Engineering Electromagnetics

W.H. Hayt Jr. and J. A. Buck

Chapter 10
Transmission Lines
Part 2: Reflection
 The Basic Reflection Problem
In the drawing below, we have an incident voltage wave, Vi , onto a load impedance
which in general is complex. The line could have losses, in which case the
characteristic impedance is complex as well. For simplicity, the load location is
chosen as z = 0, which means that all positions to the left of the load have negative z
values.

The incident voltage partially reflects from the load, forming a backward-propagting
voltage wave, Vr . A voltage across the load, VL , is thus formed. The load voltage
will have the same polarity as the incident voltage, but may be greater or less than it,
as we will see. The reflected voltage amplitude will always be equal or less than that
of the incident voltage.
 Voltage Waves
The incident and reflected waves are written in phasor form as:

+
(polarity
ZL VL assigned
_ arbitrarily)

Of interest to us are the wave voltages evaluated at the load,

(z = 0) where we have:

and
 Boundary Conditions at the Load

There are two boundary conditions: The first is that the load voltage
is equal to the sum of the incident and reflected voltages:

The second condition is that the load current is equal to the sum of the
incident and reflected currents:

Note again that all wave currents and voltages are complex amplitudes,
as we are evaluating everything at z = 0
 Voltage Reflection Coefficient
The boundary conditions are now applied, to obtain the ratio the reflected and
incident voltage amplitudes:

minus sign required for

backward wave current

Use these terms to solve for the reflection coefficient

Note that we will have no reflection if the load and characteristic impedances are
equal --- we thus have a matched load.
 Voltage Transmission Coefficient
Now, using the definition of the reflection coefficient, we may write:

… from which we get the voltage transmission coefficient, defined as

the ratio of the load and incident voltages:

Note that if Γ is greater than zero, τ will be greater than 1. This result,
although counterintuitive, is not a problem because the load current will
be less than that of the incident wave. As a consequence, the load power
will always be less than or equal to the power in the incident wave… as expected!
Power Reflection and Transmission
Next, consider the line on which the load is positioned at z = L, and with the
input at z = 0. A voltage wave is input, having amplitude V0. This wave
propagates down the line, experiencing loss, and achieving input power just in
front of the load, as shown. Reflected and transmitted powers are also indicated,
in which the transmitted power is that portion of the incident power that is dissipated
by the load impedance.

V0
ZL

z=0 z=L

The problem is to find the reflected and transmitted powers as fractions of the
power as evaluated just in front of the load.
 Reflected Power
V0
ZL

z=0 z=L
Using:

..and noting that we have a lossy line, the power just in front of the load becomes:

The reflected power is then:

 Reflected and Transmitted Power
V0
ZL

z=0 z=L

From the foregoing, we find the ratio of reflected to incident powers:

The transmitted (dissipated) power ratio is then just what is left over, or:
Example 1
Example 2
 Impedance Evaluation by Standing Wave
Measurements
Consider now a terminated line as shown below. A wave incident
onto the load is partially reflected, and a standing wave of some form is
developed, whose properties can be measured. Through such measurement,
the load impedance, if unknown, can be determined. Or, with a known load
impedance, the line characteristic impedance can be found.

In the line illustrated, the load is located at z = 0. The load impedance

is in general complex, while the characteristic impedance is assumed real (lossless line).
 Line Total Voltage

The total voltage on the line will be the sum of the incident and reflected
wave voltages, which we express in phasor form:

where the reflection coefficient is

The complex load impedance leads to a complex relection coefficient, having

non-zero phase, φ. The voltage amplitude, V0, is assumed real.
 Total Line Voltage (continued)
Incorporating the magnitude and phase of the reflection coefficient, the foregoing becomes:

We can express this in a more useful way through the trick of adding and subtracting
the term
The result is:
 Total Line Voltage (continued)
We now have:

The exponentials in parentheses become a cosine, so the final result in phasor form is:

In real instantaneous form, this becomes:

 Interpretation

The reflected wave, having amplitude reduced from that of the incident wave by
a factor of Γ, interferes with a portion of the forward wave of the same amplitude
to form a standing wave. The remainder of the incident wave remains a traveling
wave, moving in the forward z direction.

