Outline of Guide for Application of Transmission Line Surge Arresters 42 to 765 kV

4 TRANSMISSION LINE LIGHTNING PERFORMANCE PARAMETERS ...4-1Introduction ...4-1Lightning Incidence Parameters...4-2Ground Flash Density GFD...4-2Lightning Incidence to Lines ...4-4Lightni

Outline of Guide for Application of Transmission Line

Outline of Guide for Application of Transmission Line

Surge Arresters—42 to 765 kV

Extended Outline

1012313 Technical Update, October 2006

EPRI Project Manager

A Phillips

ELECTRIC POWER RESEARCH INSTITUTE

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

Kinectrics North America, Inc

CITATIONS

This report was prepared by

Kinectrics North America Inc

800 Kipling Avenue

Toronto, Ontario, Canada

Principal Investigator

W.A Chisholm

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

This publication is a corporate document that should be cited in the literature in the following manner:

Outline of Guide for Application of Transmission Line Surge Arresters—42 to 765 kV: Extended Outline EPRI, Palo Alto, CA: 2006 1012313

PRODUCT DESCRIPTION

Lightning flashovers are the most frequent cause of transmission line outages Transmission line surge arresters (TLSA) limit lightning overvoltages between phase conductors and towers, and thus eliminate most outages on protected structures This guide provides a tutorial on the relevant lightning phenomena, with an in-depth look at the operation, application, and placement of TLSA to maximize flashover protection and minimize capital investment The guide also

describes ways to improve tower grounding for better performance of overhead groundwires Results and Findings

The guide contains an in-depth description of the following areas:

• The parameters that influence transmission line lightning performance / parameters

Lightning incidence scales performance in all regions With overhead groundwires, lightning currents act against local soil resistivity to create insulator stress Adding arresters reduces the influence of grounding When OHGW are removed, leaving only arresters, the lightning charge replaces peak current as a dominant stress The section also discusses other

transmission line features that affect line lightning performance, including line, tower,

insulator and arrester air gap geometries; tower impedance; and nonlinear corona effects

• TLSA selection/specification Before selecting a TLSA, utility engineers should consider a number of design questions concerning arrester operating characteristics and rating,

temporary overvoltages, arrester protective levels and insulation coordination, TLSA energy capability, arrester failures, TLSA housings, and TLSA installation and handling Selecting

an arrester system (possibly including a series gap or insulator) for a particular transmission line is the process of simultaneously satisfying these concerns with a single arrester type

• Placement of arresters for improved lightning performance The efficient application of TLSA to improve line performance requires the investigation of all available mitigation options and weighing of the performance benefits against real cost Estimating the effects of changes in tower structure and design, shielding, grounding, and arresters on the lightning performance of transmission lines is crucial to this process This section discusses back-flashover protection, unshielded applications, and transmission lines over varying terrain Challenges and Objectives

One difficulty in focusing this report is the wide range of technical backgrounds of the readers Electrical engineers will be most interested in insulation coordination and risk management Civil engineers will be more interested in what will be gained and lost if a new line is designed without overhead groundwires (OHGW) and with compact insulation, protected by TLSA While

these readers will find what interests them, the main focus of this report is a utility project

manager facing a decision to replace existing overhead groundwires (OHGW), the

fastest-decaying transmission line component, with a typical life of 25 to 55 years What has changed in

ten years is that the decision to put up TLSA in place of OHGW is commercially and technically

viable in many areas Having this new alternative, with its reduced visual impact and peak-load loss reduction, can help the utility bottom line, especially considering that the OHGW

conductors represent 4% of the total line investment

Applications, Values, and Use

New lines with reduced visual impact are already taking advantage of TLSA to replace overhead groundwires So far, these applications have been made in areas of difficult grounding and low lightning incidence However, the alternative of buried transmission cable looms like the sword

of Damocles, motivating overhead line engineers to deliver more reliability with fewer resources EPRI Perspective

Lightning causes power outages that cost utilities more than $1 billion per year directly, in

damaged or destroyed equipment The indirect damage to customers from all power quality problems is estimated to exceed $100 billion per year, with more than half of these disturbances having lightning as a root cause This guide presents TLSA theory and design information to enable utilities to minimize the number outages due to lightning EPRI developed this guide with the understanding that users may not be familiar with either TLSA or the current standards that

do, or should, apply to them The guide is tutorial in nature and does not anticipate every

situation or utility need In general, however, experience has shown that properly designed, installed, inspected, and maintained hardware such as TLSA, counterpoise, and overhead

groundwires can significantly improve system reliability and power quality

Approach

In 1997, EPRI delivered a TLSA application guide (TR-108913), consolidating literature with results of a survey of 31 EPRI-member utilities EPRI also supported arrester energy and

mechanical tests at the EPRI Power Delivery Center-Lenox

The state of the art of TLSA has advanced considerably since the last EPRI guide was published Line arresters have proved themselves as technically and economically feasible for improving performance of conventional lines with overhead groundwires TLSA have also been used on 230-kV and 400-kV lines without overhead groundwires, where the extra surge duties raise new electrical reliability concerns Most of the difficulties found in applying TLSA relate to the spotty reliability of mechanical components This can be addressed by ensuring that TLSA components meet the same high reliability standards that apply to other line components

