RF and Microwave

Active Circuit Design

COURSE NOTES

Matthew M. Radmanesh, Ph.D.

Professor of Electrical
& Computer Engineering,
California State University,
Northridge

This course notes is a work of non-fiction. Unless otherwise noted, the

author and the publisher make no explicit guarantees as to the accuracy of
the information contained in this book and in some cases, names of people
and places have been changed to protect their privacy.
2010 Matthew M. Radmanesh, Ph.D.
All Rights Reserved.
No part of this book may be reproduced, stored in a retrieval system, or
transmitted by any means without the written permission of the author.

With permission from the author, the information

presented in this course notes is an excerpt from the book:
RF & Microwave design Essentials, Matthew M.
Radmanesh, AuthorHouse Publishing house, 2007.

CHAPTER 5

Basics of RF and
Microwaves
5.1 INTRODUCTION
This chapter lays the foundation for understanding higher frequency
wave phenomena and compartments the task of active circuit design
RF/MW frequencies into specific concept blocks. The concept
blocks create a gradient approach to understanding and designing
RF/MW circuits and represent specific realms of knowledge that
need to be mastered in order to become an accomplished designer.
Before we proceed into analysis and description of these types of
waves we need to consider why RF/Microwaves as a subject have
become so important as to be placed at the forefront of our modern
technology; and furthermore, we need to expand our minds to the
many possibilities that these signals can provide for peaceful
practices by exploring various commercial applications useful to
mankind.

5.1.1 A Short History of RF & Microwaves

Circa 1864-1873, James Clark Maxwell integrated the entire man's
extant knowledge on electricity and magnetism and introduced a

series of four coherent and self-consistent equations which described

the behavior of electric and magnetic fields on a classical level. This
was the beginning of microwave engineering as presented in a
treatise by Maxwell at that time. He predicted, purely from a
mathematical and theoretical standpoint, the existence of
electromagnetic wave propagation and that light was also a form of
electromagnetic energy both completely new concepts at the time.
From 1885 to 1887, Oliver Heaviside simplified Maxwell's work in
his published papers. From 1887 to 1891, a German physics
professor by the name of Heinrich Hertz, verified Maxwell's
predictions experimentally and demonstrated the propagation of
electromagnetic waves. He also investigated wave propagation
phenomena along transmission lines and antennas and developed
several useful structures. He could be called the first microwave
engineer.
Marconi tried to commercialize Radio at a much lower frequency for
long-distance communications, but he had a business interest in all
of his work and developments. So this was not a purely scientific
endeavor.
The possibility of electromagnetic wave propagation inside a hollow
metal tube was never investigated by Hertz or Heaviside, since it
was felt that two conductors were necessary for the transfer of
electromagnetic waves or energy. In 1897, Lord Rayleigh
mathematically showed that electromagnetic wave propagation was
possible in a waveguide, both circular and rectangular. He showed
that there are infinite set of modes of the TE and TM type possible,
each with its own cut-off frequency. These were all theoretical
predictions with no experimental verifications.
From 1897 to 1936, waveguide was essentially forgotten until it was
rediscovered by two men, George Southworth (AT&T) and W. L.
Barron (MIT) who showed experimentally that waveguide could be
used as a small bandwidth transmission media, capable of carrying
high power signals.

With the invention of transistor in 1950s and the advent of

microwave integrated circuits in 1960s, the concept of a microwave
system on a chip became a reality. There has been many other
developments, mostly in terms of application mass, which has made
RF and microwaves an enormously useful and popular subject.
Maxwell equations lay the foundation and laws of the science of
electromagnetics, of which the field of RF and microwaves is a
small subset. Due to the exact and all-encompassing nature of these
laws in predicting electromagnetic phenomena along with the great
body of analytical and experimental investigations performed since
then, we can consider the field of RF and microwave engineering a
"mature discipline" at this time.

5.1.2 Applications of Maxwell's Equations

As indicated earlier in Chapter 2, standard circuit theory can not be
used at RF and particularly at microwave frequencies. This is
because the dimensions of the device or components are comparable
to the wavelength, which means that the phase of an electrical signal
(e.g. a current or voltage) changes significantly over the physical
length of the device or component. Thus use of Maxwell's equations
at these higher frequencies become imperative.
In contrast, the signal wavelengths at lower frequencies are so much
larger than the device or component dimensions that there is
negligible variation in phase across the dimensions of the circuit.
Thus Maxwell's equation simplify into basic circuit theory as
covered in Chapters 2 and 3.
At the other extreme of the frequency range lies the optical field,
where the wavelength is much smaller than the device or circuit
dimensions. In this case, Maxwell's equations simplify into a subject
commonly referred to as geometrical optics which treats light as a
ray traveling on a straight line. These optical techniques may be
applied successfully to the analysis of very high microwave
frequencies (e.g. high millimeter wave range), where they are
referred to as "quasi-optical". Of course, it should be noted that
further application of Maxwell's equations leads to an advanced field
of optics called "physical optics or Fourier optics", which treats light

as a wave and explains such phenomena as diffraction and

interference, where geometrical fails completely.
The important conclusion to be drawn from this discussion is that
Maxwell's equations present a unified theory of analysis for any
system at any frequency, provided one uses appropriate
simplifications when the wavelengths involved are a) much larger,
b) comparable to, or c) much smaller than the circuit dimensions.

5.1.3 Properties of RF and Microwaves

An important property of signals at RF, and particularly at higher
microwave frequencies is their great capacity in carrying
information. This is due to the existence of large bandwidths that is
available at these high frequencies. For example a 10% bandwidth at
60 MHz carrier signal is 6 MHz which is approximately one TV
channel of information; on the other hand 10% of a microwave
carrier signal at 60 GHz is 6 GHz which is equivalent to 1000 TV
channels.
Another property of microwaves is that they travel by line of sight,
very much like traveling of light rays as described in the field of
geometrical optics. Furthermore, unlike the lower frequency signals,
the microwave signals are not bent by the ionosphere. Thus use of
line-of-sight communication towers or links on the ground and
orbiting satellites around the Globe are a necessity for local or global
communications.
A very important civilian as well as military instrument is Radar.
The concept of Radar is based upon Radar cross-section which is the
effective reflection area of the target. Target's visibility greatly
depends on the target's electrical size which is a function of the
incident signal's wavelength. Microwave frequencies form the ideal
signal band for Radar applications. Of course, another important
advantage of use of microwaves in Radars is the availability of
higher antenna gains as the frequency is increased for a given
physical antenna size. This is because the antenna gain is
proportional to the electrical size of the antenna, which becomes
larger as frequency is increased in the microwave band. The key

factor in all this is that microwave wavelengths are comparable to

the physical size of the transmitting antenna as well as the target.
There is a fourth and yet a very important property of microwaves
and that is the molecular, atomic and nuclear resonance of
conductive materials and substances when exposed to microwave
fields. This property creates a wide variety of applications. For
example, since almost all biological units are composed of water
predominantly, and as we know water is a good conductor, thus
microwave gains tremendous importance in the field of detection,
diagnostics and treatment of biological problems or investigations as
in medicine (e.g. diathermy, scanning, etc.). There are other areas
that this basic property would create a variety of applications such as
remote sensing, heating (e.g. industrial purification, cooking, etc.)
and many others which are listed in a later section.

5.2 REASONS FOR USING RF/MICROWAVES

Over the past several decades, there has been a growing trend toward
use of RF/Microwaves in system applications. The reasons are
many, amongst which the following are prominent:
a. Wider bandwidths due to higher frequency
b. Smaller component size leading to smaller systems
c. More available and uncrowded frequency spectrum
d. Better resolution for Radars due to smaller wavelengths
e. Lower interference due to a lower signal crowdedness
f. Higher speed of operation
g. Higher antenna gain possible in a smaller space
On the other hand, there are some disadvantages in using
RF/Microwaves such as: use of more expensive components;
availability of lower power levels; existence of higher signal losses
and use of high-speed semiconductors (such as GaAs or InP) along
with their corresponding less-mature technology, relative to the
traditional Silicon technology which is quite mature and less
expensive at this time.

In many RF/Microwave applications the need for a system operating

at these frequencies with all the above advantages, is so great that it
outweighs these disadvantages aside and spurs the engineer forward
into a high-frequency design.

5.3 RF/MICROWAVE APPLICATIONS

The major applications of RF/Microwave signals can be categorized
as follows:
A. Communication
This application includes satellite, space, long-distance telephone,
marine, cellular telephone, data, mobile phone, aircraft, vehicle,
personal and Wireless Local Area Network (WLAN) and so on.
There are two important sub-categories of applications that needs to
be considered as follows:
A1. TV and Radio broadcast
In this application, RF/Microwaves are used as the carrier
signal for the audio and video signals. An example is the Direct
Broadcast Systems (DBS) which is designed to link satellites
directly to home users.
A2. Optical Communications
In this application a microwave modulator is used in the
transmitting side of a low-loss optical fiber with a microwave
demodulator at the other end. The microwave signal acts as a
modulating signal with the optical signal as the carrier. Optical
communications is useful cases where a much larger number of
frequency channels as well as lower interference from outside
electromagnetic radiation are desired. Current applications
include telephone cables, computer network links, low-noise
transmission lines, etc.
B. Radar
This application includes air defense, aircraft/ship guidance, smart
weapons, police, weather, collision avoidance, imaging, etc.
C. Navigation
This application is used for orientation and guidance of aircraft,
ships and land vehicles. Particular applications in this area are:

C1. Microwave Landing System (MLS), which is used to guide

aircraft to land properly in airports.
C2. Global Positioning Systems (GPS) which is used to find one's
exact coordinates on the Globe.
D. Remote Sensing
In this application many satellites are used to monitor the Globe
constantly for weather conditions, meteorology, ozone, soil
moisture, agriculture, crop protection from frost, forests, snow
thickness, icebergs and other factors such as natural resources
monitoring and exploration, etc.
E. Domestic and industrial applications
This application includes microwave Ovens, microwave clothes
dryer, fluid heating, moisture sensors, tank gauges, automatic door
openers, automatic toll collection, Highway traffic monitoring and
control, chip defect detection, flow meters, power transmission in
space, food preservation, pest control, etc.

