Transmission Line Model

Timothy Vismor February 23, 2012

Abstract is document describes an overhead transmission line model that is useful for the analysis of large scale electric power systems. It establishes practical techniques for computing the series impedance and shunt admittance of arbitrary conductor con gurations. Consideration is given to computational aspects of computing transmission line impedance parameters.

Copyright 1990-2012 Timothy Vismor

CONTENTS

LIST OF TABLES

Contents
1 Introduction 2 Series Impedance 2.1 Carsons Equations . . . . . . . . . . . . . . . . . . . . . . . . 2.1.1 Approximation of P and Q in Carsons Equations . . . 2.1.2 Accuracy of Approximations to P and Q . . . . . . . . 2.1.3 Use of First Order Approximations to P and Q . . . . 2.2 Impedance of an N Conductor Transmission Line . . . . . . . 2.3 Series Impedance Computations . . . . . . . . . . . . . . . . . 2.3.1 Computation of k in the Series Approximation to P and Q . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.2 Constants in the P and Q Terms of Carsons Equations 2.3.3 Unit Conversions Associated With GMR Terms . . . . 3 Shunt Admittance 3.1 Linear Charge Density Along a Single Conductor 3.2 Capacitance of N Conductors . . . . . . . . . . . 3.3 Shunt Admittance and Reactance Matrices . . . . 3.4 Shunt Admittance Computations . . . . . . . . . 3.4.1 Potential Coecient Unit Conversions . . 3.4.2 Self Potential Unit Conversions . . . . . . 4 Units of Measure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 3 4 5 7 7 8 9 9 10 10 11 11 12 13 13 14 14 15

List of Figures
1 Transmission Line Geometry . . . . . . . . . . . . . . . . . . 6

List of Tables
1 2 Range of the Constant k . . . . . . . . . . . . . . . . . . . . . Impedance Unit Conversions . . . . . . . . . . . . . . . . . . 7 16

2 SERIES IMPEDANCE

1 Introduction
Transmission and distribution lines consist of an arbitrary spatial arrangement of one or more conductors. Information about these conductors is transformed into parameters required for power system analysis as follows: e fundamental data consists of a description of each conductor and how the conductors are arranged on their support structures. Conductor and spacing information is converted into an impedance matrix representing the self and mutual impedances of the complete con guration. e impedance matrix is reduced to eliminate elements that are not required by the analysis. e reduced impedance matrix is converted to symmetrical components when sequence impedances are required. If sequence impedances are the only available information, they can be transformed into a reduced impedance matrix. e remainder of this document examines the rst two stages of this modeling process in detail. at is, we examine techniques for transforming conductor parameters and con guration data into impedance and capacitance matrices. e analysis is limited to overhead transmission lines.

2 Series Impedance
e series impedance of an overhead transmission line is primarily a function of frequency, conductor resistance, conductor geometry, line geometry, and earth conductivity. In the fundamental work on the subject, Carson(1926) [1] developed equations for the self impedance of a conductor with earth return and the mutual impedance of two conductors with common earth return. ese equations have been discussed and elaborated upon many times over the years. Wagner and Evans(1933) [2], Clarke(1943) [3], and Anderson(1987) [4] provide excellent complementary discussions of the topic. e current formulation of the problem draws from each of these sources but follows the exposition of Clarke(1943) [3] most closely.

2.1 Carsons Equations

2 SERIES IMPEDANCE

2.1 Carsons Equations

Carsons formulas are
= + 2 = 2 2 + 4( + ) + 4( + )

(1) (2)

where
is the self-impedance of conductor with ground return. is the mutual impedance between conductors and with common

ground return. is the eective radius (or geometric mean radius) of conductor in centimeters. is the height of conductor in centimeters. is the internal resistance of conductor . the distance between conductors and in centimeters. the distance between conductor and the image of conductor in centimeters. is 2 , where is the frequency in cycles per second. Obviously, the self-impedance and mutual impedance can be decomposed into their real and imaginary components
= + = +

(3) (4)

Collecting terms in Equation 1 and Equation 2 and comparing to Equation 3 and Equation 4, it is apparent that
= + 4 = 4 = 2 2 + 4 + 4

(5) (6) (7) (8)

= 2

2.1 Carsons Equations

2 SERIES IMPEDANCE

2.1.1 Approximation of P and Q in Carsons Equations e and terms in the preceding equations are de ned by Carson as an in nite series expressed in terms of two parameters, call them and . e form of and are the same for Equation 1 and Equation 2. However, the value of and dier. For self impedances
= 4 2 = 0

(9) (10)

For mutual impedances

= 2 2 1 ( + )

(11) (12)

where
is the earth conductivity in ab/cm3 . is the angle de ned in Figure 1.

Figure 1 de nes the line geometry associated with Equation 9 through Equation 12. e rst few terms of the expansion of and follow:
2 (2) (0.6728 + ) (2) 2 = + + 2 8 16 16 32 (3) (4) + 3 4 1536 452

(13)

(2) 1 2 = 0.0386 + + 2 2 64 32 (4) (1.0895 + ) (4) + 4 4 384 384 452

3 (3) 2

(14)

2.1 Carsons Equations

2 SERIES IMPEDANCE

Figure 1: Transmission Line Geometry

2.1 Carsons Equations 2.1.2 Accuracy of Approximations to P and Q

2 SERIES IMPEDANCE

Clarke(1943) [3] states that Equation 13 and Equation 14 exhibit less than one percent error for values of up to one. Table 1 shows the wide applicability of these expressions for fundamental and harmonic analysis of power systems by examining values of for a range of geometries, frequencies, and resistivities. Table 1: Range of the Constant k
Distance 100 ft Frequency 60 Hz 660 Hz 1020 Hz 60 Hz 660 Hz 1020 Hz 60 Hz 660 Hz 1020 Hz Earth Resistivity 10 /m3 k 0.4196 1.3916 1.7300 0.1327 0.4401 0.5471 0.0419 0.1391 0.1730

100 /m3

1000 /m3

100 ft - Large double circuit transmission tower 10 /m3 - Resistivity of swampy ground 100 /m3 - Resistivity of average damp earth 1000 /m3 - Resistivity of dry earth

2.1.3 Use of First Order Approximations to P and Q At 60 Hz, it is common practice to ignore the higher order terms of the expansion of and , i.e. let
= 8

(15) (16)

1 2 = 0.0386 + 2

is practice eectively decouples the series impedance from the conductors height above ground. According to Wagner and Evans(1933) [2], this omission

2.2 Impedance of an N Conductor Transmission2 SERIES IMPEDANCE Line tends to overstate the computed resistance and understate the computed reactance. At commercial frequencies and low earth resistivities ( = 10), the rst order approximations may introduce resistance errors in the neighborhood of 10 per cent. Under similar circumstances, self reactance errors rarely exceed one per cent. However, mutual reactance errors are more volatile. For = 10, = 60, and = 200 feet, the low order approximation of understates the mutual reactance by much as 4 per cent. At higher harmonics, these tendencies are magni ed.

2.2 Impedance of an N Conductor Transmission Line

e two conductor problem of Section 2.1 can be generalized to a group of conductors with a common ground return. If currents 1 , 2 , , are owing through the conductors, the voltage drop along conductor is
= 1 1 + + + +

(17)

Similar equations can be constructed for all conductors in the group. Expressing the complete set of voltage drop equations in matrix notation yields
=

(18)

where
is the voltage vector. is the current vector. is the series impedance matrix.

e elements of the impedance matrix are computed using Carsons equations:

= + + if = if

(19)

where , , , and are de ned by Equation 5 through Equation 8. e series admittance of the conductor con guration can be determined by inverting its impedance matrix, i.e.
=

(20)

2.3 Series Impedance Computations

2 SERIES IMPEDANCE

2.3 Series Impedance Computations

is discussion of overhead transmission line series impedance concludes with a brief dicussion of computing contant factors associated with the impedance matrix and reconciling units of measure while evaluating these constants. 2.3.1 Computation of k in the Series Approximation to P and Q e parameter appears in the series expansion which approximates the and terms of Carsons equations (see Equation 9 and Equation 11 of Section 2.1.1 for details). It is of the form
= 42

(21)

where
is the earth conductivity in ab/cm3 . is a distance in centimeters.

is can be rewritten in terms of readily available quantities (i.e. commonly published units) by substituting earth resistivity (/m3 ) for conductivity and distance in conductor separation units for distance in centimeters as follows
= 4( ) 2( )

(22)

where
is conductor separation unit. In the US, conductor separation is usually measured in feet. is the number of centimeters in one conductor separation unit. is a constant converting ab/cm3 to /m3 .