This document is an extended outline that will be built upon and refined over the new few years

to develop a completed guide

ABSTRACT

In most areas, flashovers from lightning are by far the most frequent cause of transmission line outages Transmission line surge arresters (TLSA) limit the lightning over-voltages between phase conductors and the tower structure This prevents flashovers and insulation damage on that structure This guide provides a tutorial of the relevant lightning phenomena, in general, and offers an in-depth look at the operation, application, and placement of TLSA to maximize

flashover protection and minimize capital investment

The guide also considers other mitigation measures, including improved tower grounding and the application of overhead groundwires EPRI developed this guide with the understanding that users may not be familiar with either TLSA or the current standards that do or should apply to them The guide, therefore, is tutorial and does not anticipate every situation or utility need Overall, however, experience has shown that properly designed, installed, inspected, and

maintained hardware such as TLSA, counterpoise, and overhead groundwires can significantly improve system reliability and power quality

This document is an extended outline that will be built upon and refined over the new few years

to develop a completed guide

4 TRANSMISSION LINE LIGHTNING PERFORMANCE PARAMETERS 4-1Introduction 4-1Lightning Incidence Parameters 4-2Ground Flash Density (GFD) 4-2Lightning Incidence to Lines 4-4Lightning Current Parameters 4-8Stroke Current Peak Magnitudes 4-9Stroke Current Rate of Rise 4-11Stroke Current Waveshapes 4-12Total Charge Delivered 4-13Number of Strokes in a Flash 4-14Transmission Line Parameters 4-14Line Conductor Geometries 4-15Tower Geometries 4-15Insulator / Air Gap Geometries 4-16Volt-Time Curve Penetration Method 4-18

x

Disruptive Index Method 4-18Leader Progression Method 4-19Tower Ground Characteristics 4-19Buried Tower Grillage 4-20Driven Ground Rods 4-20Counterpoise 4-20Transmission Line Surge Arresters (TLSA) 4-21Nonlinear Corona Effects 4-21

5 TLSA SELECTION AND SPECIFICATION 5-1Introduction 5-1Transmission Line Arresters 5-1Arrester Operating Characteristics 5-5Arrester Rating and MCOV 5-7Temporary Overvoltages 5-7Arrester Protective Levels and Insulation Coordination 5-8Lightning Insulation Coordination 5-9Switching Surge Insulation Coordination 5-11Energy Capability of TLSA 5-12Lightning Energy 5-12Switching Energy 5-13Arrester Failures 5-15Electrical Failure Modes 5-15Arrester Disconnects 5-16TLSA Housings 5-18TLSA Installation and Handling 5-19Handling and Installation Recommendations 5-20Handling 5-20Installation and Maintenance 5-20Arrester Markings 5-20

6 PLACEMENT OF ARRESTERS FOR IMPROVED LIGHTNING PERFORMANCE 6-1Economics 6-1Backflashover Protection 6-1Phase Location of TLSA 6-1

Coupling to Overhead Groundwires 6-1Crossarm Voltage 6-3TLSA Location for High Tower Footing Resistance 6-4Unshielded Applications 6-5Arrester Energy 6-6Vertical Circuits 6-6Horizontal Circuits 6-8Transmission Lines over Unchanging Terrain 6-9Compact Transmission Lines 6-11Compact Transmission Lines with Overhead Groundwires and TLSA 6-11Compact Unshielded Transmission Lines with TLSA 6-13

7 APPLICATION SOFTWARE 7-1Introduction 7-1Lightning Performance Design Workstation (LPDW) 7-1TFLASH Overview 7-1Building a TFLASH Model 7-2TFLASH Capabilities 7-2General Procedure for Constructing a Line Model 7-2Analyzing a TFLASH Model 7-5The Classical Solution - The Average Performance of the Line 7-5Oscillographs - Line Behavior for a User-Specified Stroke 7-6General Procedure / Sample Application 7-7

8 CASE STUDIES 8-1

44 kV Case Studies 8-1Comparison of OHGW versus TLSA using Customer Momentary Disturbance

Benchmark 8-1Application Experience with 44-kV Arrester Application 8-3

46 kV Case Studies 8-4

66 kV Case Study 8-4

69 kV Case Study 8-4

115 kV Vertical Line Case Study 8-5

115 kV Horizontal H-Frame Line Case Study 8-7

115 kV Horizontal Steel Lattice Line Case Study 8-8

xii

115-kV Horizontal Line: Predicted Outage Rate 8-8

115 kV Horizontal Line: Outage Rates with Partial TLSA Treatments 8-8

115 kV Horizontal Steel Lattice Line: Lessons Learned 8-9

138 kV Case Studies 8-10

154 kV Case Study 8-10

161 kV Case Studies 8-10

230 kV Case Studies 8-10

230 kV Horizontal Line: Application Experience 8-10

230 kV Circuit with 35-kV Underbuild 8-11

B TLSA ENERGY WITHSTAND TEST DATA B-1Definition of Withstand Criteria B-1Tests of 63-mm TLSA to Destruction B-1Tests of 8.4-kV MCOV Samples to Thermal Runaway B-1