F. Medical applications
This application includes cautery, selective heating, Heart
stimulation, Hemorrhaging control, sterilization, imaging, etc.
G. Surveillance
This application includes security systems, intruder detection,
Electronic warfare (EW) receivers to monitor signal traffic, etc.
H. Astronomy and space exploration
In this application, gigantic dish antennas are used to monitor,
collect and record incoming microwave signals from outer space,
providing vital information about other planets, star, meteors, etc., in
this or other galaxies.
I. Wireless applications
Short-distance communication inside as well as between buildings in
a local area network (LAN) arrangement can be accomplished using
RF and Microwaves.

10

Connecting buildings via cables (e.g. coax or fiber optic) creates

serious problems in congested metropolitan areas, since the cable
has to be run underground from upper floors of one building to
upper floors of the other. However, this problem can be greatly
alleviated using RF and microwave transmitter/receiver systems
which are mounted on rooftops or in office windows (see Figure
5.1).
Inside buildings, RF and Microwaves can be used effectively to
create a wireless LAN in order to connect telephones, computers and
various LANs to each other. Using wireless LANs has a major
advantage in office re-arrangement where phones, computers and
partitions are easily moved with no change in wiring in the wall
outlets. This creates enormous flexibility and cost saving features for
any business entity.
A summary of RF and microwave applications is shown in table 5.1.

11

12

13

5.4 RADIO FREQUENCY (RF) WAVES

Having briefly reviewed many of the current applications of
RF/Microwaves, we can see that this rapidly advancing field has
great potential to be a fruitful source of many future applications.
As discussed earlier, electromagnetic (EM) waves are generated
when electrical signals pass through a conductor. EM waves start to
radiate more effectively from a conductor when the signal frequency
is higher than the highest audio frequency which is approximately 15
to 20 kHz. Because of this radiating property, signals of such or
higher frequencies are often known as radio frequency (RF) signals.

5.4.1 RF Bands
Since it is not practical either a) to design a circuit that covers the
entire frequency range, or b) to use all radio frequencies for all

14

purposes, therefore the RF spectrum is broken down into various

bands. Each band is used for a specific purpose and in general, RF
circuits are designed to be used in one particular band. Table 5.2
shows the most common assignment of RF commercial bands.

5.4.2 Definition of Microwaves

When the frequency of operation starts to increase toward
approximately 1GHz and above, a whole set of new phenomena
occurs that is not present at lower frequencies. The radio waves at
frequencies ranging from 1 GHz to 300 GHz are generally known
as microwaves. Signals at these frequencies have wavelengths that
range from 30 cm (at 1 GHz) to 1 millimeter ( at 300 GHz). The
special frequency range from 30 GHz to 300 GHz has a wavelength
in the millimeter range thus is generally referred to as millimeterwaves.
NOTE: It should be noted that in some texts, the range 300 MHz to
300 GHz is considered to be the microwave frequency range. This is
in contrast with the microwave frequency range defined above,
where the frequency range from 300 MHz to 1 GHz is referred as
the RF range.

5.4.3 Microwave Bands

The microwave frequency range consisting of the three main
commercial frequency bands (UHF, SHF and EHF) can further be
subdivided into several specific frequency ranges each with its own
band designation. This band subdivision and designation facilitates

15

the use of microwave signals for specific purposes and applications.

In electronics industries and academic institutions, the most
commonly used microwave bands are as set forth by the Institute of
Electrical and Electronics Engineers (IEEE) and is shown in table
5.3. In this table the "Ka to G" are the millimeter-wave (mmw)
bands.

5.5 RF AND MICROWAVE (MW) CIRCUIT DESIGN

Because of the behavior of waves at different frequencies, basic
considerations in circuit design has evolved greatly over the last few
decades and generally can be subdivided into two main categories:
a. RF circuit design considerations, and
b. Microwave (MW) circuit design considerations
Each category is briefly described next.

5.5.1 Low RF Circuit Design Considerations

RF circuits have to go through a three step design process. In this
design process the effect of wave propagation on the circuit
operation is negligible and the following facts can be stated:

16

a. The length of the circuit (l) is generally much smaller than the
wavelength (i.e. l<< )
b. Propagation delay time (td) is approximately zero ( i.e. td 0).
c. Maxwell's Equations simplify into all of the low frequency laws
such as KVL, KCL, Ohm's law, etc. Therefore at RF frequencies
(f<1 GHz), the delay time of propagation (td) is zero when l<<
and all elements in the circuit can be considered to be lumped.
The design process has the following four steps:
Step 1. The design process starts with the selection and of a suitable
device and performing a DC design to obtain a proper Q-point.
Step 2. Next, the device will be characterized (either through
measurement or calculations) to obtain its AC small signal
parameters based on the specific DC operating point selected
earlier.
Step 3. The third step consists of designing two matching circuits
that transition this device to the outside world which are the signal
source at one end and the load at the other. Various design
considerations and criteria such as stability, gain, noise, etc. are
included at this stage and must be incorporated in the design of the
final matching networks.
Step 4. In this final step, the entire circuit is put together in one
seamless design to create a functional circuit. This circuit is now
packaged properly by enclosing it in an appropriate box with correct
connectors or terminals for communication to the outside world.
The design process for RF circuits is summarized and shown in
Figure 5.2.

5.5.2 High RF and Microwave Circuits

To understand microwave circuits we should know that microwave
circuits may have one or more lumped elements but should at least
contain one distributed element. This last needs to be defined at this
point:
DEFINITION- DISTRIBUTED ELEMENT: Is defined to be an element
whose property is spread out over an electrically significant length

17

or area of a circuit instead of being concentrated at one location or

within a specific component.

______________________________________________________
EXAMPLE 5.1
Describe what a distributed inductor is?
Answer:
A distributed inductor would be an element whose inductance is
spread out along the entire length of a conductor (such as selfinductance) as distinguished from an inductor whose inductance is
concentrated within a coil.
_______________________________________________________
EXAMPLE 5.2
Describe what a distributed capacitor is?
Answer:
A distributed capacitor is an element whose capacitance is spread
out over a length of wire and not concentrated within a capacitor,

18

such as the capacitance between the turns of a coil or between

adjacent conductors of a circuit.
_______________________________________________________
Working with distributed circuits, we need to know the following
facts about them:
a. The wave propagation concepts as set forth by the Maxwell
Equations fully apply and,
b. The circuit has a significant electrical length, i.e. its physical
length is comparable to the wavelength of the signals
propagating in the circuit.
This fact brings the next point into view:
c.

The time delay (td) due to signal propagation can no longer

be neglected (i.e. td0).

To illustrate these points we will consider the following example.

_______________________________________________________
EXAMPLE 5.3
How does a two-conductor transmission line (Such as a coaxial
line, etc.) behave at low and high frequencies?
Answer:
At low frequencies this transmission line is considered to be a short
piece of wire with a negligibly small distributed resistance which
can be considered to be lumped for the purpose of analysis (since
td0).
However at higher frequencies, the resistive, capacitive and
inductive properties can no longer be separated and each
infinitesimal length (x) of this transmission line exhibits these
properties as shown in Figure 5.3.
From this Figure, we can see that the elements are series elements
(R, L) and shunt elements (G ,C) which are defined as:
R= resistance per unit length in /m
L = inductance per unit length in H/m
G = conductance per unit length in S/m
C = capacitance per unit length in F/m

19

This equivalent circuit is referred to as a distributed circuit model of

a two-conductor transmission line and will be used in the next
example to derive the equivalent circuit model of a transmission
line.
_______________________________________________________
EXAMPLE 5.4
Using KVL and KCL derive the relationship between voltage and
current in a transmission line at:
a. Low frequencies
b. High frequencies ( i.e., RF/Microwave frequencies)
Solution:
a. At low frequencies a transmission line (which can be lossy in
general), can be represented as shown in Figure 5.4.

In this Figure, "R" is the distributed loss resistance of the line, which
can be modeled as a lumped element. The voltage and current
relationship can be written as:
V1=V2+IR
Note: If the line is lossless, then we have: V1=V2

20

b. At high frequencies, based on Figure 5.3 a transmission line can

be modeled as a distributed element as shown by the equivalent
circuit in Figure 5.5.

The analysis of this equivalent circuit will be postponed until

Chapter 7- Wave Propagation, where we will examine one x
section of a transmission line and will develop the governing
equations of a transmission line in great depth.
______________________________________________________

5.5.3 High RF and Microwave Circuit Design Process

The microwave circuit design process is very similar to the RF
circuit design except for the wave propagation concepts which
should be taken into account.
The design process has the following four steps:
Step 1: The design process starts with the design of the DC circuit to
establish a stable operating point.
Step 2: The next step is to characterize the device at the operating
point (Q-point), using electrical waves to measure the percentage of
reflection and transmission that the device presents at each port.
Step 3: The third step consists of designing the matching networks
that transition the device to the outside world such that the required
specifications such as stability, overall gain, etc. are satisfied.
Step 4. In this final step, the entire circuit is put together in one
seamless design to create a functional circuit. This circuit is now
packaged properly by enclosing it in an appropriate box with correct
connectors or terminals for communication to the outside world.

21

Except for the fact that one's familiarity with wave propagation
concepts becomes crucial, Microwave Circuit Design process is
similar to the RF circuit design steps as delineated in Figure 5.6.

