Transmission Lines

Transmission Lines
1 Introduction
For efficient point-to-point transmission of power and information, the source energy must be
directed or guided. Here, the transverse electromagnetic (TEM) waves guided by
transmission lines will be discussed. The characteristics of TEM waves here are the same as
those for a uniform plane wave propagating in an unbounded dielectric medium. Basically,
these TEM waves are like “guided” waves on a line in contrast with “unguided” waves in free
space for radiated waves.
There are three common types of guiding structures that support TEM waves, namely,
parallel-plate transmission line, two-wire transmission line, and coaxial transmission line.
The following points should be first pointed out here:
1. Only perfect conductor supports TEM mode whose electric field has only component
perpendicular to the line. For good conductors, longitudinal component can exist due to
the line currents passing through the “imperfect” conductors. This mode is referred to as
the “quasi-TEM” mode.
2. In addition, if the surrounding medium is lossy (either through Ohmic loss or dielectric
loss of the medium), this additional loss should also be taken into account.

2 General Transmission-Line Equations

In general, transmission lines are used when the physical dimensions of electric networks are
usually a considerable fraction of a wavelength (typically, larger than quarter-wavelength)
and may even be many wavelengths long, whereas ordinary circuit theory is applied to
networks considerably much smaller than the operating wavelength. A transmission line is a
distributed-parameter network, or “distributed” circuit, and must be described by circuit
parameters that are distributed throughout its length, while ordinary electric networks are
considered “lumped” circuits, whose elements are discrete and currents flowing in lumped-
circuit elements do not vary spatially over the elements.
Now, consider a “lossy” transmission line consisting of two wires in xz-plane shown in
Figure 1. Applying Faraday’s law
∂B
∫ E ⋅ dl = − ∫
C S ∂t
⋅ ds (1.1)

to the contour C and surface S shown in (a) yields

∂By ( x, z; t )
∫x [E x ( x, z + ∆z; t ) − E x ( x, z; t )]dx + ∫z [E z ( x1 , z; t ) − E z ( x2 , z; t )]dz = −∫z ∫x
x2 z + ∆z z + ∆z x2
dxdz
1 1 ∂t
(1.2)
Define the voltage between the wires
x2
v( z; t ) = − ∫ E x ( x, z )dx , (1.3)
x1

then, v( z + ∆z; t ) − v( z; t ) = − ∫ [E x ( x, z + ∆z; t ) − E x ( x, z; t )]dx .

x2
(1.4)
x1

1
Transmission Lines

Fig. 1: Contours and surfaces for the derivation of the transmission-line equations

Suppose a current i (z ) exists in the upper wire in the positive z direction and returning in the
lower wire. Assume that the losses in the wires can be lumped as an impedance through
which i (z ) passes. The lossy nature of the conductors will result in the resistance per unit
length, R. There is also the internal inductance per unit length Li due to the current i (z )
partially penetrating the wires, which are not perfect conductors. Thus,
∂i ( z; t )
∫z [E z ( x1 , z ) − E z ( x2 , z )]dz = − R∆zi( z; t ) − Li ∆z ∂t .
z + ∆z
(1.5)

Furthermore, the RHS of (1.2) is related to the magnetic flux external to the wires produced
by the current, i.e.,
z + ∆z x2
i ( z; t )∆zLe = − ∫ ∫ By ( x, z; t )dxdz (1.6)
z x1

where Le denotes the external inductance per unit length.

Substituting (1.4),( 1.5),( 1.6) into (1.2) yields
 ∂i ( z; t ) 
v( z + ∆z; t ) − v( z; t ) = − ( Ri( z; t ) + ( Li + Le ) ∆z (1.7)
 ∂t 
Dividing both sides by ∆z and taking the limit as ∆z → 0 , we obtain
∂v( z; t ) ∂i ( z; t )
− = Ri( z; t ) + L , (1.8)
∂z ∂t
where L = Li + Le. Now, consider Figure 1(b). Applying the continuity equation to the surface
S yields
dQ
∫S J ⋅ ds = − dt (1.9)

where Q is the total charge enclosed by S. Over the ends of the cylinder
∫Se
J ⋅ ds = i ( z + ∆z; t ) − i ( z; t ) . (1.10)

Over the sides of the cylinder, the conductivity of the medium results in a transverse
conduction current through So so that
∫So
J ⋅ ds = G∆zv( z; t ) (1.11)

where G is the per-unit length conductance between the wires produced by the lossy medium.
In addition, let C denote the per-unit length capacitance, then over the length ∆z
Q = C∆zv( z; t ) (1.12)