C TRANSMISSION LINE LIGHTNING PERFORMANCE CASE STUDIES C-1

D MECHANICAL FORCE ANALYSIS FOR GAPLESS TLSA INSTALLATIONS D-1

LIST OF FIGURES

Figure 3-1 Power Quality Acceptability Curves Left: Computer Business Equipment

Manufacturers Association (CBEMA) 1996; Right: Information Technology Industry

Council (2000) 3-2Figure 3-2 Origins of Power Quality Disturbances < Substitute EPRI figure here > [Plata

2002] 3-2Figure 3-3 Power System Areas where Transmission Faults will Cause 50% and 70%

Sags <EPRI Ashok Sundaram> 3-3Figure 4-1 Lightning Ground Flash Density for Continental USA, 1989-1998 [Orville

Huffines] 4-3Figure 4-2 Lightning Ground Flash Density from NALDN, 2000 to 2003 (Vaisala, need

permission) 4-3Figure 4-3 Optical Transient Density Map from (NASA 2006) and Estimate of Ground

Flash Density 4-4Figure 4-4: Relation between Lateral Attractive Distance Da of Horizontal Conductor and

Average Conductor Height h Curve 1: Eriksson; Curve 2: D=2h; Curve 3: Rizk 4-5Figure 4-5 Striking Distances from Ground and Conductor to a Downward Leader 4-6Figure 4-6: Modeling of Lightning Shielding Failures using L2 Applet [Red 2005] for Peak

Stroke Currents of 5, 15 and 25 kA 4-7Figure 4-7 Relation between Lightning Leader Potential and Stroke Charge [Mazur 2001] 4-8Figure 4-8 Lightning to Instrumented Rods on Tokyo Electric Transmission Towers

[Takami 2005] 4-10Figure 4-9 Relation between Maximum Rate of Rise and Peak Amplitude of Lightning to

Tokyo Electric Transmission Towers [Takami 2006] 4-11Figure 4-10 Relation between “Virtual Front Time” and First Peak Amplitude for Lightning

to Tokyo Electric Towers [Takami 2006] 4-12Figure 4-11 Percentage of Positive Cloud-to-Ground Lightning Flashes (Left) and

Density of Large-Amplitude Positive Flashes (Right) in USA [Boccippio et al 4-14Figure 4-12 Flashover paths for a V-String Configuration 4-17Figure 5-1 Internal Construction of Silicon Carbide Lightning Surge Arrester 5-3Figure 5-2 General Electric 138-kV Gapped MOV TLSA in Virginia [Koch 1985, Zed

2004] 5-4Figure 5-3 Classification of Arrester Design Features [Richter et al 2004] 5-5Figure 5-4 TLSA Volt-Amp Curve < substitute a modern version > 5-6Figure 5-5 Typical Lightning Current Distribution on an Unshielded Transmission Line

with a Top-Phase Arrester having R=20 at 40 kA .5-11

using L-5 Applet and Step Waveshape (final to use CIGRE concave) 6-3Figure 6-2 Plot of Voltage versus Time at Various Points on Double-Circuit Tower,

Taking Into Account Relative Coupling from Shield Wires (final to use CIGRE

concave) 6-3Figure 6-3 Schematic of Traveling Waves Propagating Towards a Structure with Low

Footing Resistance: If Tower 3 has no arrester, it may flashover .6-5Figure 6-4 A Schematic showing the Shielding Angle on an Unshielded Transmission

Line with the Top Phase protected by TLSA 6-7Figure 6-5 Options for Improving Compact Line Lightning Performance 6-13Figure 6-6 Typical 115-kV Compact Line Geometry from 1980, using Polymer Post and

Semiconductive Glaze Bell Insulators [Ontario Hydro 1980] 6-15Figure 7-1 Tower Modeling Screen from EPRI TFLASH (dummy) 7-3Figure 7-2 Conductor Information Screen from TFLASH (dummy) 7-4Figure 8-1 Probability of Flashover on 44-kV Line in Delta Configuration with Overhead

Groundwire 8-2Figure 8-2 69-kV Line Configurations considered by TU Electric for Improved Lightning

Performance 8-4Figure 8-3 Application of TLSA on TU Electric 69-kV Lines [Sanders and Newman 1992] 8-5Figure 8-4 Voltage across 115-kV or 138-kV class TLSA compared to Insulator

Flashover Levels 8-5Figure 8-5 Strategy for Mounting TLSA with Suitable Mechanical Rating to Restrain

Conductor 8-6Figure 8-6 Mounting of TLSA with Insufficient Horizontal Clearance 8-7Figure 8-7 Single Circuit 115-kV Structure with Single OHGW Lightning Protection

[Tarasiewicz 2000] 8-8Figure 8-8 Typical 400-kV Line Geometries at Statnett (Norway) 8-12Figure 8-9 Compact 400-kV Unshielded Design with Top-Phase TLSA 8-12Figure 8-10 Traveling Waves near Open Terminal 8-13Figure 8-11 TLSA on AEP 765-kV Line for Switching Surge Control 8-14 Figure A-1 Shear and Tension Tests on TLSA Disconnects A-1Figure A-2 Arrester Body Bending Test Setup A-2Figure D-1 Example of Typical Conductor to Pole Suspension D-1Figure D-2 Example of Typical Conductorto Tower Mast Suspension D-2