Figure 5.6 Microwave Circuit Design Steps

5.6 FUNDAMENTALS VERSUS STRUCTURE

Before we get into specific analysis and design of RF and
microwave circuits, it is worthwhile first to examine a general
communication system in which each circuit or component has a
specific function in a bigger scheme of affairs. In general, any
communication system is based upon a very simple and yet
extremely fundamental truth, commonly referred to as the "universal
communication principle".
The "Universal Communication Principle" is a fundamental concept
which is at the heart of a wide sphere of existence called "life and
livingness", or for that matter any of its subsets particularly the field
of RF/Microwaves. This principle is intertwined throughout the
entire field of RF/microwaves and thus plays an important role in

22

our understanding of this subject. Therefore it behooves us well to

define it at this juncture.
THE UNIVERSAL COMMUNICATION PRINCIPLE: This principle states
that communication is the process whereby information is
transferred from one point in space and time (X1, Y1, Z1, t1), called
the source point, to another point in space and time (X2, Y2, Z2, t2),
called the receipt point. Usually, the receipt point at location (X2, Y2,
Z2) is separated by a distance (d) from the source point location (X1,
Y1, Z1).
The physical embodiment of the universal communication principle
is a "communication system", which takes the information from the
source point and delivers an exact replica of it to the receipt point
(see Figure 5.7).

Thus in general, it can be seen that any communication system can

be broken down into three essential elements:
1. Source point: A point of emanation or generation of information.
2. Receipt point: A point of receipt of information.
3. Distance (or Imposed space): The space existing between the
"Source point" and "Receipt point" where the information travels
through.
Furthermore, it can be observed that in order to achieve effective
communication between two systems, we need to have three more
factors present: a) There must be intention on the part of the source

23

point and the receipt point to emit and to receive the information,
respectively, b) source and receipt points must have attention on
each other (i.e. both being ready for transmission and reception), and
c) duplication (i.e. an exact replica) must occur at the receipt point of
what emanated from the source point.
Use of the universal communication principle in practice creates a
one-way communication system (such as radio and TV broadcast,
etc.), and forms one leg of a two-way communication system (such
as CB radio, telephone, etc.), where this process is reversed to create
the second leg of the communication action. An important
application of the universal communication principle is in a radar
communication system where the source point (X1, Y1, Z1) is at the
same physical location as the receipt point (X2, Y2, Z2), i.e., X1=X2,
Y1=Y2, Z1= Z2; however the times of sending and reception are
different (t1 t2). Otherwise no communication would take place.
This brings us to the obvious conclusion that one can not have a
condition where the source and the receipt points are the same,
simultaneously!
Based on this simple concept of communication, the most complex
communication systems can be understood, analyzed and designed.
Figure 5.8 is a simple and yet a very generalized block diagram of
such a practical communication system in use today.
It should be noted that the design and structure of this
communication system can change and evolve into a more efficient
system with time whereas the universal communication principle
will never change.
Of course this should be no surprise to the workers in the field
because as it turns out the foundation (which consists of fundamental
postulates, axioms and natural laws) along with fundamental
concepts (i.e., theorems, analytical techniques, theory of operation,
etc.) of any science is far superior in importance to any designed
circuitry, machinery, network, etc. This observation makes us realize
that the fundamentals are unchanging whereas the structure exists
on a constant-change basis and is always evolving. This brings us to
the following conclusion:

24

Fundamentals of any science are superior and dictate the

designed forms, structures or in general the entire application
mass of that science, and not vice versa.
This is true in all aspects of design, i.e., while the underlying
principle remains constant, the structure which is the electronic
circuit, constantly undergoes improvements with new designs and
evolves in time toward a more efficient circuitry.
This can best be described as "engineering principle as a constant"
vs. "the application mass as a constantly evolving structure" where it
approaches closer and closer to the underlying principle with each
improvement.

25

Even though rarely new discoveries may bring about new underlying
fundamentals to the forefront, nevertheless the fundamentals, as a
general rule of thumb, remain invariant.
For example, circa 1864-1873 James clerk Maxwell interrelated all
of the known data about electricity and magnetism, formulated and
presented the classical laws of Electromagnetics. Since that time,
which is over a century, tremendous technological changes and
advances have happened all over the Globe and yet Maxwell's
equations have not changed an iota. This set of celebrated
equations have remained timeless!
Of course it should be noted that quantum mechanics, dealing with
sub-atomic particles may be considered by some, to have
generalized these equations and shown that energy is not continuous
but quantized. Nevertheless, the Maxwell's equations at the classical
level of observation have not been surpassed and are still true today
and currently form the foundation of the "Electromagnetics" as a
science -- the backbone of electronics and electrical engineering.
Now to build a communication system in the physical universe that
works and is practical, one must satisfy two conditions:
1. First, it must be based on the fundamental concept of a) " the
universal communication principle" and then b) "Maxwell's
Equations"-- both in combination form a static which is unchanging!
2. Secondly, it must follow and conform to the current state of
technology in terms of manufacturing, materials, device fabrication,
circuit size and structure -- a kinetic and constantly evolving!
These two pre-requisites, in essence, clearly demonstrate and
confirm the interplay of "static vs. kinetic" which is interwoven
throughout our entire world of science and technology.
The above two steps of system design sets up the "Blue Print" for
any "general engineering system design". One must heed these
points carefully before one has gone very far in the quest for
workable knowledge.

26

5.7 ACTIVE CIRCUIT BLOCK DIAGRAMS

Considering Figure 5.8, we note several stages:
1. Energy Conversion Stage: This is a simple transducer casing
the incoming energy (e.g. sound, etc.) to be converted to
electrical energy. An example for this stage could be a
microphone.
2. Amplification Stage: This is a high gain small-signal amplifier
causing a higher signal to compensate for losses in the energy
conversion stage.
3. Frequency Conversion Stage (also called Modulation or UpConversion): This causes a carrier wave to be modulated by the
amplified signal of stage 2. This is the stage that prepares the
signal for transmission for long distance by increasing its
frequency, since higher frequency signals travel longer and
require smaller antennas. Needless to say that a local oscillator is
needed to produce the carrier wave before the modulation
process can take place.
4. Power Amplification Stage: This is the stage where the signal
power level is boosted greatly so that a higher range of reception
is allowed.
5. Transmission Link: This is the transmission media in which the
modulated signal is transported from "cause or source point" to
the "effect or the receipt point".
6. Low-Noise Amplification Stage: This is the first stage (or
Front-end) of the receiver wherein the modulated signal is
amplified and prepared by a low-noise amplifier (LNA) in such a
way that the effect of noise which could possibly be added to the
signal by later stages, is minimized.
7. Frequency Conversion Stage (Demodulation Or DownConversion): This stage demodulates the signal and brings the
carrier frequency down to workable levels. Just like the

27

modulation stage, a local oscillator of a certain frequency is

needed to make the demodulation process effective. Note: If the
local oscillator is tunable, then the same receiver can be used to
receive signals from other sources at other frequencies (a
Heterodyne receiver!).
8. Detector Stage: This stage removes the carrier wave and
reconstructs the original signal.
9. Energy Conversion Stage: This stage converts the electrical
signal back to its original form (e.g. sound). An example for this
stage could be a speaker.
10. Control Stage: This is where all the decisions with regard to
circuit connection/disconnection, routing, switching, etc. takes
place and is present at both the source and the receipt points of
the communication system.
To gain a full conceptual understanding of different types of circuit
designs one needs to have an overall idea of "how different
components fit together". To bring this point into a realm of
practicality each specific type of microwave circuit has been cast
into an exact block diagram that clearly depicts the relationship of
the device with other circuit components and sections.
The circuits considered for the purpose of the block diagram are as
follows:
A.

Amplifier: is defined to be an electronic circuit capable of

increasing the magnitude or power level of an electrical signal
without distorting the wave-shape of the quantity. The block
diagram for this circuit is shown in Figure 5.9.

28

FIGURE 5.9 An amplifier circuit block diagram.

B.

Oscillator: is defined to be an electronic circuit that converts

energy from a DC source to a periodically varying electrical
signal. The block diagram for this circuit is shown in Figure
5.10.

FIGURE 5.10 An oscillator circuit block diagram.

C.

Mixer: is defined to be an electronic circuit that generates an

output frequency equal to the sum and difference of two input
frequencies or in short a frequency converter. The block
diagram for this circuit is shown in Figure 5.11.

29

FIGURE 5.11 A mixer circuit block diagram.

D. Detector: is defined to be an electronic circuit concerned

with demodulation, i.e., extracting a signal which has
modulated a carrier wave. The block diagram for this circuit
is shown in Figure 5.12.

FIGURE 5.12 A detector circuit block diagram.

From these block diagrams we can see that the device forms the
"heart" or "engine" of the circuit around which all other circuit
components should be properly designed in order to control the
input/output flow of signals and eventually obtain an optimum
performance.
Furthermore, these four block diagrams show the irresistible fact that
the knowledge gained in earlier chapters is essential in the design of
these complicated circuits.

30

5.8 SUMMARY
To be proficient at higher frequency circuits (analysis or design),
one needs to master, on a gradient scale, all of the underlying
principles and develop a depth of knowledge before one can be
called a skilled microwave practitioner.
Figure 5.13 depicts the gradient scale of concepts that need to be
fully understood in order to achieve a mastery of circuit design skills
at higher frequencies.

31

As shown in Figure 5.13, one starts with the fundamental axioms of

sciences, fundamental concepts in electronics, and progresses toward
high frequency electronic circuit design by learning the DC and low
frequency concepts at first, then wave propagation concepts, devicecircuit characterization, matching concepts and eventually arrives at
the final destination of RF/MW active circuit design concepts, which
was originally set forth as the goal of this book.
Knowing this progressive series of concepts will enable one to
design amplifiers, oscillators, mixers, detectors, control circuits and
integrated circuits with relative ease and proficiency at RF/MW
frequencies.