Note that the standing wave is phase-shifted spatially and temporally by half
the phase of the reflection coefficient. This provides us with a measurement tool.
 Minimum Voltage Location
In the line, and to the left of the load (negative z values), the standing wave pattern
will yield minimum and maximum voltage amplitudes at certain fixed locations.
To find the locations of the voltage minima, we return to the original expression for
the total phasor voltage:

At the location of a minimum voltage amplitude, the second term in the parentheses must be
real and negative. The phase term must therefore be an odd multiple of π:

Note that the negative sign is chosen for the phase, because all z values are negative
to the left of the load.
 Minimum Voltage Amplitude
Having the location of the minimum voltage amplitude, we can now find what the voltage is

Begin with:

where again:

we find on substitution:

so that:
 Maximum Voltage Amplitude
To find the locations of the voltage maxima, we return to the original expression for
the total phasor voltage:

At the location of a maximum voltage amplitude, the second term in the parentheses must be
real and positive. The phase term must therefore be a multiple of 2π:

Following the same reasoning as with the voltage minima, we find that the voltage
maxima locations are:

which, on substitution gives the voltage amplitude:

Voltage Amplitude Variation with Location
As voltage is measured as a function of location in the line, the following plot develops,
which shows the maximum voltages achieved at each position. Remember that voltage
is oscillating sinusoidally at each position between positive and negative values in time.
Voltage Standing Wave Ratio (VSWR)
VSWR, given the symbol s, is defined as the ratio of the maximum and minimum
voltage amplitudes:

Note that in taking this ratio, the voltage wave amplitudes divide out, and so
the quantity is of general significance.

Now, the magnitude of Γ can be found by measuring the VSWR, The phase
of Γ can be determined by measuring the location of the first voltage minimum
or maximum.
Standing Wave Measurements under Differing Load
Conditions

9.5

As the line is air-filled, what is the frequency?

Example
 Finite Length Line
More realistic problems involve lines that are not infinite or semi-infinite in
length, but rather of a finite length that must be accounted for.

Here we have a loaded section of line, having length l, and driven by a

sinusoidal steady state voltage source, with internal impedance, Zg .

Note again that all positions on the line are at negative values of z
 Distributed vs. Lumped Element Circuits
This circuit problem is complicated by the fact that we must account for the phase shift that
occurs as a function of position within the line; this arises because the line length is a significant
fraction of, or multiple wavelengths. The circuit therefore consists of distributed elements.
If the line length is much shorter than a wavelength, then we can make the usual lumped element
approximation.
The strategy in solving the finite line problem is to recast the configuration in the form of
an equivalent lumped element circuit, where the line-load combination is replaced by the
input or driving point impedance, Zin .

Of great importance (and convenience) is the simple fact that whatever power is dissipated by the
input impedance in the equivalent circuit is the same power that is dissipated by the load.
 Wave Analysis of the Finite Line

Within the transmission line are forward and backward (reflected) voltage
and current waves. The problem is greatly simplifed by the fact that steady
state conditions apply…. although multiple reflections have occurred at the source
and load ends, all of these superimpose to yield just one net forward wave,
and one net backward wave. The total voltage and current phasors are written as:
 Wave Impedance
The ratio of the net voltage and current give the wave impedance:

The importance of the wave impedance is that it gives the ratio of voltage to current
as a function of position anywhere on the line.

With these, the wave impedance simplifies to:

 Reformulated Wave Impedance

We now have:

…into which is substituted the Euler identity:

..and the reflection coefficient at the load:

The result is:

This form is easiest to work with, knowing the impedances.

 Input Impedance
Having the wave impedance,

to find the input, or driving point impedance, we simply evaluate the

wave impedance at z = -l :

This is the quantity that we need to create the equivalent circuit

 First Special Case: Half-Wave Section

Having now the input impedance:

Suppose the line length is a half wavelength, or an integer multiple of this.

In this event:

… which on substitution gives:

In other words, the load impedance appears at the input….. it is as if the line section
was not even there! This works provided the operating frequency is kept at the precise value
to yield the half-wavelength condition. As the frequency is varied, the condition would of course
no longer apply.
Second Special Case: Quarter-Wave Section

Again, starting with the input

impedance expression:

Suppose the line is a quarter-wavelength long, or some odd mulitple thereof.

In this event, we would have:

.. which on substitution gives:

In other words, a quarter-wave line produces an input impedance that is proportional

to the reciprocal of the load impedance.
 Application: Half-Wave Matching
In this case, we have a lossless half-wave line section that is inserted between two
identical lines. Waves (represented by power, Pin) incident from the left would
encounter an input impedance to the first junction that is equal to the characteristic
impedance of the line on the right side.

Z01 Z02 Z01

Zin = Z01
Pin Pout

The reflection coefficient at

the first junction will be: =0

So it is as if the center section is not there at all, regardless of its characteristic impedance.
This will be true as long as the frequency is fixed at the value that gives the half-wavelength section.
As frequency varies, the transmitted power will decrease as reflections at the first junction
increase. So we have a bandpass filter!
 Application: Quarter-Wave Matching
In this case, we have a lossless quarter-wave line section that is inserted between two
lines having different characteristic impedances. Waves (represented by power, Pin)
incident from the left would encounter an input impedance to the first junction

Z01 Z02 Z03

Pin Pout

To obtain zero reflection at the first junction (total power transmission), we require a
zero reflection coefficient there, which in turn determines the required value of the intermediate
characteristic impedance: