ECE 43200/53201 Power Systems Fall 2020

AC Transmission Lines
Components and Electrical Parameters

2.4 Line Capacitance, C

Electrical Field Around a Conductor, [1]

The conductors of an energized transmission line carry charges. These charges produce an electric field
around the conductors. The line capacitance of a long straight, solid round conductor is calculated in this
section. Fig.18a shows an energized conductor and the radial electric field lines emanating from the
conductor. The field inside the conductor is zero. The constant electric potential lines are concentric
circles.

(a) (b)

Fig.18 (a) Electric field and (b) constant potential lines around a charged conductor

For the calculation of the electric flux density, D, we select a constant potential line at a radius of x, as
shown in Fig.18b.
According to Gauss' Law the electric flux density per unit length is:

⃗⃗ | = 𝑄
|𝑫 (23)
2𝜋𝑥

where Q is the electric charge per unit length.

The electric field intensity is the ratio of the flux density and the permittivity:

⃗⃗
⃗⃗ | = |𝑫| = 𝑄
|𝑬
1
(24)
0ε 2𝜋ε 0 x

𝐹
The permittivity of air can be taken as that universal constant for free space, 𝜀0 = 8.85 × 10−12 𝑚. The
potential difference between points P1 and P2 is

1
ECE 43200/53201 Power Systems Fall 2020

𝐷2
𝑃2 𝑄 1 𝑄 𝐷
𝑉12 = ⃗⃗ | 𝑑𝑥
∫𝑃1 |𝑬 = 2𝜋ε ∫ 𝑑𝑥 = 2𝜋ε ln (𝐷2) (25)
0 x 0 1
𝐷1

This equation is suitable to calculate the potential difference in the vicinity of a transmission line
conductor.

Three-Phase Transmission Line Electric Field

The energized conductors of a three-phase transmission line generate an electric field. The electric
field has undesirable environmental effects: (a) the high electric field at the conductor surface produces
corona discharges that generate radio and television disturbances; (b) the electric field at the ground
level may produce small shocks and disturb people walking under the line.

Corona Discharge
When the electric field intensity is above 20-25 kV/cm at the surface of a conductor, corona discharge
is generated. The corona discharge is the electric breakdown of the air near the conductor.

Fig.19Corona discharge on a wet conductor, [1]

Fig.19 shows a typical corona discharge originated by a water droplet on the conductor. Generally,
water droplets, dirt and metal protrusions on the conductor surface increase the local electric field and
initiate corona. The high humidity decreases the breakdown strength of the air surrounding the
conductors, which further increases the intensity of the discharge. The strong corona discharge on a
conductor is visible at night. The corona discharge intensity increases with rain and wet weather.
Other sources of discharge include polluted insulators in foggy weather, during which dry-band arcing
may occur. Corona discharge also occurs during fair weather conditions. This occurs mostly where the
hardware holding the conductors has sharp edges and surface protrusions which increase the electric
field and initiate discharge.
The electric effect of the corona is high-frequency current pulses, which generate radio and TV
disturbances. These effects cause public protest and complaints. The power companies eliminate or
reduce the TV and radio disturbances by using bundled conductors, as shown in Fig.6.

Three-phase Line Capacitance

The three-phase line capacitance for a balanced system is usually referred to positive sequence
capacitance, and assumes that the line is transposed. Fig.20 shows a simplified three-phase
transmission line without ground.

Fig.20 A simplified three-phase line

2
ECE 43200/53201 Power Systems Fall 2020

The charges (Q) on the conductor surface produce voltage differences between the conductors. These
voltage differences are equal to the line-to-line voltages. The charge on conductor A produces a
voltage difference between conductor A and B. This voltage difference is calculated by integrating the
QA generated electric field between A and B, which is achieved by the substitution of D 1 = rc and D2 =
DAB into Equation (25).
The voltage difference produced by QA between A and B is:

𝑄 𝐷𝐴𝐵
𝑉𝐴𝐵,𝑄𝐴 = 2𝜋ε𝐴 ln ( ) (26)
0 𝑟𝑐

The charge on conductor B also creates a voltage potential between A and B. This voltage is computed
by integrating the QB generated electric field between B and A, which is achieved by the substitution
of D1 = DAB and D2 = rc in Equation (25) to yield

𝑄 𝑟
𝑉𝐴𝐵,𝑄𝐵 = 2𝜋ε𝐵 ln (𝐷 𝑐 ) (27)
0 𝐴𝐵

Similarly, the charge on conductor C generates a voltage between A and B. This voltage difference is
determined by integrating the Qc produced electric field between A and B, which is achieved by the
substitution of D1 = DAC and D2 = DBC in Equation (25) to give

𝑄 𝐷
𝑉𝐴𝐵,𝑄𝐶 = 2𝜋ε𝐶 ln (𝐷𝐵𝐶 ) (28)
0 𝐴𝐶

The total voltage between A and B is the sum of the voltages generated by the three phase conductors.
The result of the summation is:

𝑄 𝐷𝐴𝐵 𝑄 𝑟 𝑄 𝐷
𝑉𝐴𝐵 = 𝑉𝐴𝐵,𝑄𝐴 + 𝑉𝐴𝐵,𝑄𝐵 + 𝑉𝐴𝐵,𝑄𝐶 = 2𝜋ε𝐴 ln ( ) + 2𝜋ε𝐵 ln (𝐷 𝑐 ) + 2𝜋ε𝐶 ln (𝐷𝐵𝐶 ) (29)
0 𝑟𝑐 0 𝐴𝐵 0 𝐴𝐶

The voltage difference between A and C can be derived using similar calculations to the above
derivation. The voltage difference between A and C is:

𝑄 𝐷𝐴𝐶 𝑄 𝐷 𝑄 𝑟
𝑉𝐴𝐶 = 2𝜋ε𝐴 ln ( ) + 2𝜋ε𝐵 ln (𝐷 𝐵𝐶 ) + 2𝜋ε𝐶 ln (𝐷 𝑐 ) (30)
0 𝑟𝑐 0 𝐴𝐵 0 𝐴𝐶