CHAPTER 5- SYMBOL LIST

A symbol will not be repeated again, once it has been identified and
defined in an earlier chapter, with its definition remaining
unchanged.
l Length of the circuit
td Time delay
- Wavelength

CHAPTER -5 PROBLEMS
5.1)
5.2)
5.3)
5.4)

5.5)

What is the difference between a lumped element and a

distributed element?
How many steps are required to design a) an RF circuit? b) A
microwave circuit? Describe the steps.
What are the similarities and difference(s) between an RF
and a microwave circuit design procedure?
Describe: a) What is meant by "fundamentals vs. application
mass"? b) What is meant by timelessness of a fundamental
truth? Give an example, c) What part of a system constantly
evolves? and d) What are the pre-requisites for any general
system design?
What is at the heart of an amplifier, an oscillator, a mixer or
a detector block diagrams?

32

5.6)
5.7)

What are the main concepts one needs to master in order to

design an RF or a microwave circuit?
Why is it necessary to understand the low frequency
electronics fully before trying to master RF/microwave
electronics?

REFERENCES
[5.1] Carlson, A. B. Communication Systems: An Introduction to
Signals and Noise in Electrical Communication. New York:
McGraw-Hill, 1968.
[5.2] Cheung, W. S. and F. H. Levien. Microwave Made Simple.
Dedham: Artech House, 1985.
[5.3] Gardiol, F. E. Introduction to Microwaves. Dedham: Artech
House, 1984.
[5.4] Ishii, T. K. Microwave Engineering. 2nd ed., Orlando: Harcourt
Brace Jovanovich, publishers, 1989.
[5.5] Laverghetta, T. Practical Microwaves. Indianapolis: Howard
Sams, 1984.
[5.6] Lance, A. L. Introduction to Microwave Theory and
Measurements. New York: McGraw-Hill, 1964.
[5.7] Radmanesh, M. M. Applications and Advantages of Fiber
Optics as Compared with other Communication Systems, Hughes
Aircraft Co., Microwave Products Div. , pp. 1-11, April 1988.
[5.8] Radmanesh, M. M. Radiated and Conducted Susceptibility
Induced Current in Bundles: Theory and Experiment, Boeing Co., HE
RF D iv., pp. 1115, Sept. 1990.
[5.9] Radmanesh, M. M. The Gateway to Understanding: Electrons to
Waves and Beyond, AuthorHouse, 2005.
[5.10] Radmanesh, M. M. Cracking the Code of Our Physical
Universe, AuthorHouse, 2006.
[5.11] Saad, T. Microwave Engineers Handbook, Vols I, II.
Dedham: Artech House, 1988.
[5.12] Scott, A. W. Understanding Microwaves. New York: John
Wiley & Sons, 1993.

33

34

CHAPTER 6

RF Electronics
6.1 INTRODUCTION
It is important to set the stage properly for the introduction of
microwave circuits. To that end we will introduce RF circuit
analysis and design to serve as a platform of fundamental
information in order to catapult us into the world of microwave
circuit design. Therefore this chapter will primarily deal with the
world of RF circuit design with the intention of preparing the reader
for a much broader field of study, namely, microwave circuit
analysis and design presented in the future chapters.

6.2 RF/MICROWAVES VERSUS DC OR LOW AC

SIGNALS
There are several major differences between signals at higher radio
frequency (RF) or microwaves (MW) and their counterpart at low
AC frequency or DC. These differences which influence the
electronic circuits and their operation greatly, become increasingly
important as the frequency is raised. The following four effects
provide a brief summary of the effects of RF/MW signals in a circuit
which are not present at DC or low AC signals:

35

EFFECT #1. PRESENCE OF STRAY CAPACITANCE - This is the

capacitance that exists:
a. Between conductors of the circuit
b. Between conductors or components and ground
c. Between components
This effect is shown in Figure 6.1.

EFFECT #2. PRESENCE OF

inductance that exists due to:

STRAY INDUCTANCE-

This is the

a. The inductance of the conductors that connect components, and

b. The parasitic inductance of the components themselves.
These stray parameters are not usually important at DC and low AC
frequencies but as frequency increases, they become a much larger
portion of the total. This concept is
shown in Figure 6.2.

EFFECT #3. SKIN EFFECT- This refers to the fact that AC signals
penetrate a metal partially and flow in a narrow band near the
outside surface of each conductor. This is in contrast to the DC
signals where they flow through the whole cross section of the
conductor as shown in Figures 6.3a and 6.3b.

36

For AC signals, the current density falls off exponentially from the
surface of the conductor toward the center. At a critical depth (),
called the skin depth or depth of penetration, signal amplitude is 1/e
or 36.8% of its surface amplitude (see Figure 6.3c) which is given
by:
1
=
f
where is the permeability (H/m), and is the conductivity of the
conductor.
_______________________________________________________
EXAMPLE 6.1
Considering copper as the conductive medium, what is the skin
depth at 60 Hz and 1 MHz?

Solution:
For copper we have:

37

=4 x 10-7 H/m
=5.8 x 107 S/m
At f=60 Hz =(1/ x 60 x 4 x 10-7 x 5.8 x 107)1/2=0.85 cm
While on the other hand for f=1 MHz, we calculate to be :
= 0.007 cm
which is a substantial reduction in penetration depth.
_______________________________________________________
As seen from example 6.1, we can observe that as frequency
increases, skin effect produces a smaller zone of conduction and a
correspondingly higher value of AC resistance compared with DC
resistance.
EFFECT #4. RADIATION- This is caused by the leakage or escape of
signals into the air. This, in essence, means that the signals bypass
the conducting medium and not all of the source energy is reaching
the load. Radiation can occur outside or within a circuit as shown in
Figure 6.4

38

The radiation factor causes coupling effects to occur as follows:

a. Coupling between elements of the circuit,
b. Coupling between the circuit and its environments, and
c. Coupling from the environment to the circuit.
"Electromagnetic interference" (EMI), also called "Radio Frequency
Interference" (RFI) or RF-noise, is due to signals at RF/MW
frequencies, is missing in DC circuits and is considered to be
negligible in most low-frequency AC circuits.

6.3 EM SPECTRUM
When an RF/MW signal radiates, it becomes an EM wave that is
propagating through a medium such as air. The range of frequencies
of electromagnetic waves known as the EM spectrum is shown in
Figure 6. 5.

Looking at this spectrum one may wonder, "How do microwaves

differ from other EM waves? The answer lies in the fact that
microwaves is a separate topic all by itself because at these
frequencies the wavelength () approximates the physical size of the
ordinary electronic components as discussed earlier in Chapter 5.
Therefore components behave differently at microwave frequencies
than they do at lower frequencies.

39

_______________________________________________________
EXAMPLE 6.2
How does an ordinary resistor element behave at microwave
frequencies?
Solution:
An ordinary carbon resistor at microwave frequencies (e.g. at f =10
GHz) has a stray capacitor and a stray inductor as well as a higher
resistance due to the skin effect ( since the cross section is reduced )
and radiation (since part of the power is lost in the air).
These factors are added into the equivalent circuit (as shown in
Figure 6.6).

the addition of extra parasitic elements in the equivalent circuit is

due to the combined length of the leads and the physical size of the
component itself which is comparable to the wavelength.
_______________________________________________________

6.4 Wavelength and Frequency

When an electromagnetic wave with a certain oscillation frequency
(f) propagates through the air or any other medium, it does so at a
certain fixed speed or velocity (also known as the phase velocity
(Vp)and a corresponding fixed wavelength () as shown in Figure
6.7.
These three factors: f, Vp and are not independent from each other
and in fact are interrelated such that the product of frequency (f) and
wavelength () is equal to the velocity (Vp), i.e.,
f = Vp
(6.1)

40

It has been shown that the velocity of propagation for any and all
EM waves through the air is approximately equal to the speed of
light(c):
c= 3 x 108 m/s
However, if the medium is not air the speed is lower than "c" and
can be shown to be :
Vp = c/ r
(6.2)
Where r is the relative dielectric constant of the medium of
propagation.

6.5 COMPONENT BASICS

In this section, the properties of resistors, capacitors and inductors at
high radio frequencies will be studied. . But first, we will take a
brief look at the most simple component of all: a piece of wire. We
will consider this element first and examine its problems at radio
frequencies.

6.5.1 Wire
A wire is the simplest element to study having a zero resistance
which makes it to appear as a short circuit at DC and low AC
frequencies. Yet at RF/MW frequencies it becomes a very complex
element and deserves special attention that will be studied in depth
shortly. Wire in a circuit can take on many forms, such as:
Wire wound resistors,
Wire wound inductors,
Leaded capacitors (see Figure 6.8), and
Element-to-element interconnect applications

41

The behavior of a wire in the RF spectrum depends to a large extent

on the wire's diameter and length. A system for different wire sizes
is the American Wire Gauge (AWG) System. In this system, the
diameters of a wire will roughly double every six gauges.
______________________________________________________
EXAMPLE 6.3
Given that the diameter of AWG 50 is 1.0 mil (0.001 inch), what is
the diameter of AWG 14?
Solution:
Starting from AWG 50 we descend downward by 6 gauges until we
reach AWG 14 as follows:
AWG 50 d = 1 mil,
AWG 44 d = 2 mils,
AWG 38 d = 4 mils
AWG 32 d = 8 mils,
AWG 26 d= 16 mils,
AWG 20 d= 32 mils,
AWG 14 d = 64 mils.
______________________________________________________
Problems Associated with a Piece Of Wire
Problems associated with a wire can be traced to two major areas:
a. Skin effect, and
b. Straight-wire inductance.
These two problems are discussed next.
a. Skin Effect in a Wire
As frequency increases, the electrical signals propagate less and less

42

in the inside of the conductor. The current density increases near the
outside perimeter of the wire and causes a higher impedance to be
seen by the signal as shown in Figure 6. 9. This is because resistance
of the wire is given by:
l
R=
,
A
and if the effective cross sectional area A decreases, it would lead
to an increase in resistance (R).

b. Straight-Wire Inductance
In the medium surrounding any current carrying conductor, there
exists a magnetic field. If the current (I) is AC, this magnetic field is
alternately expanding and contracting (and even reversing direction
if there is no DC bias present). This produces an induced voltage (as
specified by the Faraday's law ) in the wire which opposes any
change in the current flow. This opposition to change is called "selfinductance" as shown in Figure 6.10.
The concept of inductance is important because at RF/MW, any and
all conductors including hookup wires, capacitor leads, bonding
wires and all interconnections tend to become inductors and exhibit
the property of inductance as shown in Figure 6. 11.