LIST OF TABLES

Table 3-1 Typical Power Line Lightning Mitigation Options 3-4Table 4-1 Geometric and Contact Resistance for Typical Surface Electrodes 4-19Table 5-1 External Gap or MCOV Requirements for TLSA 5-21Table 6-1 Footing Resistance at Steel Lattice Towers along Hypothetical 138-kV

Transmission Line 6-4Table 6-2 Flash Incidence for 161 km (100 miles) of a Horizontal Circuit, 18 m (60 feet)

above Flat Terrain with Ground Flash Density of 3.9 per km2 per year 6-8Table 6-3 Flashover Data for 161 km (100 miles) of Unshielded Transmission Line for

Various TLSA Installations 6-10Table 6-4 Flashover Data for 161 km (100 miles) of Vertical Circuit, Steel Pole, Shielded

Transmission Line for Various TLSA Installations 6-11Table 6-5 Effects of Compact Line Insulation and Phase Spacing on Lightning

Performance (Outages per 100 km per year) 6-12Table 6-6 Options for Improving Compact Line Lightning Performance (Outages per 100

km per year) 6-13Table 8-1 Comparison of Costs for 44-kV Lightning Protection Options 8-3Table 8-2 Reduction in Total Lightning Outages for Nine Treatment Options 8-9 Table A-1 Observed TLSA Failure Loads A-2

1

PURPOSE

The purpose of this application guide is to help utilities use transmission line surge arresters (TLSA) for the reduction or prevention of outages caused by lightning on transmission lines that operate at system voltages up to and including 765 kV Other mitigation measures, such as improved tower grounding and the application of overhead groundwires, are considered in this guide

The guide was developed with the understanding that users may not be familiar with either TLSA or the current standards that do or should apply to them While it has the penalty of making this a long document, considerable tutorial material is included Even so, the guide cannot anticipate every situation or utility need The user must consider the particular

requirements of a given application and select those criteria that fit that application In certain situations the user may want to develop additional criteria to address a particular use

(Description of target audience – I want to help an electrical engineer to ask the right questions and make the right decisions when she receives a report that the overhead groundwires on a line have reached their end-of-life This help includes the technical vocabulary and basic

understanding of mechanical issues such as vibration, wind load and conductor restraint force needed to work efficiently with mechanical engineers doing tower head layouts.)

2

DEFINITIONS

Arc A continuous luminous discharge of electricity across an insulating

medium, usually accompanied by the partial volatilization of the electrodes

a network or electric installation that is normally at ground potential

Basic Lightning Impulse

Insulation Level (BIL)

For self-restoring insulation such as air The crest value of a standard 1.2/50 µs lightning impulse for which the insulation exhibits a 10% chance of flashover under standard conditions

Basic Switching Impulse

Insulation Level (BSL)

For self-restoring insulation such as air The crest value of a standard 250/2500 µs switching impulse for which the insulation exhibits a 10% chance of flashover under standard conditions

tension (International Conference on Large High Voltage Electric Systems) CIGRE is an international technical organization, similar

to the IEEE Power Engineering Society, which focuses primarily on transmission voltage systems CIGRE holds a general conference in Paris every two years The various study committees meet more frequently at other locations CIGRE'S official publication is

"Electra."

milliseconds) current that flows between strokes of a lightning flash, with moderate return-stroke channel luminosity and significant transfer of charge from cloud to ground

Definitions

mechanical loads and restrains the metal oxide valve blocks

conductor caused by a voltage gradient exceeding a certain critical value

50% probability of flashover when applied to a particular conductor and location

Critical Flashover

Voltage (CFO)

The amplitude of the voltage of a given waveshape that, under specified conditions, causes flashover through the surrounding medium on 50% of the voltage applications

prevent a permanent fault on the circuit It also provides a visual indication of a failed arrester

leader to lines, objects or ground based on electrostatic estimates of relations among leader potential, charge and current Striking distances in this model become functions of the peak magnitude of the first stroke current

electric charges at rest (with no time variation) in the frame of reference

Erosion is nonconductive and can be uniform, localized or shaped Shallow surface traces can occur on insulator surfaces after arcing

tree-Ethylene Propylene

Diene Monomer (EPDM)

A base polymer previously used in rubber housings for insulators and arrester housings, now alloyed with silicone for hydrophobicity

dielectric strength of external insulation depends on atmospheric conditions

2-2

Definitions

solid or liquid insulation, between parts of different potential or polarity, produced by the application of voltage wherein the breakdown path becomes sufficiently ionized to maintain an electric arc

that, when altered in magnitude, a voltage is induced in an electric circuit linked with the flux

straight line through two points on the front of the wave One point

is located at 90% of the crest value; the other point is either 30% or 10% of the crest value The front time is the first number in the description of a wave shape, i.e., 8 in a wave shape described as 8/20 µs

which an electric circuit or equipment is connected to the earth or to some conducting body of relatively large extent that serves in place

weather and may be equipped with weather sheds In some designs the housing may also include insulating materials between the weather sheds ad the core