2
Transmission Lines

Inserting (1.10),( 1.11),( 1.12) into (1.9) yields

∂v( z; t ) i(z,t) i(z+∆z,t)
i ( z + ∆z; t ) − i ( z; t ) + G∆zv( z; t ) = −C∆z (1.13) L∆z
∂t R∆z
C∆z
Rewriting (1.13), then dividing both sides by ∆z and v(z,t) G∆z v(z+∆z,t)
taking the limit as ∆z → 0 , we obtain
∂i ( z; t ) ∂v( z; t ) ∆z
− = Gv( z; t ) + C (1.14)
∂z ∂t
Fig. 2: The per-unit length model of a
Therefore, the per-unit length model can be shown in
lossy transmission line
Figure 2. Equations (1.8) and (1.14) are a pair of
first-order partial differential equations in v(z;t) and i(z;t), which are called the general
transmission line equations or telegrapher’s equations.
For the time-harmonic electromagnetic field, the use of phasor representation helps simplify
the analysis. Introducing the following quantities:
[ ] [ ] [ ]
E ( x, z; t ) = Re E( x, z )e jωt ; H ( x, z; t ) = Re H ( x, z )e jωt ; B( x, z; t ) = Re B( x, z )e jωt ;
[ ] [ ]
J ( x, z; t ) = Re J ( x, z )e jωt ; Qˆ ( x, z; t ) = Re Q( x, z )e jωt ;
[ ] [
v( z; t ) = Re V ( z )e jωt ; i ( z; t ) = Re I ( z )e jωt ],
where E, H, B, J are vector phasors, and V, I are (scalar) phasors, then equations (1.8) and
(1.14) become
dV ( z ) ˆ dI ( z ) ˆ
− = ZI ( z ) (1.15a) − = YV ( z ) (1.15b),
dz dz
where Zˆ = R + jωL and Yˆ = G + jωC . Equations (1.15a) and (1.15b) are called time-
harmonic transmission-line equations, which is useful for steady state analysis with
sinusoidal waveform. Note that phasors are complex quantities in general.
From (1.15a) and (1.15b), one obtains the wave equations for the lossy case as follows:
d 2V ( z ) ˆ ˆ d 2 I ( z) ˆ ˆ
2
= ZYV ( z ) = γ V ( z ) (1.16a)
2
2
= ZYI ( z ) = γ 2 I ( z ) (1.16b),
dz dz
where γ is the propagation constant given by
γ = α + jβ = ZˆYˆ = ( R + jωL)(G + jωC ) (1.17)
(α is called attenuation constant [Np/m] and β is called phase constant [rad/m]). These
equations are similar to wave equations in lossy media which can be derived from Maxwell’s
equations. They would reduce to those for the lossless case when R = G = 0, which are quite
similar to those for the waves in lossless media. The solutions to (1.16) become
V ( z ) = V0+ e −γz + V0− eγz = V0+ e −αz e − jβz + V0− eαz e jβz (1.18a)
I ( z ) = I 0+ e −γz + I 0− eγz = I 0+ e −αz e − jβz + I 0− eαz e jβz (1.18b),
where the plus and minus superscripts denote waves traveling in the +z and –z directions,
respectively, and V0±, I0± denote wave amplitudes. Substituting (1.18) into (1.15) yields the
relationships between voltage and current wave amplitudes, which are defined as the
characteristic impedance:
V0+ V0− R + jωL γ Zˆ R + j ωL
Z0 = = − = = = = . (1.19)
I0+
I0−
γ G + j ωC Yˆ G + j ωC

3
Transmission Lines

Clearly, using Z0, (1.18b) can be rewritten as

V0+ −γz V0− γz V0+ −αz − jβz V0− αz jβz
I ( z) = e − e = e e − e e . (1.20)
Z0 Z0 Z0 Z0
Note that V(z) and I(z) are similar to the plane wave (electric field, magnetic field) in a lossy
medium. The wave decays in z direction due to the e-αz and eαz terms. The velocity of
propagation or the phase velocity is given by
ω
up = with β = Im(γ ) (1.21)
β
Attenuation constant has unit Np/m where 1 Np/m means attenuation factor 1/e after 1-m
propagation. Typically represented in dB unit where αdB = 20log10eα = 8.686α and α = 1
Np/m -> 8.686 dB/m.
Note that γ and Z0 are characteristic properties of a transmission line, which depend on R, L,
G, C and ω.
Special Cases For the following special cases, the expressions are simplified.
1. Lossless Line (R = 0, G = 0)
a) Propagation constant:
γ = α + jβ = jω LC ; α = 0; β = ω LC (linear function of ω)
b) Phase velocity
ω 1
up = = (constant)
β LC
c) Characteristic impedance
L L
Z 0 = R0 + jX 0 = ; R0 = ; X0 = 0
C C
For the lossless lines, γ = α + jβ = jβ ; u = 1 / LC ; β = ω / u = 2πf / u . Thus, the propagation
constant (the phase constant) is linearly dependent on frequency, i.e., linear phase, but the
phase velocity is independent of frequency.
2. Low-loss Line or slightly lossy (R << ωL, G << ωC) low-loss conditions are more easily
satisfied at very high frequencies.
a) Propagation constant:
1/ 2 1/ 2
 R   G   R  G 
γ = α + jβ = jω LC 1 +  1 +  ≅ jω LC 1 + 1 + 
 j ωL   jωC   j 2ωL  j 2ωC 
 1  R G 
≅ jω LC 1 +  + 
 j 2ω  L C 
1 C L
α =  R +G  ; β = ω LC (linear function of ω)
2 L C 
b) Phase velocity
ω 1
up = ≅ (constant)
β LC
c) Characteristic impedance

4
Transmission Lines

1/ 2 −1 / 2
L R   G  L 1  R G 
Z 0 = R0 + jX 0 = 1 +  1 +  ≅ 1+  −  ;
C j ωL   jωC  C j 2ω  L C 
L L 1 R G
R0 ≅ ; X0 ≅ −  −  ≅ 0.
C C 2ω  L C 
3. Distortionless Line (R/L = G/C)
R G
For the lossy lines, if the condition = is satisfied, then
L C
a) Propagation constant:
L C  C
γ = ( R + jωL)(G + jωC ) = R(1 + jω )G (1 + jω ) = RG 1 + jω  = α + jβ
R G  G
C R L
or α = RG ; β = ω RG = ωC = ωC = ω LC
G G C
ω 1
b) Phase velocity: u p = = (constant)
β LC
c) Characteristic impedance
R + j ωL R(1 + jωL / R) R L L
Z 0 = R0 + jX 0 = = = = ; R0 = ; X0 = 0.
G + jωC G (1 + jωC / G ) G C C
Thus, α does not depend on frequency, β is linearly dependent on frequency, which is similar
to the lossless lines, and there is no distortion. Thus, it is called “distortionless” lines. If two
frequency components have different phase velocity, then the signal will distort. For
example, let y = cos(ω1t-β1z) cos(ω2t-β2z), and ω2 = 5ω1, β1l = 2π. Figure 3 shows the signal
at z=0 and z=l for different u1, u2.
In general, the phase constant is not a linear function of ω, thus it will lead to a up, which
depends on frequency. As the different components of a signal propagate along the line with
different velocities, the signal suffer dispersion. A general, lossy transmission line is
therefore dispersive, as is a lossy dielectric.
Signal at z=L
1

0.8
Signal at z=0
1
0.6
0.8
0.4 u2=u1
0.6
u2=.9u1
0.4 0.2
u2=.95u1
u2=1.03u1
0.2 0
u2=1.08u1
0 -0.2
-0.2
-0.4
-0.4
-0.6
-0.6
-0.8
-0.8

-1
-1
0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7
ωt ωt

Fig. 3 : Signal distortion

Example 2.1 A distortionless line has Z0 = 60 Ω, α = 20 mNp/m, u = 0.6c, where c is the
speed of light in a vacuum. Find R, L, G, C and λ at 100 MHz.