If the transmission line is transposed, the distances between the conductors are replaced by an average
distance called the GMD (geometric mean distance) that is calculated by Equation (11), which is
repeated here:

𝐺𝑀𝐷 = 3√𝐷𝐴𝐵 𝐷𝐴𝐶 𝐷𝐵𝐶 (31)

The substitution of the GMD yields:

𝑄 GMD 𝑄 𝑟 𝑄𝐴 −𝑄𝐵 GMD

𝑉𝐴𝐵 = 2𝜋ε𝐴 ln ( 𝑟𝑐
) + 2𝜋ε𝐵 ln (GMD
𝑐
)= 2𝜋ε0
ln ( 𝑟𝑐
) (32)
0 0

3
ECE 43200/53201 Power Systems Fall 2020

𝑄 GMD 𝑄 𝑟 𝑄𝐴 −𝑄𝐶 GMD

𝑉𝐴𝐶 = 2𝜋ε𝐴 ln ( ) + 2𝜋ε𝐶 ln (GMD
𝑐
)= ln ( ) (33)
0 𝑟𝑐 0 2𝜋ε0 𝑟𝑐

The sum of the two voltages is:

2𝑄𝐴 −𝑄𝐵 −𝑄𝐶 GMD 3𝑄 GMD

𝑉𝐴𝐵 + 𝑉𝐴𝐶 = ln ( ) = 2𝜋ε𝐴 ln ( ) (34)
2𝜋ε0 𝑟𝑐 0 𝑟𝑐

where QA = -QB - Qc for balanced sources.

In a balanced three-phase system the relation between the line-to-line and line-to-neutral voltages is:
° °
𝑉𝐴𝐵 = √3𝑉𝐴𝑁 ej30 , and 𝑉𝐴𝐶 = −𝑉𝐶𝐴 = √3𝑉𝐴𝑁 e−j30 , (35)

The sum of these two voltages is: 𝑉𝐴𝐵 + 𝑉𝐴𝐶 = 2√3𝑉𝐴𝑁 cos(30° ) = 3𝑉𝐴𝑁 (36)

Combining Equations (34) and (36) results in the line-to-neutral voltage:

𝑉𝐴𝐵 +𝑉𝐴𝐶 𝑄 GMD

𝑉𝐴𝑁 = = 2𝜋ε𝐴 ln ( ) (37)
3 0 𝑟𝑐

❖ Finally, the capacitance-to-neutral per unit length is calculated from (37) as:
𝑄 𝟐𝝅𝜺𝟎
𝐂=𝑉𝐴 = 𝑮𝑴𝑫 , [F/m] (38)
𝐴𝑁 𝒍𝒏( )
𝒓𝒄

Many high-voltage transmission lines are built with bundled conductors. The application of bundled
conductors increases the line capacitance. The bundled conductors are replaced by an equivalent
conductor. The radius of this equivalent conductor depends on the number of conductors in the bundle
and the distance between the conductors within the bundle. The equivalent radii of the bundles are:

Two-conductor bundle: 𝑟𝑒𝑞 = √𝑑 𝑟𝑐 (39)

3
Three-conductor bundle: 𝑟𝑒𝑞 = √𝑑 2 𝑟𝑐 (40)

4
Four-conductor bundle: 𝑟𝑒𝑞 = 1.09√𝑑 3 𝑟𝑐 (41)

where rc is the conductor radius, and d is the distance between the conductors in the bundle.

➢ In the case of bundled conductors, the conductor radius rc in Equation (38) is replaced by the
appropriate req.

2.5 Equivalent Medium and Short Line Circuits

The equivalent transmission line circuits for a balanced system are usually referred to as the positive
4
ECE 43200/53201 Power Systems Fall 2020

sequence transmission line equivalent circuits. The transmission line parameters are the resistance,
inductance (or reactance) of the conductors, and the capacitance to neutral. These parameters are
calculated per unit line length and per phase.
The parameters are distributed evenly along the line. The line can be divided into short segments and
each segment has a inductance and resistance connected in series and a capacitance connected in
parallel to the ground, as shown in Fig.21.

Fig.21 Equivalent circuit for a long transmission line

The voltages and currents in a distributed system are calculated by solving the system differential
equations. This is necessary for lines longer than 150-200 miles. The analysis of a long transmission
line is presented later in this section.
A medium-length line can be represented by a Π (pi) circuit. In this circuit, the line resistance and
inductance are concentrated in the middle section, and half of the line capacitance is placed at the
beginning and the other half at the end of the line.

(a) (b)

Fig. 22 Equivalent circuits of a medium length line (a), and of a short line (b)

Fig.22a shows the single-phase equivalent circuit of a medium length line. This circuit is used when
the line length is 50-150 miles (see also Fig.28). The capacitance of a short line is negligible, and this
simplifies the equivalent circuit, as shown in Fig.22b. This circuit is used when the line length is less
than 50 miles.