43

6.5.2 Resistors
DEFINITION-RESISTOR: Is an element specializing in the resistance
property, which is the property of a material that determines the rate
at which electrical energy is converted into thermal energy when an
electric current passes through it.
Resistors are used in almost all circuits for different purposes, such
as:
a. In transistor bias networks, to establish an operating point,
b. In attenuators (also called pads), to control the flow of power, and
c. In signal combiners, to produce a higher output power.
Once we depart from the world of DC, resistors start to behave
differently, i.e.,
At DC :
V = RI (Ohms law),
At low AC:
V RI ,

44

At high RF/MW:

V RI.

At RF/MW frequencies, a resistor (R) appears like a combination of

several elements as shown in Figure 6.12.

Figure 6.12a shows a simple resistor at DC, and as frequency

increases the lead wire inductances (L) bring about a higher resistor
value (R'>R) due to skin effect as well as parasitic capacitances, both
becoming prominent as shown in Figure 6.12b. The net effect of all
these parasitic elements, on the average, is a decrease in value of the
carbon-composition as well as metal resistors as shown in Figure
6.12c
NOTE: The 5 resistor graph in Figure 6.12c shows a slight
resonance due to the parallel combination of lead inductance and
capacitance which causes a small increase in the resistor value with
a subsequent decline as frequency is increased further.

45

There are several types of resistors, which can be briefly

summarized as follows
a. Carbon composition type resistors, which have a high capacitance
due to carbon granules parasitic capacitance,
b. Wirewound resistors which have high lead inductance,
c. Metal film resistors which are usually made up of highly resistive
films such as NiCr, etc.
d. Thin-film chip resistors which are produced on an Alumina or
Beryllia substrate and thus reduce the parasitic reactances greatly.
These four types of resistors are shown in Figure 6.13 a, b, c and d.

6.5.3 Capacitors
A capacitor was defined earlier in Chapter 3 and basically is any
device which consists of two conducting surfaces separated by an
insulating material or a dielectric. The dielectric is usually ceramic,
air, paper, mica, etc. The capacitance is that property which permits
the storage of charge when a potential difference exists between the
conductors, and is measured in Farads (see Figure 6.14).

46

The performance of a capacitor is primarily dependent on the

characteristic of its dielectric. It determines voltage and temperature
extremes of the capacitor at which it can be used. Thus any losses or
imperfections in the dielectric have an enormous effect on the circuit
operation. A few examples of different types of dielectric is shown
in Figure 6.15.

A practical capacitor has several parasitic elements which become

important at higher frequencies. The equivalent circuit of a real
capacitor is shown in Figure 6. 16.

In Figure 6.16 the elements are defined as follows:

C is the actual capacitance, L is the lead inductance, RS is the series
resistance, and RP is the insulation resistance (both creating heat and
loss).

47

The existence of parasitic elements, as shown in Figure 6.16, brings

the concept of real-world capacitors to the forefront which needs
further explanation:
PERFECT CAPACITORS: In a perfect capacitor, current will lead the
applied voltage in phase by 90 degrees. In phasor notation this can
be written as:
I = j
CV= C Vej90
(6.3)
Practical capacitors: In a real world capacitor, the phase angle ()
will be less than 90 degrees (i.e. < 90 ) as shown in Figure 6.17.
The reason < 90 is due to the existence of RS and RP (parasitic
resistances shown in Figure 6.16) which combine into one
equivalent resistor (REQ) as shown in Figure 6.18.

48

In a practical capacitor cos(), called the power factor (PF), can be

written as (see Figure 6.18):
R EQ
(6.4)
PF=cos(
)=
2
2
X C + R EQ
Usually REQ << XC where XC= 1/C. Therefore we can write:
PF=cos(
)
REQ/ XC
(6.5)
An important factor in practical capacitors or in general any
imperfect element is the Quality Factor (Q):
DEFINITION-QUALITY FACTOR (Q): Is a measure of the ability of an
element (or circuit) with periodic behavior to store energy equal to
2 times the average energy stored divided by the energy dissipated
per cycle.

Q is a Figure of Merit for a reactive element and can be shown to

be the ratio of element's reactance to its effective series resistance.
Thus For a capacitor, Q is given by:
1
Q = XC/REQ =
1/PF
(6.6)
CR EQ
From Equation (6.6), we can observe that for a practical capacitor, as
the effective series resistance (REQ) decreases, Q will increase until
REQ=0 which corresponds to a perfect capacitor having Q= ,i.e.,
REQ=0 PF=0, Q=
(a perfect capacitor)
(6.7)
The effect of these imperfections in a capacitor is shown in Figure
6.19.
From Figure 6.19, two distinct regions in the frequency response
plot of a capacitor can be identified. These two regions straddle the
resonance frequency (fr)as follows:

a. f < fr
In this region as frequency increases, the lead inductance's reactance
goes up gradually toward resonance (fr).

49

b. f > fr
In this region the capacitor acts like an inductor and is no longer
performing its intended function.
From Figure 6.19 we can conclude that we need to examine the
capacitor at RF/MW frequencies before final design and production.
This concept is shown in Figure 6.20 where the distinction between
the low AC and RF/MW is clearly shown.

6.5.4 Inductors
Definition-Inductor: is a wire which is wound (or coiled) in such a
manner as to increase the magnetic flux linkage between the turns of
the coil. The increased flux linkage increases the wire's selfinductance, as shown in Figure 6.21.

50

Inductors have a variety of applications in RF circuits such as in

resonance circuits, filters, phase shifters, delay networks and RF
chokes as shown in Figure 6.22.

Since there is no such thing as a perfect component, it is found that

amongst all components, inductors are most prone to very drastic
changes over frequency. This is due to the fact that the distributed
capacitance (Cd) and series resistance (R) in an inductor at RF/MW
plays a major role in the performance of an inductor as shown in
Figure 6.23.
From Figure 6.23, we can see that Cd exists due to a voltage drop in
the coil caused by internal resistance. The voltage drop causes a
voltage difference between two turns of the coil separated from each
other (with air as the dielectric). The aggregate of all small Cd's and

51

r's provides the equivalent circuit shown in Figure 6.24.

The effect of Cd on an inductor's frequency response is shown in

Figure 6.25. From this Figure (just like a capacitor) there are two
regions that straddle the resonant circuit.

52

These two regions can be identified as:

a. f < fr
In this region, the inductor's reactance (XL=L) increases as
frequency is increased.
b. f > fr
In this region the inductor behaves like a capacitor and as the
frequency is increased the reactance decreases.
At f = fr resonance takes place in an inductor and theoretically the
inductor's reactance is infinity; however, in practice, the total
impedance of the element is finite due to a non-zero series
resistance.
The quality factor (Q) of an inductor is defined to be:
Q= XL/RS = L/RS

(6.8a)

For a perfect inductor the series resistance is zero, thus we have:

RS= 0 Q=

(A perfect inductor)
(6.8b)
At low frequencies, Q is very large since Rs is very small, however,
as frequency increases the skin effect and winding distributed
capacitor (Cd) begin to degrade the Q of an inductor as shown in
Figure 6. 26.

From Figure 6.26, it can be seen that as frequency increases, Q will

53

increase up to Qo which is at f = fo. However, for frequencies

fo<f<fr, Rs and Cd combine to decrease the Q of the inductor toward
zero. At resonance (f=fr), where the total reactance of the element is
zero, inductor is no longer useful.
To extend the frequency range of an inductor (by increasing its Q),
we can do one of the following solutions:
a. Use a larger diameter for the wire which effectively reduces the
resistance value, or
b. Spread the winding apart which reduces the distributed
capacitance (Cd) between the windings, or
c. Increase the inductance (L) by increasing the permeability of the
flux linkage path by using a magnetic-core material. These are
shown in Figure 6.27. Solution (c ) is the most effective and
practical one of all the three solutions.

54

6.6 RESONANT CIRCUITS

A resonant circuit (also called a filter) is certainly not new and in
fact has been and is used in practically every transmitter, receiver or
piece of test equipment in existence.
Resonant circuits function is to pass selectively, a certain frequency
(or a frequency range) from the source to the load, while attenuating
all other frequencies outside of this passband as shown in Figure
6.28.

Since there is no perfect component, a perfect resonant circuit does

not exist and can not be built. However, knowing the mechanics of
resonant circuits, an imperfect resonant circuit (or filter) can be
tailored to suit our needs. A typical practical filters frequency
response is shown in Figure 6.29.