5
Transmission Lines

3 Transmission-Line Parameters
Capacitance Recall that the total charge is related to the voltage by Q = CV, the capacitance
can be found from

C=
∫ D ⋅ ds = ∫ εE ⋅ ds .
S S

− ∫ E ⋅ dl − ∫ E ⋅ dl
L L

Conductance (the shunt resistance) When the dielectric medium is lossy (having a small but
nonzero conductivity), a current will flow from the positive to the negative conductor, and a
current density field will be established in the medium. Using the Ohm’s law, J = σE, yields

G=
I ∫ J ⋅ ds = ∫SσE ⋅ ds .
= S
V − ∫ E ⋅ dl − ∫ E ⋅ dl
L L

For homogeneous media (or when σ and ε have the same spatial dependence) the following
relationship holds:
C ε
= ,
G σ
which is derived from two above equations.
Inductance The inductance can be directly calculated from
Φ ∫SB ⋅ ds ∫ µH ⋅ ds .
L= = = S
I ∫ J ⋅ ds
S ∫ σE ⋅ ds
S

However, a comparison of the propagation constant for the TEM wave on a transmission line
with R = 0 and that for the wave in a medium with constitutive parameters (µ, ε, σ)
1/ 2 1/ 2
 G   σ 
γ = jω LC 1 +  ; γ = jω µε 1 + 
 j ωC   jωε 
together with C/G = ε/σ yields
LC = µε.
Resistance Series resistance R is determined by introducing a small Ez as a slight perturbation
of the TEM wave and by finding the Ohmic power dissipated in a unit length of the line.
Typically, it is calculated by the resistance due to the skin depth.
i) Parallel plate transmission line Let w, d denote the width, separation and assume the
medium between plates has constitutive parameters (µ, ε, σ), then
w d w
C = ε ; L = µ ;G = σ .
d w d

6
Transmission Lines

The per-unit-length resistance can be calculated from R = 1/σcS, where σc denotes the
conductivity of the conducting plate and S denotes the cross section which equals the product
of width w and skin depth δ. Thus,
2 2 πfµc
R= = ,
σ c wδ w σ c
where µc denotes the permeability of the plate and f is the frequency. Also, the factor 2 comes from
the fact that the transmission line consists of two plates. Note also that a good conductor is assumed
here in the calculation of skin depth, i.e.,
1
δ= .
πfµcσ c
ii) Two-wire transmission line Assume the two conducting wires of radius a, separated by D in the
medium with constitutive parameters (µ, ε, σ), then
πε µ πσ
C= ;L = cosh −1 ( D / 2a ) ; G = , and
−1
cosh ( D / 2a) π −1
cosh ( D / 2a )
2 2 1 1 πfµc
R= = = = .
σ c S σ c 2πaδ σ cπaδ πa σ c
iii) Coaxial transmission line Assume the (inner, outer) radii be (a, b) and the medium with
constitutive parameters (µ, ε, σ), then
2πε µ b 2πσ
C= ;L = ln ; G = , and
ln(b / a ) 2π a ln(b / a)
1 1 1  1 1  1  1 1  πfµc
R= + =  + =  +  .
σ c Si σ c So σ c 2πδ  a b  2π  a b  σ c
Note that (Si, So) in the equation above denote the cross sections of the (inner,outer) conductors,
respectively.

4 Wave Characteristics on Finite Transmission Lines (Steady State Analysis)

Consider the finite transmission line shown in Fig. 4. The length of the line is l. Then,

Zg IL
Ii
Vg Vi
Zin (γ,Z0) VL ZL

z z'=l-z
z=0 z=l
z'=l z'= 0
Fig. 4: Finite transmission line terminated with load impedance ZL.
V  VL V0+ e −γ + V0− eγ V0+ e −γ + V0− eγ
ZL =   = = + −γ = (4.1).
 I  z = I L I 0 e + I 0− eγ V0+ e −γ / Z 0 − V0− eγ / Z 0
V0+ −γz V0− γz
Recall that V ( z ) = V0+ e −γz + V0− eγz (4.2a); I ( z ) = e − e (4.2b), thus,
Z0 Z0

7
Transmission Lines

V0+ −γ V0− γ

VL = V () = V0+ e −γ + V0− eγ (4.3a); I L = I () = e − e (4.3b).
Z0 Z0
Solving for V0+ ,V0− from the above equations yields

V0+ =
1
(VL + I L Z 0 )eγ = I L (Z L + Z 0 )eγ (4.4a); V0− = 1 (VL − I L Z 0 )e −γ = I L (Z L − Z 0 )e −γ (4.4b).
2 2 2 2
Substituting (4.4a),( 4.4b) into (4.2a),( 4.2b) yields
[ ]
V ( z ) = L (Z L + Z 0 )eγ ( − z ) + (Z L − Z 0 )e −γ ( − z ) (4.5a);
I
2
[ ]
I ( z ) = L (Z L + Z 0 )eγ (  − z ) − (Z L − Z 0 )e −γ (  − z ) (4.5b).
I
2Z 0
Introducing a new variable z’=l -z, then (4.5) can be rewritten as

