C62.11 specifies that
tests should be made for a range of typical lightning currents. Manufacturers publish the
lightning current protective levels for their arresters, and these data can be compared to the
withstand voltages of the line insulation to verify the line is not likely to flashover during a range
of lightning stroke conditions.

Figure 1-4 shows the voltage across two typical line surge arresters for a range of currents
through the arresters. The 84 kV MCOV and 98 kV MCOV arresters in this example are two
typical sizes that may be used on a 115 kV or 138 kV transmission line. As seen in this figure,
even for extremely large (40 kA) currents through the arresters, the voltage across the arresters is
far less than the typical insulation strength of a 7 bell or 8 bell insulator string. The currents in
this figure are not necessarily the same as the total lightning stroke current because a portion of
the stroke current will flow along the shield wire and down each of the tower grounds. Even for
a shielding failure there will be some current sharing between arresters installed on the same
phase on successive towers.

Arrester Discharge Voltage

700

7 A11 Insulator CFO = 618 kV & 8 A11 Insulator CFO = 707 kV

600

500
Voltage (kV)

98 kV MCOV
400

84 kV MCOV
300

200

100

0
0 5 10 15 20 25 30 35 40 45

Current (kA)

Figure 1-4
Voltage across typical 115 kV or 138 kV class line arresters in comparison to typical
insulator flashover levels

1-7
11178266
Line Surge Arrester – White Paper

Arrester Failure Modes

Electrical Failures

Typical line arrester currents below MCOV are in the range of hundreds of micro amperes. The
current magnitude at these low voltage levels is heavily dependent on the arrester temperature.
Any temperature rise increases the current at a given voltage. If the temperature of the arrester is
increased to a very high value, positive feedback will occur which will damage the arrester. As
temperature is increased, impedance decreases and current increases. If the heat generated by the
arrester current is greater than the heat dissipated by the arrester housing, the temperature of the
arrester will continue to increase. For this reason, thermal stability is one of the most important
application criteria for metal oxide surge arresters. When line arresters are exposed to energy
beyond their capability, the arrester will fail in a short circuit mode.

The energy capability of an arrester is affected by these thermal stability considerations. The
surge amplitude, duration, and total energy absorbed by the arrester have critical values, above
which thermal runaway of the arrester will occur. Typically, arrester MCOV is approximately
70% of the breakdown voltage for the arrester.

Surge arresters protect electrical equipment from damage by limiting overvoltages and
dissipating the associated energy. Energy handling capability of arresters is defined as the
amount of energy an arrester can absorb before the material fails. The three main electrical
failure modes of surge arresters are referred to as thermal runaway, puncture, and cracking.

Thermal runaway occurs if an arrester’s temperature is raised above the thermal stability limit,
and the energy absorbed by the arrester exceeds heat dissipated by the arrester housing.

In puncture, a small hole results from localized melting of the ZnO material when there is
excessive current concentration. Puncture failure has been studied extensively and at high
currents hot spot temperatures of 800 ºC can cause local melting.

Non-uniform current distribution can also cause non-uniform heating within the ceramic material
and if the thermal stress exceeds the strength of the metal oxide, the arrester block my crack.
Currents in the nonlinear portion of the V-I curve tend to concentrate into narrow paths within
the ZnO material.

The current density within the arrester cross section can determine the likely failure mechanism.
2
When the average current density is above 400 A/cm the arrester block is more likely to
thermally runaway rather than crack. The energy handling capability for thermal runaway and
the corresponding time to failure decrease more rapidly when the arrester current density exceeds
2
approximately 1500 A/cm because deep in the upturn region of the V-I curve, the voltage
gradient increases linearly with the current causing a rapid increase in the in the input power.
Therefore, less time is needed to heat up the disk to the stability limit temperature.

For the typical line arrester blocks, the energy absorption capability for puncture is higher than
those for thermal runaway and cracking. In contrast to station class arresters which have a larger
arrester block cross sectional area, puncture failures of typical 40 mm to 60 mm diameter line

1-8
11178266
 Line Surge Arrester – White Paper

arresters are rare. Both cracking and puncture are caused by localization of the arrester current,
which causes local heating leading to non uniform thermal expansion and thermal stresses.
Puncture and cracking do not occur when the arrester current is small, because the time duration
of the non uniform heating is slow enough for the temperature to equalize across the arrester
block.

Mechanical Failures

In addition to the previously discussed failure modes, there are a number of mechanical and
environmental stresses that may also lead to arrester failure. Housing damage during transport or
installation, or poor weathershed seal design, may allow moisture ingress and lead to flash under
along the fiberglass wrap or rods. Other failure modes include breaking of the disconnect
assembly, hardware, and connection lead failure. Depending on the installation orientation, the
arrester and associated hardware may be exposed to physical stresses due to conductor wind
motion or conductor vibration. Also, the disconnect assembly is much stronger in tension than in
shear. If possible during installation the disconnect should only be loaded in tension. For
example the disconnect in Figure 1-9 will likely last longer then the disconnect in Figure 1-5.

Although properly installed line arresters in North America have been remarkable reliable over
the last 15 years, an important design consideration is the installation orientation of the arrester
and the path the arrester will take if it does fail. Figures 1-5 through 1-9 show possible
installation methods for typical arrester configurations. In Figure 1-5, the arrester is installed at
an angle from the base of the Davit arm to the phase conductor. If an arrester in this orientation
fails and the disconnector, located between the arrester and the phase conductor, operates the
arrester will swing back down along the tower and will be clear of the phase conductor after the
breaker re-closes. In Figure 1-6, the arrester is suspended from the conductor. Depending on
whether the disconnect is installed on the line end or the grounded end of the arrester, the shorted
arrester may still be hanging from the conductor after the disconnect operates. This shorted
arrester then could swing into the tower or another phase and cause continued line faults until it
is removed or replaced.

In Figure 1-7, the arrester is installed vertically from the crossarm in parallel with the I string
insulator. In this case, if the arrester fails and the disconnect operates the short lead from the
arrester to the phase conductor will remain attached to the phase conductor and the arrester will
stay in its original orientation hanging in parallel with the insulator.

When a surge arresters fails it is permanently in a low impedance state and is effective a short
circuit. As shown in Figure 1-8, this installation method places a potential ground fault path
through the failed arrester in parallel with the insulator. The clearances in this case are probably
sufficient to withstand normal line voltage under calm conditions but any wind or swing of either
the conductor or the arrester could cause another ground fault and momentary circuit breaker
operation. This design could potentially cause multiple momentary breaker operations and may
cause the line to lock out. The failed arrester in this case would become a liability for the utility
and would probably have to be removed by a line crew shortly after it fails. A better installation
method would allow the utility to change the arrester at a later date and more convenient time.

1-9
11178266
Line Surge Arrester – White Paper

A method for attaching arresters under a dead-end or post insulator is shown in Figure 1-9. This
installation method allows the arrester to swing clear from the conductor in case of failure, and
also isolates the arrester from any mechanical damage caused by conductor vibration.
Figure 1-10 shows a typical mounting arrangement for a gapped line arrester on an I string
insulator.

Figure 1-5
Angled arrester installation method that will clear itself from the circuit (via gravity) in case
of arrester failure

1-10
11178266
 Line Surge Arrester – White Paper

Figure 1-6
Arrester installation hanging below the conductor

1-11
11178266
Line Surge Arrester – White Paper

Figure 1-7
Parallel arrester suspended from the crossarm may cause repeated circuit breaker
operations if the arrester fails and is later blown into the conductor

1-12
11178266
 Line Surge Arrester – White Paper

Figure 1-8
Example of a failed arrester that may cause repeated circuit breaker operations if the
arrester or conductor move in the wind

1-13
11178266
Line Surge Arrester – White Paper

Figure 1-9
Example of arrester mounting arrangement for dead-end or post insulators showing the
disconnect mechanism on the energized end of the insulator attachment point