55

From this Figure several important features must be first defined:

a. DEFINITION-BANDWIDTH (BW): Is the difference between upper
(2) and lower (1) angular frequencies at which the amplitude
response is 3 dB below the passband response value (also called the
half-power BW). Therefore we can write: BW=(2 - 1)
b. DEFINITION-CIRCUIT Q: Is the ratio of center angular frequency
(o) to the bandwidth (BW), i.e.,
f
(6.9a)
Q = o /BW =fo/(f2-f1) = fo/
Where
BW =(
2 - 1) = 2
f,
f= f2-f1, o=2
fo,
2=2
f 2,
and 1=2
f1.
It is important to note that circuit Q should not be confused with
component Q which is a measure of component loss, while
Circuit Q is a measure of the selectivity of a resonance circuit
which means that as the bandwidth (BW) decreases, selectivity of
the resonant circuit increases. Furthermore it should be noted that

56

the component Q does have an effect on the circuit Q but the reverse
is not true.
c. DEFINITION - SHAPE FACTOR (SF)- OF A RESONANT CIRCUIT: is
defined as being the ratio of the 60-dB bandwidth to the 3-dB
bandwidth, i.e.,
f f
SF = 4 3
(6.9b)
f 2 f1
Shape factor (SF) is simply a measure of the steepness of the skirts.
The smaller the SF number, the steeper the response skirts. A
perfect filter has SF = 1, however, in practice SF is always greater
than one (SF 1). When SF is less than 1(SF<1), we have a physical
impossibility as shown in Figure 6.30.

d. DEFINITION-ULTIMATE ATTENUATION: Is the final minimum

attenuation that the resonance circuit presents outside of the
specified passband.
A perfect resonant circuit has an ultimate attenuation of infinity. If
there are response peaks outside of the passband, then this will
detract from the ultimate attenuation specification of the circuit as
shown in Figure 6. 31

57

e. DEFINITION-INSERTION LOSS: is the attenuation resulting from

inserting a circuit between source and load.
This concept is depicted in Figure 6.32. Therefore the insertion loss
is the attenuation that results from the insertion of a resonant circuit,
usually expressed in dB.

58

f. DEFINITION- RIPPLE: Is a measure of the flatness of the frequency

response of the resonance circuit and is defined to be the attenuation
difference (in dB) of the maximum value from the minimum in the
passband, i.e.,
=|max. attenuation - min. attenuation| (in dB)
This concept is shown in Figure 6.33.

6.7 RESONANCE
The term resonance and resonant frequency have exact
meanings, which we need to define at this point as:

59

DEFINITION-RESONANCE: Is a condition which occurs when the

reactive part of Z or Y of a circuit vanishes at one or more
frequencies. In this case Z or Y becomes purely a real number.
RESONANT FREQUENCY-

The frequency at which resonance occurs

is called the resonant frequency.
Resonant circuits are used to separate out the wanted signals from
spurious or unwanted signals, thus they can be considered to be
excellent filters. At resonance, the impedance of most circuits goes
through a sharp minimum or maximum, thus the selectivity of such
circuits is often defined in terms of the half power bandwidth
centered around the resonant frequency, which effectively defines
the width of the peak (when |Z| is max.) or notch (when |Z| is min.).

6.7.1 RLC Series Resonance

Series RLC circuit is shown in Figure 6.33b.

Figure 6.33b Series RLC Circuit

Z(j
)=Vi/I=RS+j(
L-1/
C)=RS+jX
Where X=(
L-1/
C).

(6.10a)

By definition, at resonance we should have X=0, which makes |Z|

achieve a minimum value of RS and thus makes it possible to obtain
the resonant frequency (o) as follows:
(
oL-1/
oC)=0 o2LC=1
Or,
o = 2f o = LC rad/s

60

As defined earlier,
Q=XS/RS=
oL/RS=1/RSoC
Thus we can write Equation (6.10a) as:
L/RS-1/RSC)]
Z(j
)=RS[1+j(
Or,
Z(j
)= RS(1+jQ(
/
o - o/
)]

(6.10b)

Using the half power frequencies (1 and 2) for the impedance:

1=
o-
,
2=
o+
,
|Z(j
1)|=|Z(j
2)|=|Z(j
o)|/
2
(6.10c)
We can see that from Equation (6.10c), we can write:
Q=
o /BW=
o/

Where BW=
=f2-f1.

6.7.2 RLC Parallel Resonance

Parallel RLC circuit is shown in Figure 6.33c.

Figure 6.33c Parallel RLC Circuit

Dealing with a parallel circuit, we need to use the concept of
admittance as follows:
Y(j
)= I /Vi=RP+j(
C-1/
L)=GP+jB
(6.11a)
Where GP=1/RP and B=(
C-1/
L).
By definition, at resonance we should have B=0, which makes |Y|
achieve a minimum value of GP (or |Z| achieve a maximum value of
RP). Thus we can obtain the resonant frequency (o) as follows:
(
oC-1/
oL)=0 o2LC=1

61

Or,
o = 2f o = LC rad/s
As defined earlier, for a parallel circuit we have:
Q=RP/XP=RPoC=RP/
oL
Thus we can write Equation (6.11a) as:
Y(j
) =GP[1+j(
C/GP-1/GPL)]
Or,
Y(j
)= GP [1+jQ(
/
o - o/
)]
(6.11b)
Using the half power frequencies (1 and 2) for the admittance:
o-
,
1=
2=
o+
,
|Z(j
1)|=|Z(j
2)|=|Z(j
o)|/
2
(6.11c)
We can see that from Equation (6.11c), we can write:
o/

Q=
o /BW=
Where BW=
=f2-f1.
Equations (6.11a) and (6.11b) for the parallel RLC circuit are the
dual of the equations obtained for the series case (i.e., Equations
6.10a and 6.10b). They are dual of each other since one can replace
Z with Y, L with C and R with G in the series equations and obtain
the actual equations for the parallel case.

6.7.3 Analysis of a Simple Circuit in Phasor Domain

Consider the circuit shown in Figure 6.34 which consists of a series
resistance RS and shunt element with impedance ZP.

We wish to calculate the total voltage gain (Vo/Vi) of the circuit.

62

Using KVL in phasor domain, we have:

Vo = ZP Ii,
(6.12)
(6.13)
Vi = (RS+ZP) Ii
Dividing (6.12) by (6.13) we obtain the total voltage gain as:
ZP
Vo/Vi= H(
) = |H| ej = |H| =
(6.14)
RS + ZP
The gain magnitude in dB would be given by:
ZP
(dB) (6.15)
20 log10|H| =20 log10 |Vo/Vi| =20 log10
R
+
Z
P
S
NOTE 1: From Equation (6.15) we can observe that output voltage
magnitude will always be less than or at best equal to the input
voltage magnitude which will be true for all passive circuits, i.e.,
For all passive circuits:
|Vo| |Vi|
(6.16)

NOTE 2: If the shunt element (in Figure 6.34) contains a capacitor

or an inductor then the impedance of the shunt element (ZP) will be
frequency dependent and so would the output voltage(Vo) or the total
voltage gain (Vo/Vi) as the following examples illustrate this point.
______________________________________________________
EXAMPLE 6.4
If the shunt element in Figure 6.34 is a perfect capacitor, calculate
and plot the voltage gain magnitude and phase.
Solution:
From the circuit shown in Figure 6.35 and Equation (6.14), we can
write the following:
ZP=1/j
C
1
(1 / jC) =
H(
)=|H|
=Vo/Vi=
(R + 1 / jC) 1 + jRC
1
tan 1 (RC)
=
2 1/ 2
[1 + (RC) ]
Or,
|H(
)|=20 log10|H|
= - 10 log10 [1+ (
RC)2] (in dB)
-1
= - tan (
RC)

63

The magnitude and the phase are plotted in Figure 6.36a,b. From the
magnitude plot we can see that this circuit performs like a low-pass
filter.
NOTE: Attenuation is 6 dB for every octave increase of frequency
(i.e. doubling the frequency). This is due to a single reactive
element. In general, for each significant reactive element added in
the circuit, the slope will increase by an additional 6 dB.

______________________________________________________

64

EXAMPLE 6.5
If the shunt element in Figure 6.34 is a perfect inductor, calculate
and plot the voltage gain magnitude and phase.
Solution:
From the circuit shown in Figure 6.37 and Equation (6.14), we can
write the following:
ZP= j
L
H(
)=Vo/Vi=

jL
1
=
=
R S + jL 1 jR S / L

1
1 + ( R S / L )
Or,

tan 1 ( R S / L )

|H| =20 log10|H| = -20 log10 [1+(RS/

L)2] (dB)
And,
= tan-1(RS/
L)

The magnitude and the phase are plotted in Figure 6.38a,b. From the
magnitude diagram we can see that this circuit performs like a highpass filter.

65

_______________________________________________________
EXAMPLE 6.6
If the shunt element in Figure 6.39 is a combination of a perfect
capacitor in parallel with a perfect inductor, calculate and plot the
voltage gain magnitude and phase.

Solution:
From the circuit shown in Figure 6.39, we can write ZP from

66

Equation (6.14) as the parallel combination of the capacitor and

inductor as follows:
jL
ZP= j
L || -j/
C =
(1 2 LC)
H(
)=Vo/Vi=

ZP
jL
=
2
R S + Z P R S R S LC + jL

|H| =20 log10|H| = 20 log 10

R 2 1 2 LC 2 + (L )2
S

(dB)

The magnitude is plotted in Figure 6.40.

NOTE 1: Near the resonance frequency of the tuned circuit, the slope
of the resonance curve increases to 12 dB/octave, because there are
now two significant reactances present and each one is changing at
the rate of 6 dB/octave (Therefore 12 dB/octave slope.)
NOTE 2: Away from resonance, only one reactance becomes
significant, therefore there would be only a 6 dB/octave of slope in
effect.

67

6.7.4 Loaded Q
The Q of a resonant circuit was defined earlier as:
Q=

o
.
BW

The "circuit Q" is often called the "Loaded Q", because it describes
the passband characteristics of the resonant circuit under actual "incircuit" or "Loaded condition."
In general, the "loaded Q" depends on three main factors as follows:
a. The source resistance (RS)
b. The load resistance (RL)
c. The component Q (of each of the reactive elements)
Figure 6.41 shows a block diagram of a resonant circuit and its
frequency performance.

68

From Figure 6.41 we can observe that:

a. The resonant circuit sees an equivalent resistance of RS
in parallel with RL as its true load. This is shown in Figure 6.42.