Example 2-3

Calculation of Transmission Line Parameters[1]

A 500 kV transmission line in Arizona is built with lattice towers supporting a two-conductor Bluebird bundle in each
phase. Fig.23 shows the typical tower arrangement.
The line data are:
Rated line voltage Vline=500kV
Line length is lLine=206mi
Operating frequency is f = 60 Hz
The distance between conductors in the bundle is d=18in
The number of conductors in each bundle is n =2
The distance between the phases is D=32ft
The Bluebird conductor data are obtained from Table 1

5
ECE 43200/53201 Power Systems Fall 2020

Fig.23 A transmission line tower, 500 kV [1]

1.762 𝑖𝑛
The conductor radius is 𝑟𝑐 = 2
The GMR is GMRC = 0.0588 ft
Ω
The ac resistance per mile at 75°C is R75 = 0.0544 𝑚𝑖
𝐹 𝐻
The universal constants are 𝜀0 = 8.85 × 10−12 𝑚 and 𝜇0 = 4𝜋 × 10−7 𝑚
The GMR of the two-conductor bundle is found using Equation (20):

𝐺𝑀𝑅 = √𝑑 𝐺𝑀𝑅𝑐 = √18 × 2.54 × 10−2 × 0.0588 × 0.3048 = 0.091m

The equivalent radius of the two-conductor bundle is determined from Equation (39):

1.762
𝑟𝑒𝑞 = √𝑑 𝑟𝑐 = √18 × 2.54 × 10−2 × 2
× 2.54 × 10−2 = 0.101m

Using Equation (11), the geometric mean distance (GMD) between the phase conductors is:
3 3 3
𝐺𝑀𝐷 = 3√𝐷𝐴𝐵 𝐷𝐴𝐶 𝐷𝐵𝐶 = √D ∙ D ∙ 2𝐷 = D√2 = 32 × 0.3048√2 = 12.289𝑚

The line reactance per unit length is calculated by using Equation (19) and is

𝜇0 GMD 12.289
𝑋𝐿 = 𝜔𝐿 = 2𝜋𝑓 × 𝑙𝑛 ( ) = 60 × 4π × 10−7 × ln ( )=
2𝜋 𝐺𝑀𝑅𝑐 0.091
Ω Ω
= 3.698 × 10−4 = 0.596
m mi

The total reactance of the line is:

𝑋𝐿𝑖𝑛𝑒 = 𝑋𝐿 𝑙𝐿𝑖𝑛𝑒 = 0.596 × 206 = 122.75Ω

The line operating temperature is determined by the ambient temperature, sun radiation and the load current produced
heating. In Arizona, it is prudent to use the 75°C (167°F) temperature for the calculation of the conductor ac
resistance.
6
ECE 43200/53201 Power Systems Fall 2020

The two conductors in the bundle are connected in parallel. The line resistance of this two-conductor bundle is:
𝑅75 0.0544
𝑅𝐿𝑖𝑛𝑒 = 𝑙𝐿𝑖𝑛𝑒 = (206) = 5.6 Ω
2 2

The line resistance and reactance are connected in series as shown in Fig.24. The overall impedance of this line can
be written as

𝑍𝐿𝑖𝑛𝑒 = 𝑅𝐿𝑖𝑛𝑒 +𝑗𝑋𝐿𝑖𝑛𝑒 = 5.6 + j122.75 Ω.

Fig.24 Single phase equivalent circuit of the transmission line

The capacitance-to-ground per unit length is calculated by Equation (38) and (39). The capacitance to ground is:
2𝝅𝜺0 2𝝅𝜺0 nF
𝐶𝑙𝑖𝑛𝑒 = 𝑮𝑴𝑫 = 𝑮𝑴𝑫
= 18.62
𝒍𝒏( ) 𝒍𝒏( ) mi
𝒓𝒄 √𝑑 𝑟𝑐

The line performance is studied using a single-phase equivalent circuit. In this circuit, the line capacitance is divided
into two parts. Half of the total line capacitance is placed on the beginning and the other half at the end of the line, as
shown in Fig.24.

The capacitive reactance at the supply side is:

1
𝑋𝐶𝑆 = − 𝐶𝑙𝑖𝑛𝑒 = −1.383 𝑘𝛺
𝜔 𝑙𝐿𝑖𝑛𝑒
2
The capacitive reactance at the load side is the same: 𝑋𝐶𝐿 = 𝑋𝐶𝑆 = −1.383 𝑘𝛺.

2.6. Long Transmission Lines Differential Equations

In linear circuit analysis the R, L, C parameters were considered lumped parameters, the displacement
current was neglected as well as the conduction currents through the dielectric between conductors.

These assumptions cannot be taken anymore in the case of long transmission lines, when the displacement
and conduction currents through dielectric are directly proportional with the conductor surfaces (length,
respectively). Because of these, the conductor currents vary with the length x of the line: 𝑖 = 𝑖(𝑥, 𝑡).

There is a direct analogy between the field distribution of a uniform transmission line and the uniform plane
wave, [4]. The field distributions for the uniform plane wave and for the uniform transmission line are
both known as transverse electromagnetic (TEM) waves because E and H are both perpendicular to the
direction of propagation, or both lie in the transverse plane. In the transmission line, however, it is possible
and customary to define a voltage and a current. These quantities are the ones for which we shall write
equations, obtain solutions, and find propagation constants, reflection coefficients, and input impedances.

Maxwell's equations can be solved to obtain the differential equations which the voltage or current must
7
ECE 43200/53201 Power Systems Fall 2020

satisfy on a transmission line. Instead we will consider here a circuit model for an incremental length of line
and write two circuit equations (this is a way to begin to tie field theory and circuit theory together).
Our circuit model will contain the line parameters in per-length units of Q/m, , and.: inductance L [H/m],
capacitance C [F/m], shunt conductance G [S/m], and series resistance R [Ω/m]. Knowing the operating
frequency and the dimensions, we can then determine the values of R, G, L, and C on a per-unit-length
basis by using formulas developed in earlier sections.

In order to account for the distributed nature of transmission-line constants, consider the circuit
shown in Fig.25, which represents a line section of incremental length Δx. If at the sending end of the
line a sinusoidal voltage is applied then 𝑣(𝑥, 𝑡) and 𝑖(𝑥, 𝑡)have a sinusoidal variation in time.
V(x) and I(x) denote the voltage and current phasors at position x, which is measured in meters from
the right, or receiving end of the line. Similarly, V(x + Δx) and I(x + Δx) denote the voltage and
current phasors at position (x + Δx).