[ ] [ ]
V ( z ' ) = L (Z L + Z 0 )eγz ' + (Z L − Z 0 )e −γz ' (4.6a) I ( z ' ) = L (Z L + Z 0 )eγz ' − (Z L − Z 0 )e −γz ' (4.6b).
I
2
I
2Z 0
The use of hyperbolic functions simplifies the equations above to be
V ( z ' ) = I L [Z L cosh γz '+ Z 0 sinh γz '] (4.7a) I ( z ' ) = L [Z L sinh γz '+ Z 0 cosh γz '] (4.7b).
I
Z0
The ratio of (V(z’)/I(z’) is the impedance when one looks toward the load end of the line at a
distance z’ from the load, which is given by
V ( z' ) Z cosh γz '+ Z 0 sinh γz ' Z + Z 0 tanh γz '
Z ( z' ) = = Z0 L = Z0 L (4.8).
I ( z' ) Z L sinh γz '+ Z 0 cosh γz ' Z 0 + Z L tanh γz '
At the source end of the line z’=l, the generator looking into the line sees an input impedance
Zin, which is given by
Z + Z 0 tanh γ
Z in = Z ( z ' = ) = Z 0 L (4.9).
Z 0 + Z L tanh γ
Then the load impedance and the transmission line can be replaced by the input impedance
Zin as depicted in Fig. 5. The input voltage Vi and input current Ii in Fig. 5 are found from the
equivalent circuit as follows:
Z in
Vi = Vg (4.10) g Z
Z g + Z in
Vg
Ii
Ii =
Z g + Z in
(4.11) Vg Zin Vi
The average power delivered by the generator to the
input terminals of the line is
Fig. 5: Equivalent circuit for finite
(Pav )i = 1 Re[Vi I i *] , (4.12) transmission line at generator end
2
and the average power delivered to the load is
2

(Pav )L 1
= Re[VL I L *] =
1 VL 1 2
RL = I L RL (4.13).
2 2 ZL 2
For a lossless transmission line, conservation of power requires that (Pav)i = (Pav)L.
Note that once Vi, Ii are obtained, V0± can be calculated from (4.2a),(4.2b) as

8
Transmission Lines

V0+ V0−
V ( z = 0) = Vi = V0+ + V0− ; I ( z = 0) = I i = − ,
Z0 Z0
then the voltage and current along the transmission line can be found.
Voltage Reflection Coefficient and Voltage Standing Wave Ratio (VSWR)
Define the “complex” voltage reflection coefficient as
V0− eγz V0− 2γz
Γ( z ) = + −γz = + e , (4.14)
V0 e V0
then (2a),(2b) can be rewritten as
V0+ −γz
V ( z ) = V0+ e −γz [1 + Γ( z )] (4.15a) I ( z) = e [1 − Γ( z )] (4.15b).
Z0
It follows that
V ( z) 1 + Γ( z )
Z ( z) = = Z0 (4.16)
I ( z) 1 − Γ( z )
and
V ( ) 1 + Γ ( ) 1 + ΓL
Z ( z = ) = = Z L = Z0 = Z0 . (4.17)
I ( ) 1 − Γ ( ) 1 − ΓL
Therefore, the voltage reflection coefficient at the load is given by
Z − Z0
ΓL = L =| ΓL | e jθΓ . (4.18)
Z L + Z0
Then from (4.14),
V0−
+
= Γ()e −2γ = ΓL e −2γ , (4.19)
V0
thus, Γ( z ) = ΓL e 2γ ( z − ) . (4.20)
The magnitude ratio of the maximum to the minimum voltages along a finite, terminated line
is defined as the voltage standing wave ratio (VSWR), i.e.,
| V | 1+ | Γ |
VSWR = S = max = , (4.21)
| Vmin | 1− | Γ |
which measures “degree of mismatch”. S=1 denotes the matched-load condition.
Power Flow Due to losses, the power is a function of the position on the line.
 *
 V0+ −αz − jβz V0− αz jβz  
(
Pav ( z ) = Re[V ( z ) I * ( z )] = Re V0 e e
1 1  + −αz − jβz
+ V0− eαz e jβz ) e e − e e  (4.22)
2 2   Z0 Z0  
 
Recall that
V0+ −γz
V ( z ) = V0+ e −γz [1 + Γ( z )] (4.15a) I ( z) = e [1 − Γ( z )] (4.15b),
Z0
then,

9
Transmission Lines

 * 
 V0+ −γz 
Pav ( z ) = Re V0 e [1 + Γ( z )] e  [1 − Γ( z )]
1  + −γz *
 
2   Z0  
  (4.23)
1 + 2 −2αz 1 − Γ( z ) + j 2 Im Γ( z ) 
2

= V0 e Re  
2  Z 0* 
The law of energy conservation requires that the rate of decrease of Pav(z) with distance along
the line equals the time-average power loss PL per unit length. Thus,
∂P ( z )
− av = PL ( z ) = 2αPav ( z ) , (4.24)
∂z
from which we obtain the following formula:
P ( z)
α= L (Np/m) (4.25)
2 Pav ( z )
Example 4.1 A certain transmission line operating at ω = 106 rad/s has α = 8 dB/m, β = 1
rad/m, and Z0 = 60 + j40 Ω, and is 2 m long. If the line is connected to a generator of 10∠0
V, Zg = 40 Ω and terminated by a load of 20 + j50 Ω, determine
a) The input impedance
b) The sending end current
c) The power at the sending end and the load