1-14
11178266
 Line Surge Arrester – White Paper

Figure 1-10
Typical gapped line surge arrester installation

Arrester Application Considerations

Special care is necessary in applying line surge arresters on transmission lines. The use of surge
arresters instead of an overhead shield wire is not recommended when the dominant flashover
mechanism is caused by direct strokes (back flashover and shielding failures). When there is no
overhead shield wire, installing surge arresters at every insulator location will prevent most
flashovers, but the lack of a shield wire prevents multiple towers from sharing a portion of the
total lightning current. Individual surge arresters must carry much more of each lightning
strokes total current and consequently, must absorb much more energy than if a shield wire was
installed.

1-15
11178266
Line Surge Arrester – White Paper

Applying arresters to transmission towers to limit lightning flashovers is entirely different from,
and also more complex than, substation arrester applications. This type of arrester installation
should be handled with care if a major performance improvement is desired. For example, if the
arresters are applied only on certain phases and certain towers, the result will likely be that the
flashovers are transferred to adjacent unprotected towers. Careful analysis of arrester locations
and tower footing impedances, potential shield wire benefits, and arrester energy sharing is
recommended before any installation is made. Also the probability of arrester failures should be
added to estimates of overall line reliability. The EPRI TFlash program is designed to allow
engineers to examine all arrester options and their potential benefits to improving line
performance. The product ID for TFlash 4.0 software is 1011388. Eligible members can
download TFlash 4.1 from www.epri.com.

Installing arresters on every other structure, or some other further separation, can be an effective
way to reduce induced flashovers. Induced flashovers are the dominant flashover mechanism for
distribution lines because the distribution line is typically the same height, or maybe lower than
surrounding natural shielding (trees or buildings). Induced flashovers are not the dominant
flashover mechanism for transmission lines. Transmission line flashovers are typically caused
by direct strokes to the wires that result in either backflash or shielding failures. If induced
flashovers are not likely to occur on a line, then spreading arresters at periodic intervals can
actually worsen overall lightning performance by lowering the peak current required to cause a
flashover and transferring a problem from one location to the next unprotected insulator.

When an arrester is installed on a transmission line to prevent a backflashover it will start to

conduct a portion of the lightning current from the tower onto the protected phase conductor long
before the insulator string would normally have flashed over. The amount of current conducted
onto the phase conductor depends on the lightning stroke current and all the impedances of the
possible paths for the lightning current. When a lightning stroke occurs to a shielded line, some
lightning current will flow down the shield wire in each direction, and some will flow down the
tower through the tower surge impedance and the footing impedance at the base of the tower.
The lightning current flowing down the tower will develop voltages across the arrester that are
much higher than the arrester MCOV and the arrester will start to conduct some lightning current
onto the phase conductor. The problem occurs when the arrester has conducted enough lightning
current onto the phase conductor that a transient over voltage has developed on the phase
conductor larger than the nearby unprotected insulator CFO. The lightning current conducted
onto the phase wire through the arrester effectively looks like a smaller shielding failure current
to the next tower down the line.

If arresters are installed one some multiple of every other, or every third, or fourth, tower on a
transmission line, the result may be worse lightning performance for the line than if there were
no arresters installed at all. An engineer attempting to improve a power line’s lightning
performance must understand the types of lightning flashovers that are most likely to occur and
the impact any potential solutions may have elsewhere on the line or the power system. Also,
rules of thumb used for arrester placement on distribution systems are not appropriate for
transmission line lightning protection because the dominant methods of lightning flashover are
different between distribution lines and transmission lines.

The transferred flashover from the arrester protected insulator to the next unprotected insulator
along the line is even more of a problem when smaller arresters are used on transmission lines.

1-16
11178266
 Line Surge Arrester – White Paper

The smaller arresters have a lower MCOV, and consequently a lower “turn on voltage”, which
will conduct more lightning current onto a phase conductor during a stroke to the tower or shield
wire. The smaller arrester actually produces a large transferred apparent shielding failure current
to the next unprotected phase and will result in further decreased lightning performance for a
line.