Fig.25 Incremental length section of a transmission line [3]

The circuit in Fig.25 is analyzed considering its impedance and admittance constants, in sinusoidal
steady state:

𝑧 = 𝑅 + 𝑗𝜔𝐿, [Ω/m]
𝑦 = 𝐺 + 𝑗𝜔𝐶, [S/m] (42)

where the isolation conductance G is usually neglected for overhead 60-Hz lines.

Writing a KVL equation for this circuit

𝑽(𝑥 + Δ𝑥) = 𝑽(𝑥) + (𝑧Δ𝑥)𝑰(𝑥) (43)

𝑽(𝑥+Δ𝑥)−𝑽(𝑥)
Rearranging (43), and taking the limit lim
Δ𝑥→0 Δ𝑥

𝑑𝑽
= 𝑧𝑰(𝑥), or
𝑑𝑥

𝑑𝑽
= (𝑅 + 𝑗𝜔𝐿)𝑰(𝑥) (44a,b)
𝑑𝑥

Similarly, writing a KCL equation for the circuit

𝑰(𝑥 + Δ𝑥) = 𝑰(𝑥) + (𝑦Δ𝑥)𝑽(𝑥 + Δ𝑥) (45)

𝑰(𝑥+Δ𝑥)−𝑰(𝑥)
Rearranging (45) and taking the limit lim
Δ𝑥→0 Δ𝑥

8
ECE 43200/53201 Power Systems Fall 2020

𝑑𝑰
= 𝑦𝑽(𝑥) or
𝑑𝑥

𝑑𝑰
= (𝐺 + 𝑗𝜔𝐶)𝑽(𝑥) (46a,b)
𝑑𝑥

The significance of Equations (44) and (46) can be interpreted respectively as follows: the line voltage
drop (from sending- to receiving-end) per length unit equals the voltages drop on the line's conductors
resistance and inductance. The current decrease per length unit is due to both the conduction current
through dielectric and the charging current with electric charge of the conductor's surfaces.

The differential Equations (44) and (46) have two unknowns, V(x) and I(x). We can eliminate I(x) by
differentiating (44a) and using (46a):

𝑑2 𝑽(𝑥) 𝑑𝑰
= 𝑧 𝑑𝑥 = z𝑦𝑽(𝑥), or
𝑑𝑥 2

𝑑2 𝑽(𝑥)
= (𝑅 + 𝑗𝜔𝐿)(𝐺 + 𝑗𝜔𝐶)𝑽(𝑥) = γ2 𝑉(𝑥), (47a,b)
𝑑𝑥 2

where the so called propagation constant 𝜸 whose units are [𝑚−1 ] is

❖ 𝛾 = √𝑧𝑦 = √(𝑅 + 𝑗𝜔𝐿)(𝐺 + 𝑗𝜔𝐶) (48)

The propagation constant γ is a complex quantity with positive real and imaginary parts, that is 𝛾 = α
+ jβ.

Equation (47) is a linear, second-order, homogeneous differential equation with one unknown, V(x).
By inspection, its solution is of the form

𝑽(𝑥) = 𝑉1 𝑒 𝛾𝑥 + 𝑉2 𝑒 −𝛾𝑥 , (49)

where 𝑉1and 𝑉2 are integration constants (that are calculated based on the boundary
conditions).
Next, substituting (49) in (44)
𝑑𝑽
== 𝛾𝑉1 𝑒 𝛾𝑥 − 𝛾𝑉2 𝑒 −𝛾𝑥 = 𝑧𝑰(𝑥), and solving for 𝑰(𝑥):
𝑑𝑥

𝛾𝑉1 𝑒 𝛾𝑥 −𝛾𝑉2 𝑒 −𝛾𝑥 𝑉1 𝑒 𝛾𝑥 −𝑉2 𝑒 −𝛾𝑥

𝑰(𝑥) = = , (50)
𝑧 𝑍𝑐

where the characteristic impedance (also called surge impedance [2]) is

𝑧 𝑧 𝑧 (𝑅+𝑗𝜔𝐿)
❖ 𝑍𝑐 = 𝛾 = = √𝑦 = √(𝐺+𝑗𝜔𝐶) [Ω] (51)
√ 𝑧𝑦

Let's evaluate the integration constants 𝑉1and 𝑉2from the boundary conditions. At x=0, the receiving
9
ECE 43200/53201 Power Systems Fall 2020

end of the line (see Fig.25), the receiving-end voltage and current are:
𝑉1 −𝑉2
𝑽𝑹 = 𝑽|𝑥=0 = 𝑉1 + 𝑉2 , and 𝑰𝑹 = 𝑰|𝑥=0 = (52a,b)
𝑍𝑐

𝑽𝑹 +𝑍𝑐 𝑰𝑹 𝑽𝑹 −𝑍𝑐 𝑰𝑹
Solving for 𝑉1 and 𝑉2: 𝑉1 = , and 𝑉2 = (53a,b)
2 2

❖ Substituting 𝑉1 and 𝑉2 into (49) and (50), and recognizing the hyperbolic functions cosh and
sinh, we finally obtain the voltage and current equations at any point x along the line in terms
of the receiving-end voltage and current as follows:

𝑽(𝑥) = cosh(𝛾𝑥)𝑽𝑹 + 𝑍𝑐 sinh(𝛾𝑥)𝑰𝑹 (54)

1
𝑰(𝑥) = 𝑍 sinh(𝛾𝑥)𝑽𝑹 + cosh(𝛾𝑥) 𝑰𝑹 (55)
𝑐

𝑒 𝛾𝑥 +𝑒 −𝛾𝑥 𝑒 𝛾𝑥 −𝑒 −𝛾𝑥
where cosh(𝛾𝑥) = , and sinh(𝛾𝑥) = .
2 2

❖ At the sending end of the line, where x = l, the sending-end voltage and current are:

𝑽𝑺 = cosh(𝛾𝑙)𝑽𝑹 + 𝑍𝑐 sinh(𝛾𝑙)𝑰𝑹 (56)

1
𝑰𝑺 = 𝑍 sinh(𝛾𝑙)𝑽𝑹 + cosh(𝛾𝑙) 𝑰𝑹 , (57)
𝑐

where the quantity 𝛾𝑙 is dimensionless.