(in liquid or vapor form) that can lead to electrical and mechanical degradation

Induction Field

(Magnetostatic Field)

The electric and magnetic fields created by a constant current

conductors or equipment and to insulate these conductors or equipment from ground or from other conductors or equipment An insulator comprises one or more insulating parts to which

connecting devices (metal fittings) are often permanently attached

Internal insulation is usually not self-restoring

Definitions

Insulator Arcing Horn /

Insulator Arcing Ring

A metal part, usually shaped like a (Horn / Ring), placed at one or both ends of an insulator or a string of insulators to establish an arc- over path, thereby reducing or eliminating damage by arc-over to the insulator or conductor or both

expressed in days per year (TD) Used in the past to estimate lightning ground flash density but no longer recommended for this purpose, as the relation between KL and GFD varies with location

from a cloud followed by one or more return strokes

Lightning impulse

protective level

The maximum lightning impulse voltage expected at the terminals

of a surge protective device under specified conditions of operation The lightning impulse protective levels are given by: a) Front-of-wave impulse sparkover or discharge voltage, and b) the higher of either a 1.2/50 impulse sparkover voltage or the discharge voltage for a specified current magnitude and waveshape

downward stepped leader meets an upward leader from the earth

Maximum Continuous

Operating Voltage

(MCOV)

The maximum line-to-ground power frequency voltage (RMS) that

is specified by the manufacturer

Metal Oxide Varistor

(MOV)

Zinc Oxide, sintered with a number of other metal elements to give

a highly non-linear semiconducting material used in modern surge arresters In contrast to the older silicon carbide arresters, MOV arresters do not require spark gaps but can benefit from them

purpose of intercepting direct strokes in order to project the phase conductors from direct strokes They may be grounded directly or indirectly through short gaps

impedance, travels back down the line towards the source The voltage reflection coefficient is ρv = (Zb-Za)/(Zb+Za) where Za is the initial impedance and Zb is the new impedance encountered

2-4

Definitions

surge impedance, travels on in the new impedance The refraction coefficient for voltage is (2Zb)/(Zb+Za) where Za is the initial impedance and Zb is the new impedance encountered

stepped leader and pilot streamer have established a highly-ionized path between charge centers Lightning current flow removes the charge deposited by the stepped leader along the stroke channel

Self-Restoring Insulation Insulation, such as porcelain or non-ceramic insulators, that is not

damaged by flashovers and soon regains all or most of its insulation strength after a flashover event

the transmission line and another conducting electrode that is connected to ground, or to the high-voltage terminal of a line surge arrester Spark gaps were the first lightning protection devices used

on transmission lines and active spark gaps, with series nonlinear elements, are still an important surge proactive device technology

Standard Impulse

Voltage Wave

A voltage waveshape with a front time of 1.2µs and a time to half value of 50 µs that is used to test insulation in the laboratory Testing standards such as IEEE Standard 4 describe the allowable tolerances on this waveshape

magnitudes that are associated with stepped leaders are small (on the order of 100 A) in comparison with the return stroke current The stepped leaders progress in a random direction in discrete time steps of 10 to 80 m in length Their most frequent velocity of propagation is 0.05% of the speed of light It is not until the stepped leader is within the striking distance of the point to be struck that the stepped leader is positively directed toward this point

Strokes typically last less than 100 µs Each component stroke of a flash is separated by several tens of milliseconds In many cases, a small continuing current flows between strokes

Definitions

losses, Z = √(L/c)w, here L and C are the inductance and capacitance per unit length for a transmission line or cable For lumped circuits, L and C are the total equivalent inductance and capacitance

Transmission Line Surge

Arrester (TLSA)

Arresters designed specifically for application on transmission lines

to prevent line insulation flashovers They may be gapped or gapless

such as a transmission tower, skyscraper, or mountain-top Most stepped leaders are downward traveling Downward leaders tend to have higher potential and charge, both of which lead to higher first return stroke peak current magnitudes

to the time of chopping, which may occur on the front, at the crest

or on the tail The curve is obtained by applying impulse voltages

of constant shape, but with different peak values

a circuit Neglecting losses, v = √(1/LC), where L and C are the inductance and capacitance per unit length for a transmission line or cable

insulating surfaces (leakage distance) between the conductive parts

of an insulator Weathersheds also provide protected bottom surfaces that tend to stay dry in wet weather, further improving electrical flashover performance

2-6

3

WHY PROTECT TRANSMISSION LINES FROM

LIGHTNING

Economic Impact of Power Quality Problems

Citation from 2001 EPRI report, noting “power disturbance problems cost the US economy between $119B and $188B per year” [IEEE Spectrum Jan 2006]

Classification of Power Quality Problems

Standards have been developed by several interested parties to negotiate what constitutes

acceptable power, through the use of time-duration curves of disturbances Important standards have been recommended by:

• Computer Business Equipment Manufacturing Association (CBEMA)

• Information Technology Information Council (ITIC)

• Semiconductor Equipment and Materials International (SEMI-47)