Line surge arresters are most frequently installed to prevent lightning related flashovers of line
insulation and prevent circuit breaker momentary operations during the lightning transients.
Another very common source of transient overvoltage on a transmission line is switching surge.
Typical transmission switching surges have a lower peak magnitude than lightning transients, but
they waveshape is much longer. Switching surges can cause electrical failure of transmission
line arresters because of the long waveshape. In order to effectively control the switching
overvoltage and prevent excessive line arrester failures, the protective level of line surge
arresters and substation surge arresters must be coordinated. It is critically important to ensure
the larger diameter (higher energy capability) arrester blocks in the substation class surge
arresters respond to the switching surge over voltage and the line arresters do not absorb
switching energy. The voltage rating (MCOV) of station class arresters must be lower than the
voltage rating of the line arresters on the same circuit to ensure the substation arresters start to
conduct first and prevent the line arrester from absorbing switching energy. The higher voltage
lightning surges are still effectively absorbed by the line arresters and the associated line
insulation is protected from lightning flashover.

Arrester Design Alternatives

There are currently four different design alternatives for metal oxide surge arresters; porcelain
arresters, tube design polymer arresters, cage design polymer arresters and wrapped design
polymer arresters. The relatively large weight disadvantage of porcelain housed surge arresters
precludes there use to directly protect high voltage transmission line insulation. Hanging large
porcelain arresters directly on the conductor or on the tower cross arm is seldom practical above
distribution level voltages. This leaves the three polymer housed arrester designs as viable
alternatives for transmission line surge arresters.

Wrapped Design

The wrapped design polymer arrester consists of a fiber glass cloth, impregnated with an epoxy
resin, wound around the stack of metal oxide disks and the metal end fittings. This type of
design provides a low weight arrester with sufficient strength to support itself in nearly any
orientation. A wrapped design arrester should never be used to support the weight or force of
any other power system component or conductor.

There is a balanced design approach to the overall strength of the fiberglass wrap for this type of
arrester. In case of an arrester overload and subsequent short circuit through the failed metal
oxide material, the maximum pressure built up inside the arresters is determined by the
mechanical strength of the fiberglass wrap. If the wrap thickness is chosen too large, an
excessive pressure will build up inside the arrester before the housing is burst and the pressure is
released. If the wrap thickness is too thin, than the arrester may not be able to withstand typical

1-17
11178266
Line Surge Arrester – White Paper

installation and self supporting forces and may break and fall apart. Each arrester manufacturer
with a wrap design arrester has determined their own construction method and overall strength
for their fiberglass wrap. This type of construction provides the manufactures’ with an easily
scalable construction method whereby individual smaller arrester sections may be joined
together with double ended studs to create a larger arrester before the weather shed housing is
installed. The wrapped design construction is used by most of the arrester suppliers in North
America.

Figure 1-11
Typical wrapped arrester design (from Ohio Brass sales literature)

1-18
11178266
 Line Surge Arrester – White Paper

Tube Design

A tube design polymer arrester is similar to wrapped design arrester, such that the fiberglass tube
is placed around the stack of metal oxide disks and the metal end fittings. The mechanical
strength of the fiberglass tube determines the overall strength of the arrester. A similar balance
between mechanical strength and the ability to vent gas pressure during an arrester failure must
be selected when designing this type of arrester. The tube design construction is only used by
two suppliers in North America.