The loaded Q can be calculated by noting that:

RP = RS ||RL,
(6.17)
XP = the inductive or capacitive reactance of either of the reactive
components (since they are equal at resonance!)
Therefore:
Q = RP/XP

(6.18)

b. If RS or RL increases, then the equivalent resistance increases,

which will reduce the energy losses and thus will narrow the curve.
This will increase the selectivity and as a result the loaded Q .
c. For a fixed RP, if XP is decreased by choosing a smaller "L" or a
larger C, Q will increase. This point is illustrated in the next
example.
_______________________________________________________
EXAMPLE 6.7
Design a resonant circuit with a loaded Q=1.1 at f=142.4 MHz that
operates between a source resistance of 100 and load resistance
of 100 . Discuss how to increase Q. Use perfect components.
Solution:
RP=100 || 100 = 50
XP =RP/Q= 50/1.1= 45.45 = L = 1/C
Choose: L=50 nH, C=25 pF ,
Given fixed RS and RP, we can increase Q by 20 times through
scaling up the capacitor value by 20 while scaling down the inductor

69

value by 20, i.e.,

Q=22 Cnew=500 pF, Lnew= 2.5 nH
These two cases are shown in Figure 6.42.
Figure 6.43 shows two equivalent circuits for two different Qs
obtained by scaling the inductors and the capacitors values
appropriately.

_______________________________________________________
Therefore a circuit designer has two design approaches in designing
resonant circuits:
a. Select an optimal value of RS and RL to get the specified Q, or
b. Given RS and RL, select component values of L and C to
optimize Q.
NOTE 1: If poor quality reactive components (i.e. low Q) are used in
highly selective resonant circuits, the net result is that we effectively
place a low-value shunt resistor directly across the circuit which
will drastically reduce its loaded Q and increase the bandwidth.
NOTE 2: At resonance, an ideal LC parallel circuit has a very high
(ideally infinite) total impedance as shown in Figure 6.44

70

NOTE 3: Usually we only need to involve the Q of the inductors in

the loaded-Q calculations, since the Q of most capacitors are quite
high over their useful frequency range which means that they have a
very small resistive passive element.

6.7.5 Impedance Transformation

The most common type of impedance transformation is from the
series elements to shunt elements as shown in Figure 6.45

We first define the component Q (represented by Qc) as:

QC = QS = QP, (component Q)
(6.19)
Equation (6.19) states that the Q remains the same in the process of
series-to-shunt transformation which is a correct assumption since
we are still dealing with the same element even though we are
changing its equivalent circuit.
Through simple mathematical manipulation, we can write:
RP = (QC2 +1) RS
(6.20)
XP = RP/QC
(6.21)
Using (6.21) for a shunt configuration, QC is defined as:
QC = RP/XP (shunt)
(6.22)
which is in contrast with the series configuration earlier defined as:
QC =XS/RS (series)
(6.23)
_______________________________________________________
EXERCISE 6.1
It will be a worthwhile exercise to derive the above impedance

71

transformation equations, i.e. Equations (6.20) and (6.21).

HINT: Set Zin of each circuit (Real and Imaginary), equal to each
other.
_______________________________________________________
EXAMPLE 6.8
An imperfect inductor has an inductance of 50 nH with a series loss
resistance of 10 . Find the following:
a. QC at 100 MHz
b. The equivalent parallel configuration at f=100 MHz
Solution:
a. QC = XS/RS=2f L/RS = 2 x 100x106 x 50x10-9 /10 = 3.14
b. RP=(QC2+1)RS=(3.142+1) x 10= 108.7
XP=RP/QC=108.7 / 3.14 =34.62
XP= LP LP= XP/ =55.1 nH
The equivalent circuit is shown 6.46.

______________________________________________________

6.7.6 Insertion Loss(IL)

If the inductor and the capacitor were perfect components with no
internal loss, then the insertion loss for LC resonant circuits would
have been zero dB.

72

In actuality, this is not the case and insertion loss is a very critical
parameter in specifying a resonant circuit as shown in Figure 6.47.
DEFINITION- INSERTION LOSS : Is a positive number expressing the
difference (in dB) between the power received at the load before and
after the insertion of a circuit or component in the transmission line
connecting a source to a load.
Mathematically, we can write:
Insertion Loss =IL(dB)= -10log10(Vo'/Vo)2 =-20log10(Vo'/Vo)
Where Vo is the voltage at the load before the insertion, whereas Vo
is the voltage after the circuit is placed in the line. For example, a
perfect LC circuit with no resistive loss has zero dB insertion loss as
shown below.

The following example will illustrate the concept of insertion loss

further.
______________________________________________________
EXAMPLE 6.9
Calculate the insertion loss of the LC resonant circuit shown in
Figure 6.48 at f=1430 MHz.
RS=RL= 1 k
Inductor: L= 0.05 H, QC1=10
Capacitor: C=25 pF, QC2=

73

Solution:
a. Removing the LC circuit gives:
Vo=1000/(1000+1000) x Vi= 0.5 Vi
b. Next we convert the inductor's series configuration into parallel:
QC1=XSL/RSL RSL=2 x 1430x106 x .05x10-6/10 = 45
RPL=(QC12 + 1) RSL RPL= 4.5 k
Therefore at resonance we have a circuit as shown in Figure
6.49:

(RL)EQ= 1 k || 4.5 k = 820

Thus the load voltage with resonant circuit in place is:
Vo' = 820/(1000+820) Vi = 0.45 Vi
Insertion Loss (IL)=Vo'/Vo = 0.45/0.5=0.9
IL(dB)= -20log10(0.9) = 0.92 dB
An insertion loss of 0.92 dB may not appear much but can add
up very quickly if we cascade several resonant circuits together.
_______________________________________________________
EXAMPLE 6.10
Design a simple parallel resonant circuit to work between a source
resistance (RS) of 1 k and load resistance (RL) of 1 k to provide a

74

3dB bandwidth of 10 MHz at a center frequency (fo) of 100 MHz.

Also calculate the insertion loss. Assume all capacitors are perfect
and the inductor has a Q of 85.
Solution:
Considering the circuit shown in Figure 6.50, we have:
Q = fo/f = 100/10 = 10

For inductor:
QC=RP/XP = 85 RP=85 XP
(1)
The loaded Q for the circuit is:
Q=10= Rtot/XP = (RP||1 k||1 k)/ XP
(2)
Solving (1) and (2) for RP and XP we obtain:
L=XP/o=70 nH
XP=44.1 =oL=1/oC C=1/oXP = 36 pF
RP = 3.75 k
To find the insertion loss we note two cases as follows:
a) Without the resonant circuit
Vo= 1 k/ (1 k+1 k)=0.5 Vi
b) With the resonant circuit in place
(RL)EQ=RP||RL = 3.75 k||1 k=789.5
Vo' = Vi x 789.5/(1000+789.5) = 0.44 Vi
I.L.(dB)= 20 log10 (0.44 Vi/0.5 Vi)= 1.1 dB
_______________________________________________________

6.8 IMPEDANCE TRANSFORMERS

By observation, it becomes apparent that low values of RS and RL
tend to load down a given resonant circuit, leading to a decrease of
its Q and broadening of its bandwidth. Thus it is very difficult to

75

design a high-Q simple LC resonant circuit that would function well

between two low values of RL and RS as shown in Figure 6.51.

To solve this design problem, two types of impedance transformer

may be used as shown in Figures 6.52 and 6.53. The impedance
transformer placed between the load and the source in the circuit
converts RS or RL to a much larger resistance. Thus the Q of the
circuit is pushed to a higher value, since the resonating circuit as a
whole is presented with a higher resistance value.
As shown in Figures 6.52 and 6.53 there are two types of
transformers:
a. Tapped-C transformer and,
b. Tapped-L transformer.
These are described next.

6.8.1 Tapped-C Transformer

RS'= RS(1+C1/C2)2,
(6.24a)
C1 C 2
CT =
(6.24b)
C1 + C 2
Where "CT" is the equivalent capacitance that resonates with "L".

76

6.8.2 Tapped-L Transformer

RS' = RS (n/n1)2
(6.25)
LT = L
(6.26)
The transformer circuits (a) and (b) present a much larger RS and RL
that is actually present. For example for a circuit with RS = 50 , the
transformer would turn the 50 into a 500 and the circuit will be
able to see a higher RS and thus its Q would be much higher.
______________________________________________________
EXAMPLE 6.11
Design a resonant circuit (as shown in Figure 6.54) such that it
operates between RS=50 and RL=2000 with a Q=20 at the
center frequency fo=100 MHz.. The inductor has a QC = 100 at 100
MHz. You may use a tapped-C transformer to achieve the desired Q.
Solution:
We will use a tapped-C transformer to step up RS=50 to 2000 in
order to match the load resistance for maximum power transfer.
RS'=RS(1+C1/C2)2 2000=50(1+C1/C2) 2 C1/C2= 5.3
Inductor: QC=RP/XP=100 RP=100 XP
(1)

77

Q= Rtot/XP = (RS' || RL || RP)/XP

(2)

Using (1) and (2), we solve for RP and XP:

RP=4 k
XP=40 L=XP/o = 63.6 nH
CT= 1/ XPo = 39.8 pF
CC
CT=39.8= 1 2
C1/C2=5.3
C1 + C 2
If we select C1=250 pF, then:
C2=47 pF
The final design is shown in Figure 6.55.

6.9 RF IMPEDANCE MATCHING

Impedance matching is often necessary in the design of RF circuitry
to provide maximum possible transfer of power between a source

78

and its load as shown in Figure 6.56.

One of the applications of an RF impedance matching network is in

the front end of any sensitive receiver where the signal is extremely
weak and none of it can be wasted due to mismatch. Therefore use
of an appropriate matching network becomes crucial to the overall
performance of the circuit.
The maximum power transfer theorem states that:
For DC: ZS = RS = RL = ZL
(i.e. There is no reactance)

(6.27)

For AC: ZL = ZS* RS = RL

XS = -XL

(6.28a)
(6.28b)

NOTE: XS = -XL is valid only at one frequency (the frequency of

resonance). Therefore, a perfect match can occur only at the
resonant frequency which poses a problem in broadband matching
of circuits.
At all other frequencies removed from the matching center
frequency, the impedance match becomes progressively worse and
eventually non-existent as shown in Figure 6.57.
There are an infinite number of possible networks that could be used
to perform the impedance matching function. For Example a circuit
as simple as a 2-element LC network or as elaborate as a 7-element
matching network would work equally well. But first we will
analyze a simple matching network as illustrated in the following

79

example.