Note: The Equations (56), (57) use exact parameters valid for any line length. For accurate calculations, these equations
must be applied for overhead lines (60 Hz) longer than 250 km (150 miles), [3]. However, the following simplifications
can be made for easier hand calculations:

• For short length lines, |𝛾𝑙| ≪ 1, then: cosh(𝛾𝑙) ≈ 1, and sinh(𝛾𝑙) ≈ 𝛾𝑙.
(𝛾𝑙)2
• For medium length lines: cosh(𝛾𝑙) ≈ 1 + , and sinh(𝛾𝑙) ≈ 𝛾𝑙.
2!
(𝛾𝑙)2 (𝛾𝑙)3
• For long length lines: cosh(𝛾𝑙) ≈ 1 + , and sinh(𝛾𝑙) ≈ 𝛾𝑙 + .
2 3!

The long line equations are given often in matrix form using the so called ABCD parameters. For
example, the Equations (56) and (57) are rewritten as follows:

𝑽𝑺 𝐴 𝐵 𝑽𝑹
[ ]=[ ][ ] (58)
𝑰𝑺 𝐶 𝐷 𝑰𝑹
where
𝐴 = 𝐷 = cosh(𝛾𝑙), per unit
𝐵 = 𝑍𝑐 sinh(𝛾𝑙), [Ω]
1
𝐶 = 𝑍 sinh(𝛾𝑙). [S] (59)
𝑐

10
ECE 43200/53201 Power Systems Fall 2020

Example 2-4

Calculation of Exact Parameters of a Long Transmission Line [3]

A 765 kV, three-phase, 60 Hz, 300 km, completely transposed line has the following positive-sequence impedance
and admittance:
°
𝑧 = 0.0165 + 𝑗0.3306 = 0.331𝑒 𝑗87.14 [Ω/km]
𝑦 = 𝑗4.674 × 10−6 [S/km]

Calculate the exact ABCD parameters of the line assuming positive-sequence of operation.

Solution: From Equations (48) and (51):

°
0.331𝑒 𝑗87.14 ° °
𝑍𝑐 = √ ° = √7.082 × 104 𝑒 −𝑗2.86 = 266.1𝑒 −𝑗1.43 [Ω],
4.674×10−6 𝑒 𝑗90
and
° °
𝛾𝑙 = √0.331𝑒 𝑗87.14 × 4.674 × 10−6 𝑒 𝑗90 × 300 = 0.3731𝑒 𝑗88.57° = 0.00931 + j0.373 per unit
Then
cosh(𝛾𝑙) = 0.9313𝑒 𝑗0.209° , and sinh(𝛾𝑙) = 0.3645𝑒 𝑗88.63°.

Finally, from(59)

𝐴 = 𝐷 = cosh(𝛾𝑙) = 0.9313𝑒 𝑗0.209° per unit

°
𝐵 = 𝑍𝑐 sinh(𝛾𝑙) = 266.1𝑒 −𝑗1.43 × 0.3645𝑒 𝑗88.63° = 97.0𝑒 𝑗87.2° [Ω]

1 0.3645𝑒 𝑗88.63°
𝐶= sinh(𝛾𝑙) = ° = 1.37 × 10−3 𝑒 𝑗90.06° [S]
𝑍𝑐 266.1𝑒−𝑗1.43

2.7 Lossless Lines and Surge Impedance Loading

When line losses can be neglected, simpler expressions for the line parameters are obtained and the
above concepts are more easily understood. Losses due to corona are usually very small, hence the
shunt conductance is reasonable assumed zero, G = 0, [2]. Also, in a lossless line R=0, resulting in
only line distributed inductance and capacitance, without a significant sacrifice of accuracy.
Therefore, for a lossless line, R=G=0, and the line impedance and admittance are expressed only as

𝑧 = 𝑗𝜔𝐿 [Ω/m], and 𝑦 = 𝑗𝜔𝐶 [S/m] (60)

❖ The characteristic impedance from (51) is now called surge impedance (for a lossless line) and is
pure resistive, that is

𝑧 𝑗𝜔𝐿 𝐿
𝑍𝑐 = √𝑦 = √𝑗𝜔𝐶 = √𝐶 , [Ω] (61)

and the propagation constant from (48) is purely imaginary, that is

11
ECE 43200/53201 Power Systems Fall 2020

𝛾 = √𝑧𝑦 = √(𝑗𝜔𝐿)(𝑗𝜔𝐶) = 𝑗𝜔√𝐿𝐶 = j𝛽, where 𝛽 = 𝜔√𝐿𝐶 [m-1] (62)

❖ Surge impedance loading (SIL) is the power delivered by a lossless line to a load resistance equal
𝐿
to the surge impedance 𝑍𝑐 = √𝐶.

Fig.26 shows a lossless line terminated by a resistance equal to its surge impedance. This line
represents either a single-phase line or one phase-to-neutral of a balanced three-phase line.