10
Transmission Lines

5 Steady State Analysis of Lossless Transmission Lines

In most practical applications, low-loss transmission lines are used and thus they can be
approximated as lossless to simplify the analysis. As previously mentioned, for lossless lines,
γ = α + jβ = jω LC (5.1) ; α = 0 (5.2); β = ω LC (5.3).
Hence, the input impedance becomes
Z + jZ 0 tan β
Z in = Z 0 L . (5.4)
Z 0 + jZ L tan β
where Z0 is real. Note that βl = ωl /up = 2πl /λ, thus it is more convenient to express l in terms
of wavelength, i.e., λ = up/f. Also, the voltage reflection coefficient becomes
V0− e jβz V0− j 2 βz
Γ( z ) = + − jβz = + e = ΓL e j 2 β ( z − ) = ΓL e jθΓ . (5.5)
V0 e V0
and V ( z ) = V0+ e − jβz + V0− e jβz = V0+ e − jβz [1 + Γ( z )] . (5.6)
+ −
Let V0+ = V0+ e jθ , V0− = V0− e jθ , then Im
V0− )|
Γ(z
Γ( z ) = = ΓL , and θ Γ = 2 βz − θ + + θ − = 4πz / λ − θ + + θ − . |1+

z)|
θΓ

|Γ(
+
V0 1 Re
1+Γ(z) can be illustrated in the complex plane as shown in Fig.
6, which is often referred to as the crank diagram. Fig. 6: The crank diagram
It follows that
V ( z ) = V0+ e − jβz [1 + Γ( z )] = V0+ 1 + Γ( z ) , thus
V ( z ) max = V0+ (1 + Γ( z ) ) (5.7) V ( z ) min = V0+ (1 − Γ( z ) ) (5.8)
Clearly, for a matched load |ΓL|=0 and |V(z)|max=|V(z)|min.
The VSWR then becomes
| V | 1+ | Γ( z ) | 1+ | ΓL |
VSWR = S = max = = . (5.9)
| Vmin | 1− | Γ( z ) | 1− | ΓL |
S −1
Alternatively, Γ( z ) = ΓL = . (5.10)
S +1
Power Flow The average power is given by
 *
− jβz  V0
+
V0− jβz  
Pav ( z ) = Re[V ( z ) I * ( z )] = Re V0 e
1
(
1  + − jβ z
)
+ V0 e  e − jβ z
− e 
2 2   Z0 Z0  
 
(5.11)
 *  2

( )
+
 V + e − jβ z  1 V0
= Re V0+ e − jβz (1 + Γ( z ) ) 0  (1 − Γ * ( z ) ) =
1
1 − ΓL
2

2   Z0   2 Z0
 
(5.11) can be found from summing power traveling in +z and –z directions as follows:

1  + − jβz V0 * jβz  1 V0
+ + 2
+ 1 +
[ +
 ]
P ( z ) = Re V ( z ) I * ( z ) = Re V0 e
2 
e  = (5.12)
 2 Z0
av
2 Z0
 

11
Transmission Lines

1  − − jβz V0 * jβz  1 V0
− − − 2 2
1 V0 ΓL
2
− 1 −
[
−

2 
]
P ( z ) = Re V ( z ) I * ( z ) = Re V0 e e = = (5.13)
 2 Z0
av
2 Z0 2 Z0
 
2

( ).
+
+ − 1 V0
Pav ( z ) = P ( z ) + P ( z ) = 1 − ΓL
2
av av (5.14)
2 Z0
Note that the power is uniform along the transmission line, i.e., independent of the position
on the line, thus the power delivered to the line is the same as the power delivered to the load.
P P−
Clearly, av ,reflected = av+ = ΓL .
2
(5.15)
Pav ,incident Pav
Also, the power delivered to the line and to the load can be found from
1 V (0)V * (0)  1 Vi
2

Re[V (0) I * (0)] = Re 

1 Z in
Pav ,to line =  = cos ∠Z in , V (0) = Vg (5.16)
2 2  *  2 Z Z + Z
 Z in  in in g

1 V ()V * ()  1 VL
2

= Re[V () I * ()] = Re 

1
Pav ,to load = cos ∠Z L . (5.17)
2 2  Z *L  2 Z L
Special Cases
a) Short-circuit load (ZL = 0)
Z in = jZ 0 tan β = jZ 0 tan(2π / λ ) , which implies an “inductive” reactance.
|ΓL|=1 → No power delivered to load (i.e., the load does not consume any power).
b) Open-circuit load (ZL = ∞)
Z in = Z 0 /( j tan β) = − jZ 0 / tan(2π / λ ) , which implies a “capacitive” reactance.
|ΓL|=1 → No power delivered to load (i.e., the load does not consume any power).
c) Matched load (ZL = Z0)
Z in = Z 0
|ΓL|=0 → Pav = Pav+ → all power is delivered to the load.
d) Quarter-wavelength transmission line (l = λ/4)
Since βl =π/2, tan βl → ∞, thus
Z 02
Z in = .
ZL
It follows that Γ(0) = ΓL e − j 2 β = −ΓL , and
for ZL = 0, Γ L= -1, Γ(0) = 1, Zin=∞ : short-circuit → open-circuit
ZL = ∞, Γ L= 1, Γ(0) = -1, Zin=0 : open-circuit → short-circuit
e) Transmission lines’ length equal multiples of half-wavelength (l = nλ/2)
Since βl =nπ, tan βl = 0, thus Zin = ZL, i.e., input impedance is equal to load impedance.
Note also that since function tan(x) is a “periodic” function with period π,
tan β = tan( β ± nπ ) = tan β ( ± nπ / β ) = tan β ( ± nλ / 2), n = 1,2,3,  ,
thus adding multiples of half-wavelength does not change the input impedance.

12
Transmission Lines

Example 5.1 A 10-m section of lossless transmission line having Z0 = 50 Ω and u = 200 m/µs
is driven by a 26-MHz generator having an open-circuit voltage of Vg = 100 V and generator
impedance Zg = 50 Ω. The line terminated in a load impedance of ZL = 100 + j50
Ω. Determine the input impedance to the line and the instantaneous voltage at the input to the
line and the at the load, i.e., V(0, t) and V(l, t).

Example 5.2 A 1-m section of lossless transmission line having C = 200 pF/m and L = 0.5
µH/m is driven by a 30-MHz generator having an open-circuit voltage of Vg = 1 V and
generator impedance Zg = 10 Ω. The line terminated in a load impedance of ZL = 100 + j50
Ω. Determine the load voltage and the average power delivered to the line and to the load.