• American National Standards Institute (ANSI) C84.1

Customer computer equipment used to meet the CBEMA standard for power quality, shown in Figure 3-1 Ride-through for short-duration sags (70% voltage dip for 100 ms) in the 1970s was provided by rotating machines or other locally stored energy for mainframe computers Linear power supplies with large electrolytic capacitors provided this function in electronic equipment

of the period

Revisions to the CBEMA curve were made by its replacement, the Information Technology Industry Council (ITIC), starting in 1996 and adopted in 2000 Figure 3-1 shows that the general nature is the same, but there are differences in detail The ITIC:

• Raised the tolerance level for short-duration overvoltages, because these are easy to eliminate with surge protective devices inside the equipment

• Reduced the tolerance level for short-duration voltage sags, because providing additional energy storage in typical switching power supplies adds cost, consumes more power

Why Protect Transmission Lines from Lightning

-100 -50 0 50 100 150 200 250

ACCEPTABLE POWER

10% +

ACCEPTABLE

POWER

Figure 3-1

Power Quality Acceptability Curves Left: Computer Business Equipment Manufacturers

Association (CBEMA) 1996; Right: Information Technology Industry Council (2000)

Figure 3-2 classifies typical voltage dip – duration curves for six different power quality

disturbance root causes Nearby Distribution Faults are classed as under-voltage conditions in

both the CBEMA and ITIC curves Fuse Operations sit right on the CBEMA curve of

acceptable power, but are classed as unacceptable in the ITIC graph

Transmission faults lead to short-duration voltage dips of three to ten ac cycles (50 to 200 ms)

that affect a large number of customers at once For the Transmission Faults in Figure 3-2:

• 15% are classified as under-voltage conditions in the CBEMA curve

• 35% are classified as under-voltage conditions in the ITIC curve

Figure 3-2

Origins of Power Quality Disturbances < Substitute EPRI figure here > [Plata 2002]

3-2

Why Protect Transmission Lines from Lightning

Lightning as a Root Cause of Short-Duration Faults

EPRI has made extensive simulations of power system disturbances, leading for example to the area map in Figure 3-3 where transmission faults will cause unacceptable 70% voltage dips

trans-If the transmission lines are completely exposed to lightning, like distribution lines, then every flash would cause a flashover Protective relaying would operate each time, leading to 120 unacceptable voltage dips at the customer load every year These would not be spread evenly

over the year, however – about 80 of the 120 would occur in July and August

Normal practice would be to protect the transmission lines with overhead groundwires (OHGW) These will steer lightning away from the phase conductors and lead it safely to ground The combination of OHGW, high insulation strength and good grounding from wide-base towers is highly effective, especially in the central US where the soil resistivity is low

With OHGW and good grounding, the lightning outage rate for a typical transmission line will

be less than 1 tripout per 100 km per year For the sensitive area in Figure 3-3, this works out to (190 km x 1 tripout per 100 km per year) or an average of 1.9 voltage dips every year Put into

other terms, 2 of the 120 flashes will cause faults on the protected line, an efficiency of 98.3%

Why Protect Transmission Lines from Lightning

Typical Power Line Lightning Mitigation Options

The consequences of a lightning fault depend to a large extent on the operating voltage of the

line Figure 3-3 has shown that there will be a region around the fault where customers may

record an unacceptable voltage dip The number of customers in the affected area will scale

somewhere between linearly and quadratically with the system voltage, as higher-voltage lines tend to use larger conductors or bundles

The efficiency of various types of lightning protection also varies with operating voltage High system voltage usually calls for larger electrical clearances, and lightning impulse strength scales linearly at about 500-540 kV per meter of clearance As an additional factor, however, the

insulation requirements for uniform lightning performance also vary with local soil type

150 to 550 kV 0.3 to 1 m

550 to 1100 kV

1 to 2 m

1100 to 2200 kV

2 to 4 m Induced

Overvoltages 30% extra stress Add TLSA**, or 300-400 kV CFO Induced flashovers not likely

Shielding

Failures No effect

Add TLSA to top and poorly protected phases

Add second overhead groundwire, and/or move them outboard Add TLSA to poorly protected phases

Backflashover 9tower with poor soil x more stress on

Add TLSA to top phase(s)

Add TLSA to bottom phase(s)

Improve grounding Add TLSA on bottom phases

Improve grounding Add gapped TLSA

* Poor soil = 1000 Ωm where 100 Ωm is typical; ** TLSA = Transmission Line Surge Arrester

Utility Investment in Lightning Protection using Overhead Groundwires

Overhead groundwires (OHGW) in Table 3-1 are used for systems operating above 100 kV to provide effective lightning protection over a long service life Simple visual cues such as broken strands or missing wire signal the end of OHGW service life However, they do not carry any power – in fact, they dissipate power at peak loads because nearby phase conductors induce

circulating currents One evaluation of the value of OHGW protection was carried out [Red

2006] using a benchmark of the cost per avoided customer momentary disturbance Using the example in Figure 3-3, this benchmark is:

The cost of avoiding 118 of the 120 unacceptable voltage sags using OHGW (The number of affected customers in the area) times (118 avoided disturbances)

3-4

Why Protect Transmission Lines from Lightning

A rough guide is that the OHGW protection adds a total of 10% to the line cost This breaks down as 4% for the extra conductors (there are usually two), 3% for stronger towers and 3% to make up generation capacity for the peak-power losses Using typical transmission line cost of