Figure 1-12
Typical tube arrester design (from NGK sales literature)

Cage Design

Instead of using a fiberglass wrap to support the stack of metal oxide disks, some manufactures
use a number of solid fiberglass rods (forming a “cage”) placed around the outside of the disk
stack. The arrester strength is derived from the clamping of each end of the fiberglass rods in a
large metal end fitting. The metal oxide elements have a very high compressive strength and the
fiberglass rods have a very high tensile strength. Combining these two opposite strengths in the
cage design produces an arrester that is much stronger mechanically than the wrap or tube
designs. The cage design can, therefore, support some mechanical load in addition to supporting
itself. Each manufacturer has their own mechanical specifications for each size and class of cage
design line arrester. The fiberglass cage around the metal oxide disks also provides some
protection from impact during transport and installation. The cage design allows for direct
venting of pressure built up during an arrester fault, because the entire circumference of the
arrester disks is not restrained by fiberglass as it is with the wrapped design. Due to the larger
structural end fittings of the cage design, this alternative is not modular as previously described
for the wrap design arrester. The cage design construction is only used by two suppliers in North
America.

1-19
11178266
Line Surge Arrester – White Paper

Figure 1-13
Typical cage design surge arrester (from Siemens sales literature)

1-20
11178266
 Line Surge Arrester – White Paper

Current Manufacturers Offerings

Table 1-1
North America Line Surge Arrester Suppliers

Manufacturer Product Maximum Gap or Arrester Weathershed Company Website

Tradename System Gapless Design Material
Voltage (Disconnect) Structure
(www.______)
Design

ABB Pexlink 170 kV Gapless Cage Silicone “abb.com”

Cooper UltraSil 145 kV Gapless Wrapped Silicone “cooperpower.com”

Joslyn ZQPT 170 kV Gapless Wrapped EP/Silicone “joslynmfg.com”

Blend

NGK None 500 kV Gap Tube Silicone “ngk.co.jp”

Ohio Brass ProtectaLite 250 kV Gapless Wrapped Alloy EP “hubblepowersystems.com”

Sediver None 300 kV Both Available Tube Silicone “sediver.fr”

Siemens 3EL2 340 kV Gapless Cage Silicone “siemens.com/arrester”

Tyco 2HSR 175 kV Gapless Wrapped Silicone “bowthorpe-emp.com”

1-21
11178266
Line Surge Arrester – White Paper

Summary of Line Arrester Performance Experience

A total of twenty two Utility companies in the United States participated in the 2005 EPRI
Survey on Transmission Line Surge Arrester use. EPRI would like to thank the following
companies for participating:

• Alabama Power

• Alabama Electric Coop (AEC)

• Alliant

• American Electric Power (AEP)

• Bonneville Power Administration (BPA)

• Central Hudson

• Central Power & Light (Texas)

• City Public Service of San Antonio (CPS)

• Consolidated Edison (ConEd)

• Entergy

• First Energy

• Florida Power and Light (FP&L)

• Georgia Power

• Georgia Transmission Corp (GTC)

• Hawaii Electric Light Company (HELCO)

• Hawaiian Electric Company (HECO)

• Mid American

• National Grid (USA)

• Northeast Utilities (NU)

1-22
11178266
 Line Surge Arrester – White Paper

• Public Service Electric and Gas (PSE&G)

• Public Service of New Mexico (PSNM)

• Tennessee Valley Authority (TVA)

Figure 1-14
Area covered by EPRI TLSA survey (survey region shown in color)

The following questions were asked during the telephone survey, which was conducted in the
spring and early summer of 2005:

1. Do you use transmission line surge arresters?

(Voltages, manufacturers, shielded/unshielded lines)

2. How do you determine arrester placement?

(High footing resistance areas, high fault areas, software models (TFlash, EMTP,
SigmaSLP, other program, simple patterns?)

3. Have you had any line arrester failures?

(Electrical or mechanical, on shielded or unshielded lines)

4. Do you use FALLS/NLDN data?

1-23
11178266
Line Surge Arrester – White Paper

5. Do you track line performance with a company wide database?

6. Do you have target footing impedance at each structure?

7. Do you have target performance levels (outages/ 100 miles / year)?

The individual company responses were recorded by hand during the telephone survey. The
following list provides a summary of the responses:

• Ten (45%) of the surveyed utilities used TLSA to improved lightning performance
– Arresters were in use on 46 kV, 69 kV, 115 kV, 138 kV, and 161 kV
– Arresters were used on both shielded and unshielded transmission lines