13
Transmission Lines

6 Transient Analysis
In this section, the transient behavior of lossless transmission lines will be discussed. Such
practical situations include cases where non-time-harmonic signals are used or where the
conditions are not steady-state. Examples are digital (pulse) signals in computer networks and
sudden surges in power and telephone lines.
Recall that the general transmission line equations previously derived are given for lossless
transmission lines by
∂v( z; t ) ∂i ( z; t ) ∂i ( z; t ) ∂v( z; t )
− =L (6.1a) and − =C (6.1b)
∂z ∂t ∂z ∂t
and the wave equations derived from (6.1a), (6.1b) are
∂ 2v ( z; t ) ∂ 2v( z ) ∂ 2i ( z ; t ) ∂ 2i ( z )
= LC (6.2a) and = LC (6.2b),
∂z 2 ∂t 2 ∂z 2 ∂t 2
The general solutions to (6.2a), (6.2b) are
v( z; t ) = v + (t − z / u ) + v − (t + z / u ) (6.3a) i ( z; t ) = i + (t − z / u ) + i − (t + z / u ) (6.3b)
+ - + -
where +,- signs denote the waves traveling in +z, -z, respectively. v , v , i , i depend on t, z
and u.
Proof

As before, i± can be related to v± via the characteristic impedance as

i + (t − z / u ) = v + (t − z / u ) / Z 0 ; (6.4a) i − (t + z / u ) = −v − (t + z / u ) / Z 0 (6.4b)
where Z 0 = L / C . Hence,
[ ]
i ( z; t ) = v + (t − z / u ) − v − (t + z / u ) / Z 0 (6.5)
Now, consider a terminated transmission line where a generator with generator resistance Rg
is applied at t = 0, as depicted in Fig. 7.
t=0
+
+ +
Rg
Vg v(0,t) v+(z,t) v(z1,t) (u,Z0) -
v (z,t) v(l,t) RL
- -
-
z=0 z = z1 z=l

Fig. 7: A generator applied to a terminated lossless line at t = 0.

At the load end (z = l), v(; t ) = RLi (; t ) (6.6)
The discontinuity at the load results in the reflection. Recall the voltage reflection coefficient
mentioned previously, which is given by
v − (t +  / u )
ΓL = + . (6.7)
v (t −  / u )
Using ΓL to rewrite v, i at the load as

14
Transmission Lines

v(; t ) = v + (t −  / u ) + v − (t +  / u ) = v + (t −  / u )(1 + ΓL ) (6.8a)

{ + −
} +
i (; t ) = v (t −  / u ) − v (t +  / u ) / Z 0 = v (t −  / u )(1 − ΓL ) / Z 0 (6.8b)
Likewise, the current reflection coefficient can be defined as
i − (t +  / u )
− ΓL = . (6.9)
i + (t −  / u )
Substituting (6.8a),(6.8b) into (6.6) yields
v + (t −  / u )(1 + ΓL ) = RL v + (t −  / u )(1 − ΓL ) / Z 0 or
1 + ΓL
RL = Z 0 . (6.10)
1 − ΓL
Solving for ΓL yields
R − Z0
ΓL = L , (6.11)
RL + Z 0
which is the same as that for the steady-state analysis, except this ΓL is real. Note that at the
load, reflection is like a mirror; reflected wave v-, a replica of v+ which is flipped around and
is multiplied by ΓL, is a “mirror image” of v+.
Now, consider the portion of the line at the generator end z = 0. Let T = l /u be the time delay
and the generator is turned on at t = 0.
During 0 ≤ t ≤ 2T , no backward-traveling waves will appear at z = 0. Thus,
v(0; t ) = v + (t − 0 / u ) , 0 ≤ t ≤ 2T (6.12a) i (0; t ) = v + (t − 0 / u ) / Z 0 , 0 ≤ t ≤ 2T (6.12b)
Hence, the ratio of the voltage and current on the line is Z0 for 0 ≤ t ≤ 2T , i.e., the input
impedance seen by the generator is Z0 during 0 ≤ t ≤ 2T . Thus,
v(0; t ) = Vg (t ) − Rg i (0; t ) = Vg (t ) − Rg v(0; t ) / Z 0 , (6.13)
Z0
so that v(0; t ) = Vg (t ) , 0 ≤ t ≤ 2T . (6.14)
Z 0 + Rg
Therefore, the initial v+ has the same shape as Vg(t) and is multiplied by Z0/(Z0+Rg). Likewise,
V (t )
i (0; t ) = g , 0 ≤ t ≤ 2T . (6.15)
Z 0 + Rg
After 2T, there exist backward-traveling waves at the generator end. The discontinuity at this
end will cause the reflection in the same manner as that at the load end. Define the voltage
reflection coefficient at the generator as
v − (t + 0 / u )
Γg = + , (6.16) then
v (t − 0 / u )
v(0; t ) = v + (t − 0 / u ) + v − (t + 0 / u ) = v + (t − 0 / u )(1 + Γg ) (6.17a)

[ ]
i (0; t ) = v + (t − 0 / u ) − v − (t + 0 / u ) / Z 0 = v + (t − 0 / u )(1 − Γg ) / Z 0 (6.17b)
Since v(0; t ) = Rg i (0; t ) ,
v + (t − 0 / u )(1 + Γg ) = Rg v + (t − 0 / u )(1 − Γg ) / Z 0 or
1 + Γg
Rg = Z 0 . (6.18)
1 − Γg

15
Transmission Lines

Solving for Γg yields

R − Z0
Γg = g , (6.19)
Rg + Z 0
which is similar to the voltage reflection coefficient at the load. The reflections at both ends
will continue until the system reaches the steady-state.
Reflection Diagram or Bounce Diagram To calculate the
voltage and current at a particular time and location on a t
transmission line with arbitrary resistive load tends to be 6T
tedious and difficult to visualize when there are many
reflected waves. In such cases, the graphical construction of g
Γ 2 Γ 2V1+
L 5T
a reflection diagram (or a bounce diagram) is helpful. Fig. 8 t5
+ 4T 2 +
shows a typical reflection diagram. V1 in the figure denotes
t4 Γg Γ V1 L

the initial voltage, i.e., v(0;t) in (6.14). It will take the time T 3T
Γg ΓLV1+
for this wave to reach the load end and then the reflection t3
will take place creating a reflected wave, which will reach 2T
the generator end at the time 2T. These processes will t2 ΓLV1+
continue until the steady-state. V1+
T
If one is interested in finding the voltage at a specific t1 z
location on the line, say z1, first find the time when each 0 z1 l
wave reaches that location, as denoted here by t1, t2 and vice
Fig. 8: A reflection diagram
versa in the figure. These are time instants when the voltage
discontinuities occur. Then the voltage at these instants can be obtained by simply adding all
components together.
Example 6.1 Consider a 400-m section of lossless transmission line having Z0 = 50 Ω and u =
200 m/µs connected to a load ZL = 100 Ω. At t = 0 a 30-V battery with zero generator
resistance is connected to the line. Sketch the distribution of voltage along the line for
several instants of time. Then sketch the voltage at the load to the line, v(l ;t), as a function of
time for 16 µs.

16
Transmission Lines

Example 6.2 Repeat the previous example where the voltage source is a pulse of 30 V but
duration 1 µs. The generator resistance remains zero.

17
Transmission Lines

Example 6.3 Consider a 400-m length of coaxial cable with C = 100 pF/m and L = 0.25
µH/m. The cable is terminated in a short circuit and is driven by a pulse source with internal
resistance of 150 Ω. The pulse has a magnitude of 100 V and duration of 6 µs. Sketch the
voltage at the input to the line, v(0;t), as a function of time for 18 µs.

Capacitive load termination

Assume that a lossless transmission line is terminated with a capacitive load CL, and a DC
voltage source Vg with internal resistance Rg is applied. When the switch is closed at t = 0, a
voltage wave of an amplitude

18
Transmission Lines

Z0
V1+ = v(0; t ) = Vg , 0 ≤ t ≤ 2T (6.20).
Z 0 + Rg
travels toward the load. Upon reaching the load at t = l /u = T, a reflected wave V1-(t) is
produced because of mismatch. At z = l , for all t ≥ T,

v(; t ) = vL (t ) = V1+ + V1− (t ) (6.21); iL (t ) =

1 +
Z0
[ ] dv (t )
V1 − V1− (t ) (6.22); iL (t ) = C L L (6.23).
dt
From (6.21), (6.22), one obtains
vL (t ) = 2V1+ − Z 0iL (t ) . (6.24)
Substituting (6.24) into (6.23) yields
dvL (t ) 1 2V1+
CL + vL (t ) = , t ≥ T. (6.25)
dt Z0 Z0
The solution of (6.25) is given by
{ }
vL (t ) = 2V1+ 1 − e − (t −T ) / Z0CL , t ≥ T, (6.26)
which can be easily obtained via Laplace’s transform.
It follows that

iL (t ) =
2V1+ −(t −T ) / Z0CL
Z0
e [ ]
, t ≥ T (6.27) ; V1− (t ) = 2V1+ 1 / 2 − e − (t −T ) / Z0CL , t ≥ T. (6.28)

7 The Smith Chart

The Smith chart is a graphical tool for calculating the characteristics of transmission lines; it
1
is constructed with a unit circle (radius 1), i.e., |Γ| ≤ 1. Here, assume that Z0 is real , and
Z − Z0
recall that Γ = ΓL = L (7.1) or Γ = ΓL = ΓL e jθ L = Γr + jΓi (7.2).
Z L + Z0
To generalize for use with different Z0, the chart is “normalized” by Z0. For the load
Z
impedance Z L , normalized impedance zL is given by z L = L = r + jx (7.3)
Z0
Using (7.3) in (7.1), (7.2) yields
z −1 (1 + Γr ) + jΓi
Γ = ΓL = Γr + jΓi = L (7.4) or z L = r + jx = . (7.5)
zL + 1 (1 − Γr ) − jΓi
(1 + Γr ) + jΓi [(1 + Γr ) + jΓi ][(1 − Γr ) + jΓi ]
Since = ,
(1 − Γr ) − jΓi (1 − Γr ) 2 + Γi2
1 − Γr2 − Γi2 2Γi
r= (7.6a) and x = (7.6b)
(1 − Γr ) 2 + Γi2 (1 − Γr ) 2 + Γi2
Rearranging terms in (7.6) yields
2 2 2 2
 r   1   1 1
 Γr −  + Γi =  (7.7a) and (Γr − 1) +  Γi −  =  
2

2
(7.7b)
 1+ r  1+ r   x  x

1
Recall that most practical transmission lines are assumed lossless or slightly lossy, where the characteristic
impedance is real.

19
Transmission Lines

These represent circles with (7.7a)

centered at (r/(1+r),0) with radius 1/(1+r)
and (7.7b) centered at (1,1/x) with radius
1/x. Note also that VSWR = (1+|Γ|)/(1-
|Γ|) is also a circle, thus if Γ is known, so
is VSWR.
Figure 9 shows some of constant r circles
and constant x circles given by (7.7).
Observations
1. Short circuit (r = x = 0) corresponds
to ZL =0 and open circuit (r = x = ∞)
corresponds to ZL = ∞ + j∞, but short
circuit is transformed into open
circuit with l = λ/4 (quarter-
wavelength transformer), i.e., Zin =
Z02/ZL, and vice versa. Fig. 9 : Constant r circles and constant x circles on
2. Complete revolution (2π) around the the Smith chart
Smith chart represents a distance of λ/2 on the line with clockwise denoting “toward the
generator (G)” and counterclockwise denoting “toward the load (L)”.
3. Three scales around the periphery of the Smith chart :λ toward G, λ toward L, λ in
degree.
4. Vmax at Zin,max when V+, V- in phase and Vmin at Zin,min when V+, V- out of phase, which are
λ/4 apart.
5. Can also be used as admittance chart with normalized admittance, y = g + jb.
Application of Smith Chart
Example 7.1: Input impedance calculation A load impedance of 130+j90 Ω terminates a 50-Ω
transmission line that is .3λ long. Find ΓL, Γ(z=0), Zin, SWR.
Solution The normalized load
impedance z L = (130+j90)/50 =
2.6+j1.8. Plot this value on the Smith
chart, one can find that ΓL =
0.60∠21.8° and SWR = 3.98. Since
the line is .3λ long, by moving the
d i s t a n c e . 3 λ × 4 π /
λ = 1.2π = 216° toward the generator
along the |Γ|=0.60 circle, one can find
the input impedance and reflection
coefficient at the generator to be:
Zin = Z0zin = 50(0.255+j0.117) =
12.7+j5.8 Ω and Γ(z=0) = Γin =
0.60∠165.8°.