Fig.26 Lossless transmission line

terminated by its surge impedance [3]

𝑒 𝑗𝛽𝑥 +𝑒 −𝑗𝛽𝑥
From (54) and (55) for a lossless line, recognizing that cosh(𝛾𝑥) = cosh(𝑗𝛽𝑥) = =
2
𝑒 𝑗𝛽𝑥 −𝑒 −𝑗𝛽𝑥
cos (𝛽𝑥) and sinh(𝛾𝑥) = sinh(𝑗𝛽𝑥) = = 𝑗sin(𝛽𝑥), we obtain
2

𝑽(𝑥) = cos (βx) 𝑽𝑹 + 𝑗𝑍𝑐 sin(𝛽𝑥)𝑰𝑹 (63)

1
𝑰(𝑥) = 𝑍 𝑗sin(𝛽𝑥)𝑽𝑹 + cos(𝛽𝑥) 𝑰𝑹 (64)
𝑐

Assuming the receiving end voltage to be the reference phasor: 𝑽𝑹 = 𝑉𝑅 𝑒 𝑗0 , then, at (SIL) surge
𝑉
impedance loading, 𝑰𝑹 = 𝑍𝑅 𝑒 𝑗0 , and the Equations (63) and 64) become
𝑐

𝑽(𝑥) = (cos (𝛽x) + jsin(𝛽𝑥))𝑉𝑅 = 𝑉𝑅 𝑒 𝑗𝛽𝑥 [V] (65)

𝑉 𝑉𝑅
𝑰(𝑥) = (cos (𝛽x) + jsin(𝛽𝑥)) 𝑍𝑅 = 𝑒 𝑗𝛽𝑥 [A] (66)
𝑐 𝑍𝑐

From (65) and (66), V(x) and I(x) change phase by 2𝜋 radians when x = 2 𝜋/𝛽. Denoting wavelength by 𝜆, and using 𝛽
from (62),
2𝜋 2𝜋 1
𝜆=
𝛽
=
𝜔√𝐿𝐶
=
𝑓√𝐿𝐶
(67)
1
or 𝜆𝑓 =
√𝐿𝐶
(68)
1 1
The term ( ) is the velocity of propagation of voltage and current waves along a lossless line. For overhead lines, ( )≈
√𝐿𝐶 √𝐿𝐶
8
3 x 10 m/s, and for f = 60 Hz, (67) gives

1 3 x 108
𝜆= ≈ = 5 x 106 m = 5000km = 3125 mi
𝑓√𝐿𝐶 60

12
ECE 43200/53201 Power Systems Fall 2020

Typical power-line lengths are only a small fraction of the above 60-Hz wavelength.

At SIL, the voltage profile is flat, that is from Eq.(65): |𝑽(𝑥)| = |𝑉𝑅 | volts (69)

That is, the voltage magnitude at any point x along a lossless line at SIL is constant (see Fig.28).
Using (65) and (66), the complex power flowing at any point x along the line is

|𝑉𝑅 |2
𝑆(𝑥) = 𝑃(𝑥) + 𝑗𝑄(𝑥) = 𝑽(𝑥)𝑰∗ (𝑥) = (70)
𝑍𝑐

Thus the real power flow along a lossless line at SIL remains constant from the sending end to the
receiving end. The reactive power flow is zero.

V2rated
❖ At rated line voltage, the real power delivered, or SIL, is: 𝑆𝐼𝐿 = (71)
𝑍𝑐

where rated voltage is used for a single-phase line and rated line-to-line voltage is used for
the total real power delivered by a three-phase line.

Example 2-5
(Ex. 4-2, Ex. 4-3, [2])

A 345 kV, 60 Hz three phase transmission line, built with Bluebird ACSR conductors that are arranged similarly to
Fig.17, has the following data:
Phase spacing between conductors: DAB = 7.5 m; DBC = DAB; DAC = 2DAB.
Conductor diameter: 1.762 inches.
T he length of the line is: l = 100 km
Line resistance: R=0.031 Ω/km
Ignoring the effect of ground and conductor sags, calculate:
(a) the distributed line inductive reactance ωL [Ω/km], and line capacitive admittance [S/km]; (b) the surge impedance Zc
and the Surge Impedance Loading (SIL); (c) the percentage loss if the line is surge-impedance loaded.

Solution
(a) Using Equation (19) the distributed line inductive reactance is:
𝜇 GMD
𝑋𝐿 = ωL = 𝜔 0 𝑙𝑛 ( ),
2𝜋 𝐺𝑀𝑅 𝑐
where the GMD equivalent distance between conductors is calculated from Equation (11): 𝐺𝑀𝐷 = 3√𝐷𝐴𝐵 𝐷𝐴𝐶 𝐷𝐵𝐶 =
3
√7.5 × 15 × 7.5 = 9.45 m,

and from Table 1: 𝐺𝑀𝑅𝑐 = 0.0588 ft = 0.0588 × 0.3048 m = 0.0179 m

The line inductive reactance per unit length is

GMD 9.45 𝑚Ω Ω
𝑋𝐿 = ωL = 𝑓𝜇0 𝑙𝑛 (𝐺𝑀𝑅 ) = (60)(4𝜋 × 10−7 )𝑙𝑛 (0.0179) = 0.473 𝑚 = 0.473 𝑘𝑚
𝑐
The total reactance of the line results as: 𝑋𝐿𝑖𝑛𝑒 = 𝑋𝐿 𝑙𝐿𝑖𝑛𝑒 = 0.473 × 100 = 47.3 Ω

2𝜋𝜀0 (2𝜋8.85×10−12 )
Using Equation (38) the distributed line capacitive admittance is: 𝜔C = (2𝜋𝑓) 𝐺𝑀𝐷 = (2π60) =
𝑙𝑛( ) 𝑙𝑛(
9.45
)
𝑟𝑐 1.762
( 2 ×2.54×10−2 )
S μS
3.467 × 10−9 = 3.467
m km

13
ECE 43200/53201 Power Systems Fall 2020

μS
The total line capacitive reactance is 𝑌𝐿𝑖𝑛𝑒 = 𝜔C𝑙𝐿𝑖𝑛𝑒 = 3.467 km × 100km = 346.7μS = 0.347mS

𝐿 ω𝐿 0.473×10−3
The surge impedance from Equation (61) is 𝑍𝑐 = √𝐶 = √ω𝐶 = √3.467×10−9 ≅ 369Ω.
V2rated 3452 ×106
From Equation (71): 𝑆𝐼𝐿 = 𝑍𝑐
= 369
= 3.225 × 108 W ≅ 323MW.

(c) At surge-impedance loading of this transmission line (see Fig.26), the per-phase current through the line is

𝑉𝑝ℎ𝑎𝑠𝑒 𝑉𝐿𝐿 345×103

𝐼= 𝑍𝑐
= = = 539.8 𝐴,
√3𝑍𝑐 √3×369
and the power loss as a percentage of SIL in this 100km long line, is

𝑃𝑡𝑜𝑡𝑎𝑙.𝑙𝑜𝑠𝑠 3×𝐼 2 𝑅𝐿𝑖𝑛𝑒 3×(539.82 )×(0.031×100)

%𝑃𝑙𝑜𝑠𝑠 = 100 × = 100 × = 100 × ≅ 0.84%
𝑆𝐼𝐿 𝑆𝐼𝐿 3.225×108

Table 3, [6], lists surge impedance and SIL values for typical overhead 60-Hz three-phase lines.

Table 3

Voltage Profiles
In practice, power lines are not terminated by their surge impedance. Instead, loadings can vary from
a small fraction of SIL during light load conditions up to multiples of SIL, depending on line length
and line compensation, during heavy load conditions.
If a line is not terminated by its surge impedance, then the voltage profile is not flat. Fig.27 shows
voltage profiles of lines with a fixed sending-end voltage magnitude V s for line lengths l up to a
quarter wavelength.

Fig.27 Voltage profiles of an uncompensated lossless line [3]

14
ECE 43200/53201 Power Systems Fall 2020

This figure shows four loading conditions: (1) no-load, (2) SIL, (3) short circuit, and (4) full load,
which are described as follows [3]:

1. At no-load, IR = 0 and (63) yields 𝑽(𝑥) = cos (𝛽𝑥) 𝑽𝑹 . The no-load voltage increases from
𝑽𝑺 (𝑥) = cos (𝛽𝑙) 𝑽𝑹 at the sending end to 𝑽𝑹 at the receiving end (where x = 0).

2. From (69), the voltage profile at SIL is flat.

3. For a short circuit at the load, VR= 0 and (63) yields 𝑽(𝑥) = 𝑗𝑍𝑐 sin(𝛽𝑥)𝑰𝑹 . The voltage decreases
from 𝑽𝑺 (𝑥) = 𝑗𝑍𝑐 sin(𝛽𝑙)𝑰𝑹 at the sending end to VR = 0 at the receiving end.

4. The full-load voltage profile, which depends on the specification of full-load current, lies above
the short-circuit voltage profile.

Fig.27 summarizes these results, showing a high receiving-end voltage at no-load and a low
receiving-end voltage at full load. This voltage regulation problem becomes more severe as the line
length increases.

Line Loadability

The Surge Impedance Loading (SIL) provides a benchmark in terms of which the mount of maximum
loading of a transmission line can be expressed. This loading is a function of the length of the
transmission line so that certain constraints are met. This approximate loadability is a function of the
line length:
• Short line less than 50 miles (80 km) in length can be loaded more than 3 times SIL, based on
not exceeding the thermal limits.
• Medium length lines 50-150 miles in length (80-240 km) can be loaded to 1.5 to 3 times the
SIL before the voltage drop across them exceeds more than 5%.
• Long lines in be 150-300 miles (240-480 km) range can be loaded to around the SIL because of
the ability limit (discussed later in this course) so that the phase angle of the voltage between
the two ends does not exceed 40 to 45 degrees.

The loadability of medium and long length lines can be increased by providing series and shunt
compensation (will be discussed later in this course when dealing with stability). These results are
summarized in the table below.

Lumped Transmission Line Models in Steady State [2]

Long-Length Lines

In balanced sinusoidal steady state, the distributed-parameter model of the transmission line based on
Equations (56) and (57) can be represented by a two-port as shown in Fig.28a, where, the symmetry is
necessary from both ports due to the bilateral nature of the transmission line.

15
ECE 43200/53201 Power Systems Fall 2020

Fig.28. Per-phase representation of the transmission lines

(a) long length, medium lines (b) and short lines (c), [2]

The parameters Zseries and Yshunt/2 of a long line, in the two-port representation of Fig.26a, can be
found as follows:
Short-circuiting the receiving end, VR = 0, and from (56)

𝑽𝑺
❖ | = 𝑍𝑠𝑒𝑟𝑖𝑒𝑠 = 𝑍𝑐 sinh(𝛾𝑙) (72)
𝑰𝑹 𝑽 =0
𝑹

Note: Observe that the parameter 𝑍𝑠𝑒𝑟𝑖𝑒𝑠 , in Equation (72), is identical with parameter B from Equation (59) of the matrix
form of the long line equations.

Open-circuiting the receiving end, IR = 0, from (56) and using the voltage division

𝑽𝑺 𝑌𝑠ℎ𝑢𝑛𝑡
| = cosh(𝛾𝑙) = 1 + 𝑍𝑠𝑒𝑟𝑖𝑒𝑠 (73)
𝑽𝑹 𝑰 =0 2
𝑹

cosh(𝛼)−1 𝛼
From (72) and (73), and recognizing that = tanh ( 2 ), the shunt admittance is
sinh(𝛼)

𝛾𝑙
𝑌𝑠ℎ𝑢𝑛𝑡 tanh( )
❖ = 2
(74)
2 𝑍𝑐

Medium-Length Lines

For lines less than 200 miles in length at 60 Hz frequency, it makes almost no difference to assume the
transmission line resistance as a lumped value Rline = Rlline, rather than distributed. In addition, the
shunt conductance can be neglected, with G = 0, so that the rest of the line is lossless. Hence the
𝐿
characteristic impedance is real, 𝑍𝑐 = √𝐶 , as given by Equation (61), and the propagation constant
is purely imaginary, 𝛾 = j𝛽 = 𝑗𝜔√𝐿𝐶, as given by Equation (62) .