Assuming that the frequency and resitivity are constant for any set of impedance computations the bulk of the expression
4( ) 2( )

(23)

is a constant which is computed once then stored for reuse.

2.3 Series Impedance Computations

2 SERIES IMPEDANCE

2.3.2 Constants in the P and Q Terms of Carsons Equations After and are computed, the terms 4 and 4 in Equation 5 through Equation 8 of Section 2.1.1 produce impedances in units of ab/cm. If impedances are expressed in / , these terms expand to
4( )( )

(24)

and
4( )( )

(25)

where
is line length unit. In the US, line length is usually measured in miles. is the number of centimeters in one line length unit. is a constant converting ab to . i.e. 1 109 .

Assuming that the frequency is constant, both and are multiplied by the same factor
4 2( )( )

(26)

e rst terms of of the inductive reactance equations (Equation 7 and Equation 8 of Section 2.1) are also multiplied by half of this value, i.e.
2 2( )( )

(27)

Once again, both of these constants are calculated once then stored. 2.3.3 Unit Conversions Associated With GMR Terms When the logarithmic term in Equation 7 of Section 2.1 is computed, the conductors GMR must be converted to conductor separation units, i.e.
2

(28)

is actually evaluated as
2 ( )

(29)

where converts conductor radius units to conductor separation units. Factoring out a constant in this expression yields 10

3 SHUNT ADMITTANCE

(30)

where
= 2

(31)

e factor is also calculated once and stored.

3 Shunt Admittance
e capacitance of an overhead transmission line is primarily a function of conductor geometry and line geometry. All of the references cited in Section 2 with regard to series impedance also touch upon the subject of self and mutual capacitance. e current discussion most closely follows the work of Anderson(1987) [4].

3.1 Linear Charge Density Along a Single Conductor

Assuming that a group of conductors carrying linear charge densities 1 , 2 , , are located above the ground plane, the voltage of conductor to ground is
= 1 1 + + + +
1

(32)

where
is the charge of conductor in coulombs/meter. is the radius of conductor . is the distance between conductor and its image (i.e. 2 in Figure 1). is the distance between conductor and conductor . is the distance between conductor and the image of conductor as

illustrated in Figure 1. is the permittivity of the medium. Note: e distances associated with each logarithmic ratio (e.g. and or and ) of Equation 32 must be expressed in the same units.

11

3.2 Capacitance of N Conductors

3 SHUNT ADMITTANCE

3.2 Capacitance of N Conductors

Given a group of conductors carrying linear charge densities 1 , 2 , , that are located above the ground plane, equations of the same form as Equation 32 (Section 3.1) can be constructed for all conductors in the group. Expressing the complete set of potential equations in matrix notation yields
=

(33)

where
is the voltage vector. is the charge vector. is the potential coecient matrix.

e elements of the potential matrix (with units of F1 m) are de ned as follows:

= 2 2

if = if

(34)

Recall that the permittivity of a medium is often expressed as

= 0

(35)

where
0 is the permittivity of free space (i.e. 8.8541853 1012 F/m). is the relative permittivity of the medium (e.g. 1 for air).

In matrix notation, the capacitance of the con guration is

=

(36)

solving Equation 33 for the charge vector yields

=

(37)

By inspection it is apparent that

=

(38)

e matrix is sometimes known as the capacitance coecients (or Maxwells coecients) of the line. 12

3.3 Shunt Admittance and Reactance Matrices SHUNT ADMITTANCE 3

3.3 Shunt Admittance and Reactance Matrices

If the charge density along the transmission line is sinusoidal rather than linear, Equation 36 is a phasor equation. Multiplying Equation 36 by yields
=

(39)

Recalling that the current phasor associated with a sinusoidal variation in charge is expressed as
=

(40)

It is apparent that
=

(41)

An alternate expression for the charging current is

=

(42)

erefore, the charging admittance (which is pure susceptance) must be

=

(43)

e preceding discussion suggests a computational procedure for determining the capacitive parameters of a conductor con guration: 1. Compute the con gurations potential matrix using Equation 34. 2. Compute its capacitance matrix by inverting . 3. Multiply the capacitance matrix by the scalar to obtain the shunt admittance matrix . 4. Invert the the shunt admittance matrix to determine the capacitive reactance .