Quiz 1 Repeat example 1 for a 100-Ω transmission line.

20
Transmission Lines

Example 7.2: Admittance calculation Let yL denote the normalized load admittance, then yL =
yL − 1 1 / z L − 1 1 − z L z −1
1/zL and = = . Recall that Γ = ΓL = L , it follows that
yL + 1 1 / z L + 1 z L + 1 zL + 1
yL − 1
= −Γ = Γe jπ . Thus, the normalized admittance can be found by rotating the normalized
yL + 1
impedance by 180°.
Use the previous example as an example. yL = 0.26-j0.18, thus YL = 5.2-j3.6 mS. Then by
moving 1.2π toward the generator, one can find that yin = 3.24-j1.48, thus Yin = 64.8-29.6 mS,
which is the reciprocal of 12.7+j5.8 Ω.
Quiz 2 Suppose the input impedance is found to be 25 + j20 Ω and the transmission line has
Z0 = 50 Ω and is 0.3625λ long. Find the load impedance.
Example 7.5: Quarter-wavelength transformer Recall that when l = λ/4, Z in = Z 02 / Z L , and
thus on the Smith chart, it becomes zin = 1/zL or yin = zL. Therefore, the input impedance of
the quarter-wavelength transformer can be found by rotating zL by π. For example, the short
circuit is transformed to the open circuit and vice versa.
The quarter-wavelength transformer has one significant application in matching load
impedance ZL to a transmission line with characteristic impedance Z0. By inserting a quarter-
wavelength transformer with characteristic impedance Z 0 = Z L Z 0 between the transmission
line and the load, the input impedance becomes Zin = Z0, thus the line is now matched.
The Slotted Line A slotted line is a transmission line configuration (usually waveguide or
coax) that allows the sampling of the electric field amplitude of a standing wave on a
terminated line. With the device, the VSWR and the distance of the first voltage minimum
from the load can be measured, and from this data the load impedance can be determined.
Note that the load impedance is generally a complex number, two distinct quantities must be
measured to determine the impedance.
Although the slotted line used to be the principal way to measure an unknown
impedance at microwave frequencies, it has been superseded by the modern vector network
analyzer in terms of accuracy, versatility and convenience. The slotted line is still of some
use in certain applications such as high-millimeter wave frequencies or where it is desired to
avoid connector mismatches by connecting the unknown load directly to the slotted line, thus
avoiding the use of imperfect transitions. Another reason for studying the slotted line is that it
provides an excellent tool for learning basic concepts of standing waves and mismatched
transmission lines.
Assume that, for a certain terminated line, the VSWR on the line and lmin, the distance
from the load to the first voltage minimum on the line, are measured. Recall that |Γ| =
(VSWR - 1)/(VSWR + 1) and a voltage minimum occurs when e j (θ −2 β ) = −1 , where θ is the
phase angle of the reflection coefficient, Γ = Γ e jθ . The phase angle of the reflection
coefficient is then θ = π + 2 β min . Actually, since the voltage minima repeat every λ/2, any
multiple of λ/2 can be added to lmin without changing the reflection coefficient, i.e.,

21
Transmission Lines

θ = π + 2 β ( min + nλ / 2) . Thus, VSWR and lmin can be used to determine the reflection
coefficient and the load impedance can be determined from the reflection coefficient.

Figure 10: A slotted line

Example 7.6 Impedance measurement with a slotted line The following two step procedure
has been carried out with a 50-Ω coaxial slotted line to determine an unknown impedance:
1. A short circuit is placed at the load plane,
resulting in a standing wave on the line with
infinite VSWR and sharply defined voltage
minima as shown in figure (a). Voltage minima
are recorded at z = 0.2, 2.2, 4.2 cm.
2. The short circuit is replaced by the unknown
load. VSWR=1.5 is measured and voltage
minima are recorded at z = 0.72, 2.72, 4.72 cm.
Find the unknown impedance.
Solution Since the voltage minima repeat every
λ/2, λ = 4.0 cm. In addition, because the reflection coefficient and the input impedance also
repeat every λ/2, we can consider the load terminals to be effectively located at any of the
voltage minima locations listed in step 1. Thus, if we say the load is at 4.2 cm, then the next
voltage minimum away from the load occurs at 2.72 cm, giving lmin = 4.2 – 2.72 = 1.48 cm =
0.37λ. Therefore,
VSWR − 1
Γ= = 0.2;
VSWR + 1
θ = π + 2β min = π + 2(2π )0.37 = 0.48π
→ Γ = 0.2e j 0.48π = 0.0126 + j 0.1996
The load impedance is then
1+ Γ
Z L = Z0 = 47.3 + j19.7 Ω .
1− Γ
To use the Smith chart, first noticing
that the voltage minimum is located on
the horizontal axis to the left of the
origin. Thus, beginning with this point
and then moving 0.37λ toward the load,
we can get zL = 0.95 + j0.39 and thus ZL
= 47.5 + j19.5 Ω.

22