In Equation (72), recognizing that sinh(𝑗𝛽𝑙) = 𝑗sin(𝛽𝑙),

𝐿
𝑍𝑠𝑒𝑟𝑖𝑒𝑠 = j√ sin(𝛽𝑙) (75)
𝐶

Recognizing that for small values of 𝛽𝑙, sin(𝛽𝑙) ≈ 𝛽𝑙, and making use of 𝛽 = 𝜔√𝐿𝐶, from (75) we
16
 ECE 43200/53201 Power Systems Fall 2020

obtain

𝑍𝑠𝑒𝑟𝑖𝑒𝑠 = 𝑗𝜔𝐿𝑙 = 𝑗𝜔𝐿𝑙𝑖𝑛𝑒 [Ω], (76)

where 𝐿𝑙 = 𝐿𝑙𝑖𝑛𝑒 .

(𝛾𝑙)2
Similarly from (73), recognizing that for small values of 𝛾𝑙, cosh(𝛾𝑙) ≈ 1 + , we obtain
2!

(𝛾𝑙)2
𝑌𝑠ℎ𝑢𝑛𝑡 cosh(𝛾𝑙)−1 2! 𝑗 2 𝜔 2 𝐿𝐶𝑙2 𝑗𝜔𝐶𝑙 𝑗𝜔𝐶𝑙𝑖𝑛𝑒
= = = = = [S], (77)
2 𝑍𝑠𝑒𝑟𝑖𝑒𝑠 𝑍𝑠𝑒𝑟𝑖𝑒𝑠 2𝑗𝜔𝐿𝑙 2 2

where 𝑙𝐶 = 𝐶𝑙𝑖𝑛𝑒 .

Short-Length Lines

In the case of short lines less than 100 miles in length, the effect of capacitive reactive power is small
compared to the system strength to provide Q and can often be neglected. This results in the
simplified circuit model shown in Fig.28c with only the series impedance, and where the series
line resistance is neglected in more simplified preliminary studies.

References

1. George G. Karady and Keith E. Holbert, "Electrical Energy Conversion and Transportation - An
Interactive Computer-Based Approach", IEEE Press, Wiley-Interscience, 2005

2. N. Mohan, "First Course on Power Systems", MNPERE, Minneapolis, 2006, ISBN 0-9715292-7-2.

3. Duncan J. Glover, Mulukutla S. Sarma, and Thomas J. Overbye, "Power Systems - Analysis and
Design", 4th Ed., Thomson, 2008

4. William H. Hayt, Jr., "Engineering Electromagnetics", McGrawHill.

5. Electric Power Research Institute (EPRI), "Transmission Line Reference Book: 345kV and above", 2nd
edition.

6. Electric Power Research Institute (EPRI),www.epri.com

17
ECE 43200/53201 Power Systems Fall 2020

Appendix A
Skin Effect - Propagation of Plane Wave in Good Conductors [4]

Let’s consider a wave that is established by an electromagnetic field existing in some external dielectric that adjoins the
conductor surface. We shall see that the primary transmission of energy must take place in the region outside the
conductor, because all time-varying fields attenuate very quickly within a good conductor.
The good conductor has a high conductivity and large conduction currents. The energy represented by the wave traveling
through the material therefore decreases as the wave propagates because ohmic losses are continuously present. The ratio
of conduction current density to the displacement current density in a material is given by σ ⁄ (ωϵ). The general expression
for the propagation constant, 𝛾 = 𝛼 + 𝑗𝛽, is
𝜎
𝛾 = 𝑗𝜔√𝜇𝜖√1 − 𝑗
𝜔𝜖
which is simplified for the case of conductors when is σ ⁄ (ωϵ) >> 1 to obtain

𝜎
𝛾 = 𝑗𝜔√𝜇𝜖√−𝑗 = j√−jωμσ,
𝜔𝜖
° ° 1 1
but √−𝑗 = √𝑒 −𝑗90 = 𝑒 −𝑗45 = −𝑗 , therefore
√2 √2

𝛾 = (1 + 𝑗)√𝜋𝑓𝜇𝜎
Hence
𝛼 = 𝛽 = √𝜋𝑓𝜇𝜎

If we assume only an Ex component traveling in the + z direction, then

𝐸𝑥 = 𝐸𝑥0 𝑒 −𝑧𝛼 cos (𝜔𝑡 − 𝑧𝛼)

We may tie this field in the conductor to an external field at the conductor surface. We let the region z > 0 be the good
conductor and the region z < 0 be a perfect dielectric.
The conduction current density at any point within the conductor is directly related to E:

𝐽𝑥 = 𝜎𝐸𝑥 = 𝜎𝐸𝑥0 𝑒 −𝑧𝛼 cos (𝜔𝑡 − 𝑧𝛼)

The last two equations contain a wealth of information. Considering first the negative exponential term, we find an
exponential decrease in the conduction current density and electric field intensity with penetration into the conductor
(away from the source).
The exponential factor is unity at z = 0 and decreases to e -1 =0.368 when 𝑧 = 1/√𝜋𝑓𝜇𝜎. This distance is denoted by δ
and is termed the depth of penetration, or the skin depth:
𝟏 1 1
𝜹= = = ,
√𝝅𝒇𝝁𝝈 𝛼 𝛽
where f is the frequency, 𝜇 is the permeability, and 𝜎 is the conductivity (reciprocal of resistivity 𝜌),

Next, let us determine expressions for the velocity and wavelength within a good conductor. We already have
1
𝛼=𝛽= = √𝜋𝑓𝜇𝜎
𝛿
2𝜋
Since the wave length is defined as the distance that provides a shift of 2π rad, then 𝛽 = , and we find the
𝜆
wavelength to be: 𝜆 = 2𝜋𝛿.
Also
𝜔
𝑣 = = 𝜔𝛿.
𝛽

For example, for aluminum at 60Hz, with 𝛿 = 18.75𝑚𝑚, the wave velocity is 𝑣 = 𝜔𝛿 = 7𝑚/𝑠 or about
15.65mi/h which is a very small value. In free space, a 60Hz wave has a wavelength of 3100 mi and travels
at the velocity of light.
18