$US 100k per mile, the utility capital investment in OHGW in the sensitive area of Figure 3-3 works out to $1.2 million With about 100,000 customers in the area, the benchmark works out

to about 10¢ per avoided customer momentary disturbance

Depending on the system voltage, this value was found to range from 1¢ to 14¢ in (Red 2006) The value is lower for areas of dense lightning and for EHV lines that take advantage of low-reactance phasing to limit induced-current power consumption at peak loads

Utility Investment in Other Lightning Protection Methods

Utilities have other options to OHGW, and have used them with moderate success in areas of minimal lightning like California, British Columbia, New Brunswick, Newfoundland and

Quebec Some utilities reduce their sensitivity to single-phase to ground flashovers by providing single-pole reclosing Others provide redundant transmission paths However, high rates of multiple-pole flashover from lightning have forced some of these utilities to retrofit protective measures, such as transmission line surge arresters (TLSA)

This guide covers two important applications of TLSA as lightning protection:

1 To improve the efficiency of OHGW protection If the OHGW is reaching 98.3%, there is not much more to be gained However, if the ground resistivity is high, OHGW efficiency can fall off to 30-40% Spot treatment of high-resistance towers (and their neighbors) is also

1 Those that govern lightning incidence to a line

2 Those that govern the development of insulator and air gap voltages when lightning hits a line

or hits the earth near a line

The designer can substantially improve line lightning performance by paying careful attention to the parameters in both categories

It should be recognized that lightning flashovers are meteorological phenomena that can vary widely from year to year As such, it is easy to understand that 100% accuracy in forecasting flashover rates is no more possible than 100% accuracy in long-range weather forecasting [r1], except for the situation where TLSA suppress nearly all flashovers that would otherwise

normally occur It should also be noted that the number of lightning flashovers can vary widely from one year to the next One should not be surprised at a variation of 3:1, and sometimes substantially more than that

Lightning performance calculations are useful for the comparison of line designs in the same meteorological environment The designer can compare design options, understanding that while the absolute performance of the line may be uncertain, the design is the best option

evaluated based on performance requirements, economic considerations, soil conditions and terrain, and the estimated average ground flash density (GFD) where the line is to be located

It is possible to perform some limited lightning performance calculations manually, but the evaluation of multiple design options generally requires the use of a high-speed computer

Transmission line lightning computer programs forecast an average expected flashover rate by first assuming a prescribed average lightning incidence to a line that is based on average or median thunderstorm weather records Some programs also provide estimates of return periods, i.e., probabilities that flashover rate X will be exceeded only once in Y years, based on statistical distributions of lightning currents and GFD statistics Transmission line lightning programs

Transmission Line Lightning Performance Parameters

perform these calculations using concepts outlined in this chapter Application software is discussed in more detail in Section 7

Lightning Incidence Parameters

"Lightning incidence" is a generic term relating to the likelihood of lightning hitting a line over a designated period of time To assess line performance, engineers usually refer to the number of flashes to a line per 100 miles, or 100 kilometers, per year With this information they can use modern analytical procedures to determine the number of flashes to each overhead groundwire and phase conductor These values are fixed by GFD and the "electrogeometric" twists and contortions of each lightning leader as it approaches the line or the earth

Ground Flash Density (GFD)

Every transmission line is located in a specific meteorological environment Therefore, a key lightning incidence parameter is the average number of flashes to earth per square kilometer, per year along the line corridor This parameter, called the ground flash density (GFD), is

determined by averaging years of ground flash counts recorded by electronic locating systems This technology has been operated for fifteen years or more, giving satisfactory average values

of GFD even in areas of North America where thunderstorms vary substantially from year to year and the density is low Figure 4-1 shows a ten-year average GFD for the USA

4-2

Transmission Line Lightning Performance Parameters

Figure 4-1

Lightning Ground Flash Density for Continental USA, 1989-1998 [Orville Huffines]

Figure 4-2

Lightning Ground Flash Density from NALDN, 2000 to 2003 (Vaisala, need permission)

For those regions where the GFD is unknown or of dubious accuracy, engineers ten years ago used the local "keraunic level" as an alternative means of estimating the GFD The keraunic level

is defined as the average number of days per year at a specific location on the ground that

someone can expect to hear thunder Weather bureaus in many countries keep yearly records of the keraunic level at their meteorological observation points From these records, national or international isokeraunic maps can be prepared to show contours of constant keraunic level

A rough relationship between local GFD and keraunic level is given by:

1 1 25

/ 1

054.004

GFD = =

Equation 4-1: Ground Flash Density from Annual Thunder-Days or Thunder-Hours

Where: GFD = average flashes to earth/km2/year, Td = average thunder days per year (keraunic level) and Th = average thunder hours per year (keraunic level)

This relationship, recommended by IEEE [2] and CIGRE [3] ten years ago, assumes that the ratio of the number of cloud-to-cloud flashes to the number of cloud-to-ground flashes is the same in both tropical and temperate zones There is considerable evidence that this is not the