• Three companies (30%) placed arresters on every phase and every structure for some
lines

• Two companies (20%) focused arrester placement in high footing resistance areas

• Eight companies (80%) focused arrester placement in high fault areas

• Six companies (60%) used EPRI’s TFlash software to determine arrester placement

• One company (10%) used the Sigma SLP program to determine arrester placement

• Of the ten companies using line arresters, two did not know if they had experienced any
failures, the remaining eight all experienced some failures
– Most, but not all, of the electrical failures were on unshielded lines
– Most of the mechanical failures involved the disconnect breaking or failures of the
end studs on the arrester
– One company experienced an arrester breaking in half during installation

• Sixteen companies (73%) used the National Lightning Detection Network data to
correlate lightning outages

• Thirteen companies (59%) have specific targets for structure footing impedance for each
line voltage

• Five companies (23%) have specific flashover performance targets for each line voltage

1-24
11178266
A
REFERENCES

1. Handbook for Improving Overhead Transmission Line Lightning Performance, EPRI,

Palo Alto, CA 2004. #1002019

2. IEEE Standard C62.11-1999, IEEE Standard for Metal Oxide Surge Arresters for AC
Power Circuits (>1kV)

3. IEEE Standard C62.22-1997, IEEE Guide for the Application of Metal Oxide Surge
Arresters for AC Systems

4. IEC 60099-4, International Standard for Surge Arresters – Part 4, Metal Oxide Surge
Arresters without Gaps for AC Systems, 1998

5. IEC 60099-5, International Standard for Surge Arresters – Part 4, Selection and
Application Recommendations, 2000

A-1
11178266
11178266
11178266
Export Control Restrictions The Electric Power Research Institute (EPRI)
Access to and use of EPRI Intellectual Property is The Electric Power Research Institute (EPRI), with
granted with the specific understanding and major locations in Palo Alto, California, and Charlotte,
requirement that responsibility for ensuring full North Carolina, was established in 1973 as an
compliance with all applicable U.S. and foreign export independent, nonprofit center for public interest energy
laws and regulations is being undertaken by you and and environmental research. EPRI brings together
your company. This includes an obligation to ensure members, participants, the Institute’s scientists and
that any individual receiving access hereunder who is engineers, and other leading experts to work
not a U.S. citizen or permanent U.S. resident is collaboratively on solutions to the challenges of electric
permitted access under applicable U.S. and foreign power. These solutions span nearly every area of
export laws and regulations. In the event you are electricity generation, delivery, and use, including
uncertain whether you or your company may lawfully health, safety, and environment. EPRI’s members
obtain access to this EPRI Intellectual Property, you represent over 90% of the electricity generated in the
acknowledge that it is your obligation to consult with United States. International participation represents
your company’s legal counsel to determine whether nearly 15% of EPRI’s total research, development, and
this access is lawful. Although EPRI may make demonstration program.
available on a case-by-case basis an informal
assessment of the applicable U.S. export classification Together…Shaping the Future of Electricity
for specific EPRI Intellectual Property, you and your
company acknowledge that this assessment is solely
for informational purposes and not for reliance
purposes. You and your company acknowledge that it
is still the obligation of you and your company to make
your own assessment of the applicable U.S. export
classification and ensure compliance accordingly. You
and your company understand and acknowledge your
obligations to make a prompt report to EPRI and the
appropriate authorities regarding any access to or use
of EPRI Intellectual Property hereunder that may be in
violation of applicable U.S. or foreign export laws or
regulations.

© 2005 Electric Power Research Institute (EPRI), Inc. All rights

reserved. Electric Power Research Institute and EPRI are registered
service marks of the Electric Power Research Institute, Inc.

Printed on recycled paper in the United States of America 1010233

ELECTRIC POWER RESEARCH INSTITUTE

3420 Hillview Avenue, Palo Alto, California 94304-1395 • PO Box 10412, Palo Alto, California 94303-0813 • USA
800.313.3774 • 650.855.2121 • [email protected] • www.epri.com

11178266