3.4 Shunt Admittance Computations

is discussion of overhead transmission line shunt admittance concludes with a brief dicussion of computing contant factors associated with the potential matrix and reconciling units of measure while evaluating these constants.

13

3.4 Shunt Admittance Computations 3.4.1 Potential Coecient Unit Conversions

3 SHUNT ADMITTANCE

e constant associated with the computation of potential coecients in Equation 34 depends only upon the medium in which the conductors reside. Assuming that the conductors are suspended in air ( = 1), the potential constant (in F1 m) is
1 1 = = 1.79751087 1010 12 20 2 8.8541853 10 1

(44)

To compute potential coecients in line length units rather than meters, an additional conversion factor is required, i.e. the multiplier in Equation 34 is actually
2

or
( )1.79751087 1010

(45)

which produces potential coecients with units F1 . is product is computed once and stored. 3.4.2 Self Potential Unit Conversions e self potential in Equation 34 is

When this term is computed, the numerator and denominator of the logarithmic factor must be in the same units. Assuming that the conductors diameter (in ) is readily available, the distance must be converted to conductor separation units and the diameter must be converted to a radius. erefore, the computed logarithmic factor is
( ) 2

where converts conductor diameter to conductor separation units. Factoring out a constant in this expression

14

UNITS OF MEASURE

(46)

where
= 2

(47)

e factor is computed once and saved. Note: In the context of the current discussion, the clear choice of unit for capacitive reactance is . However, the capacitive reactance found in American reference materials is often M or more speci cally Mmile. Hence, an additional factor may be required when converting capacitive reactance from computational units to commonly published units (ie. 106 for converting M to ).

4 Units of Measure
A number of unit systems are involved in overhead transmission line impedance calculations. e current section is intended to make their distinctions clear. An engineer provides data for the calculations in what we will refer to as the user unit system (also referred to as generally available units or commonly published units in other sections of this document). User units may vary along the following lines: Conductor separation units, , are associated with conductor-to-conductor and conductor-to-image distances. In US applications, conductor separation is usually measured in feet. Conductor radius units, , are associated with eective radius (GMR) measurements. In American reference materials, conductor GMR is usually reported in feet. Conductor diameter units, , are associated with outside diameter measurements. In American reference materials, conductor diameter is usually reported in inches. Line length units, , are associated with the length of a line section. In US utility applications, line length is usually measured in miles.

15

REFERENCES

REFERENCES

e units in which equations are expressed in this document are called problem formulation units. e units in which impedance calculations are actually implemented are referred to as computation units. Table 2 describes these unit systems in detail. Table 2: Impedance Unit Conversions
Units Quantity Frequency Earth Resitivity k Resistance Inductive Reactance Potential Coecients Maxwells Coecients Capacitive Susceptance Capacitive Reactance Conductor Diameter Conductor GMR Conductor Separation Formulation Hz ab/cm3 ab/cm ab/cm ab/cm F1 m F/m /m m m cm cm Computational Hz /m3 / / / F1 F/ /

User Hz /m3 n/a / / n/a n/a n/a (M)

- conductor diameter unit - conductor radius unit - conductor separation unit - line length unit

References
[1] J. Carson, Electric Circuit eory and Operational Calculus, McGraw-Hill, New York, 1923. 3 [2] C. Wagner and R. Evans, Symmetrical Components, McGraw-Hill, New York, 1933. 3, 7 [3] E. Clarke, Circuit Analysis of AC Power Systems, John Wiley & Sons, New York, 1943. 3, 7

16

REFERENCES

REFERENCES

[4] J. Anderson, Transmission Line Reference Book, 2nd ed. Rev., Electric Power Research Institute, Pal Alto, CA, 1987. 3, 11

17