Transmission Line Lightning Performance Parameters

case, but Equation 4-1 should be roughly correct for temperate zones Other possible

relationships have been tabulated in [Red Book 1981]

In tropical areas with more than 100 thunder-days per year, estimates of GFD using Equation 4-1

Figure 4-3

nsient Density Map from (NASA 2006) and Estimate of Ground Flash Density

Lightning Incidence to Lines

From the regional value of GFD, the approximate number of flashes per year collected by a

Equation 4-2: Eriksson Expression for Number of Flashes to Transmission Line

Where N s = number of flashes to a line 100 km/yr, GFD = ground flash density (flashes/km2/yr),

h t = overhead groundwire height at the tower (m) and b = horizontal separation distance between

overhead groundwires (m)

4-4

Transmission Line Lightning Performance Parameters

Eriksson’s expression in Equation 4-2 uses the overhead groundwire height at the tower, h t, which can be confusing Most other expressions use the average height of the wire over ground

h given by this height at the tower minus two-thirds of the sag, assuming that the terrain is flat

Other equations for h have been proposed for rolling and mountainous terrain If no overhead

int

groundwires exist, h in Equation 4-2 becomes the average height of conductor attachment po

at the tower, and b is the distance between the outmost phases If only one shield wire exists, b is

also zero A preferred expression for average flash incidence as a function of first peak return

stroke current I and average conductor height h is:

( I h b)

GFD

N s = ⋅ 0 69⋅ 0 45 +

14.310

Equation 4-3: Preferred Expression for Lightning Flash Incidence to Power Lines

This expression gives similar numerical estimates to the Equation 4-2, as shown in Figure 4-4, and has a much stronger basis in the physics of switching-surge gap flashover

Figure 4-4: Relation between Lateral Attractive Distance Da of Horizontal Conductor and

Average Conductor Height h Curve 1: Eriksson; Curve 2: D=2h; Curve 3: Rizk

Flashes to a line terminate either on one, of the overhead groundwires (if any exist) or on one of

very high insulation strength, and the first stroke is weak, one of the subsequent strokes is likely phases A strike to a phase, known as a "shielding failure”, will usually cause flashover lation at one or more towers which may involve one or more phases Even if th

Transmission Line Lightning Performance Parameters

to have enough energy to cause flashover This means that every shielding failure can be

considered to cause a shielding failure flashover

The "electrogeometric theory" of shielding failures is reviewed in [Red Books, IEEE 1243] The model is based on an assumed perfect correlation between leader charge and flash current stepped downward leader of a flash approaches a li

As a

ne, the last step has a choice of striking the earth or jumping to a shield wire or phase wire (Figure 4-2) The "striking distance" to a shield

m Lines to Vertical Leader (IEEE 1243)

The

wire or phase wire is approximated by Equation 4.3

65 0

10I

r c =

Equation 4-4: Recommended Striking Distance fro

)43ln(

7.16

Equation 4-5: Recommended Striking Distance from Earth to Vertical Leader (IEEE 1243)

where: r g= striking distance to earth (m) and h = average conductor height, < 40 m

Transmission Line Lightning Performance Parameters

Further research has adjusted some of the electrogeometric relationships, with a view towards

accurately The expression in Equation 4-3 is one model that accomplishes this

mines the

,

ell-eloping a unified model that predicts both the flash incidence and the sh

The flash termination point depends not only on the location of its tip as it approaches the line

but also on the current, I, that will be delivered by the first stroke in the flash This current is

proportional to the leader charge near the leader tip, and it is this charge that deter

striking distances r c and r g The final breakdown of the air is assumed to occur over the shortest distance and this determines the anchoring point for the stroke current Electrogeometric theory

as outlined in [Red Books] or [Rizk 1990], can be applied over the expected wide range of

lighting amplitudes (from 3 to 200 kA), and the results integrated to estimate the number of expected hits to each shield or phase wire Alternately, numerical simulations can be

performed.An applet (LI2) provided with this guide can be used to explore the various models of shielding failure Figure 4-6 illustrates the occurrence of a few shielding failures, even for a wshielded double circuit line, when some randomness is introduced into the leader propagation

P(I≤5 kA) = 1%

Transmission Line Lightning Performance Parameters

P(I≤15 kA) = 13%

P(I≤25 kA) = 36%

Figure 4-6: Modeling of Lightning Shielding Failures using L2 Applet [Red 2005] for Peak Stroke Currents of 5, 15 and 25 kA

Lightning Current Parameters

For computational purposes, a lightning flash to a line is idealized as a vertical,

infinite-impedance, surge current source The actual surge impedance of a flash channel is still a matter

of discussion after a century of lightning research Values from 300 Ω to infinity have been used In principle, the flash acts like a single-conductor lossy transmission line that has been lowered from an overhead thundercloud into the vicinity of a line, charged to extremely high

voltage and then suddenly connected to the line Estimates of the leader potential V can be plotted against leader charge Q, both obtained from multiple-point synchronized measurements

of electric fields at ground level, as shown in Figure 4-7 The relation C Leader =Q/V then gives an

estimate that these vertical charged rods with their corona envelopes have capacitances of about (13.2 MV per coulomb) or 75 nF in the area where [Mazur 2001] carried out these studies

4-8