______________________________________________________
EXAMPLE 6.12
Analyze the LC matching network (shown in Figure 6.58) which
transforms a source resistance RS = 100 to a load of RL = 1000
.

Solution:
From Figure 6.59 we have:

80

Z1= 1000||-j333 = 100-j300

Z2= j300 + (100-j300) = 100

Therefore the source sees a 100 as the total load which creates a
matched condition with the source resistance (RS=100 ) and
therefore maximum power transfer (for Vi=1 V) occurs as follows:
Vo=0.5Vi
(PL)max =1/2 (Vo2/Rin)
= 1/2(0.52 /100) = 1.25 mW= 0.97 dBm
These calculations bring us to an important question:
How much power would have been transferred if matching were not
placed between RS and RL?
From the diagram shown in Figure 6.60, we can write:
Vi=1 V
Vo= (1000/1100)Vi= 0.91 Vi
PL= (Vo2/RL)/2 = (0.912 /1000)/2 = 0.41 mW = -3.83 dBm

Therefore compared to the matched case the power loss would have
been:

81

Power loss= 0.41/1.25=0.328= - 4.89 dB

Which shows that only 1/3 of the available power is transferred and
the 2/3 remaining is wasted (or reflected back to the source) due to
mismatch.
NOTE : The function of the shunt component is to transform a larger
impedance down to a smaller value with a real part equal to the
real part of the source impedance and the reactive component
capable of resonating(or canceling out) with the reactive part of the
source impedance (see Figure 6.61).

_______________________________________________________

6.9.1 The L-Network

The simplest and most widely used matching circuit for lumped
elements is the L- network as shown in Figures 6.62 a, b, c, and d.

82

The generalized configuration is shown in Figure 6.63 where we can

write:
RP = (Q2 +1)RS
(6.29)
(6.30)
Q = QS = QP
QS = QP = ( R P / R S ) 1
(6.31)
QS = XS/RS
(6.32)
QP = XP/RP
(6.33)
where QS, RS and XS are for the series leg and QP, RP and XP are for
the parallel leg.

NOTE: XS and XP may be either capacitive or inductive reactances

but must be of opposite types. The following example illustrates the
concept of L-networks more clearly.
_______________________________________________________
EXAMPLE 6.13
Design a circuit to match a 100 source resistance to a 1000
load resistance at fo = 100 MHz. Assume that a DC voltage must
also be transferred from the source to the load and all elements are
perfect.
Solution:
The need for DC at the output dictates the need for an inductor in the
series leg as shown in Figure 6.64.
RS=100 , RP=1000
QS=QP=(1000/100 - 1)1/2 = 3
QS=XS/RS XS=3RS=300
QP=RP/XP XP=RP/3=1000/3=333

83

XS=L L=XS/ = 477 nH

XP=1/C C=1/XP= 4.8 pF
NOTE: This circuit was analyzed earlier in example 6.12.
_______________________________________________________
The previous examples dealt with matching two real impedances.
However in actual practice we deal with transistors, transmission
lines, antennas, etc., which all present complex input and output
impedances as shown in Figure 6.65.
There are two basic approaches in handling complex impedances :
a. The Absorption Method;
b. The Resonance Method.
These two methods are explained below:

84

6.9.2 The Absorption Method

This is a method in which any stray reactances can be absorbed into
the impedance-matching network by prudent placement of each
matching element such that the following occurs:
1. Element capacitors (C) are placed in parallel with stray capacitors
(Cp),
2. Element inductors (L) are placed in series with any stray inductors
(Lp).
3. Next, the stray component values are then subtracted from the
calculated element values to arrive at the final matching network.
The new element values C' and L', are given by:
(6.34)
C = Cp + C' C' = C-Cp
'
'
L = Lp + L L = L-Lp
(6.35)
The following example further illustrates this point.
_______________________________________________________
EXAMPLE 6.14
Use the absorption method to match the source (100+j126 ) to a
load (1000 +j795.8) at 100 MHz as shown in Figure 6.66.

85

Solution:
Step 1: Totally ignore the reactances and simply match 100 to
1000 such that the inductors are in series and capacitors are in
parallel. This step has already been done in example 6.13 and we use
the results for the matching network directly (see Figure 6.67),i.e.
L=477 nH,
C=4.8 pF.

Step 2: The new elements are given by:

L' = L-Lp=477-200=277 nH
C' = C-Cp = 4.8 -2 =2.8 pF
Based on these values, the final design is shown in Figure 6.68.

86

NOTE: This method is workable only if the calculated element values

(L,C) are higher than the stray values (Lp,Cp).
_______________________________________________________

6.9.3 The Resonance Method

is a method in which any stray reactances are resonated with an
equal and opposite reactances at the frequency of interest. Once this
is done, the design proceeds the same way as two pure resistances:
one at the source and the other at load.. The following example
illustrates this point.
_______________________________________________________
EXAMPLE 6.15
Design a matching network that will match a source resistance of 50
to a capacitive load (at fo= 75 MHz) as shown in Figure 6.69. The
matching circuit should block the DC to the output. Use the
resonance method.

Solution:
Step 1: Resonate 40 pF with a shunt inductor (L1) with the following
value (see Figure 6.70):

L1=1/C L1=1/C2 L1=112.6 nH

Step 2: Now we match 50 to 600 by the same technique as
before (see Figure 6.71):

87

QS=QP== (600 / 50) 1 = 3.32

XS=QSRS=50 x 3.32 =166
XP= RP/QP = 600/3.32 = 181

Using a series cap to block the DC: C=1/XS=12.8 pF

L2= XP/ = 384 nH
Combining L1 and L2 in parallel (as shown in Figure 6.72) we
obtain:
LL
Ltot= 1 2 = 87 nH
L1 + L 2

Figure 6.72 Using L1 and L2 to match a capacitive load to 50 .

The final circuit is shown in Figure 6.73.

88

6.10 THREE-ELEMENT MATCHING

We can observe with the L-networks that once Rs and Rp are given,
the Q of the network is defined and the designer no longer has a
choice over its value because:
Q = (R P / R S ) 1
(6.36)
If a narrow bandwidth is desired, this will cause a design problem.
An example of a circuit with a low Q as shown in Figure 6.74.

The use of 3-element networks overcomes this disadvantage and can

be used for narrow-band high-Q applications. The minimum Q
available is the circuit Q established with an L-matching network.
There are Two types of three-element networks as shown in Figure
6.75.

Using more than 3 elements in the matching network design would

bring about a greater amount of design flexibility which can lead to
tedious mathematical equations. For that reason mathematical
calculations of the reactive elements X1,X2 and X3, which are too
complicated, is omitted here. Instead we will show a simpler way to
calculate these values later using a graphical tool called a Smith
chart. As will be seen shortly, the use of a smith chart greatly
simplifies a very complex design process.

89

Chapter 6- Symbol List

A symbol will not be repeated again, once it has been identified and
defined in an earlier chapter, with its definition remaining
unchanged.
AWG American Wire Gauge
BW Bandwidth
Cd- Distributed capacitance
CT Total capacitance of a center tapped capacitor
f0 Frequency where Q is maximum; Center of passband.
fr Resonance frequency
H Voltage Gain
IL Insertion Loss
PF Power Factor
- Permeability
Q Quality Factor
Q0 Maximum Quality Factor where f= f0
RP Insulation resistance
RS Series resistance
SF Shape Factor
VP Phase velocity
- Conductivity
- Skin depth or depth of penetration
- Phase angle
Dielectric constant of a material
r Relative dielectric constant of a material (=/o)
- Resistivity

CHAPTER-6 PROBLEMS
6.1)

6.2)
6.3)

Design a resonant circuit with a loaded Q of 50 that operates

between source of 100 and a load of 2000 at a frequency
of 100 MHz.
Transform a series configuration of a 250 nH inductor into an
equivalent parallel configuration at 50 MHz (Q=10).
Using the tapped-C method, design a resonant circuit with a
loaded "Q" of 40 at a center frequency of 100 MHz that
operates between a source resistance of 100 an a load

90

6.4)

resistance of 3000 . The capacitors are all lossless and the

inductor has a Q of 100 at 100 MHz.
Design a simple parallel LC resonant circuit to provide a
bandwidth of 10 MHz at a center frequency of 100 MHz.
The resonant circuit is operating between a source and a load
impedance of 2000 each. The capacitor is lossless and the
Q of the inductor is 85. Calculate the insertion loss of the
resonant circuit in operation.

Figure P6.4
Using an L-network, design a circuit to match a 50 source
resistance to an 850 load at 50 MHz. Assume that the DC must
also be transferred from the source to the load.
6.5) Using the absorption method, design a matching network to
match the source and the load at 50 MHz as shown in Figure
P6.6.

6.6)

Figure P6.6
Using the resonance method, design an impedance matching
network that will block the flow of DC from the load as
shown in Figure P6.7. Assume f=100 MHz.

91

Figure P6.7

REFERENCES
[6.1] Bowick, C. RF Circuit Design. Carmel: SAMS-Prentice Hall,
1982.
[6.2] Carr, J. J. Secrets of RF Circuit Design. New York: McGrawHill, 1991.
[6.3] Gottlieb, I. W. Practical RF Power Design Techniques. New
York: McGraw-Hill, 1993.
[6.4] Harsany, S. C. Principles of Microwave Technology.
Upper Saddle River: Prentice Hall, 1997.
[6.5] Krauss, H. L., C. W. Bostian, and F. H. Raab. Solid State
Radio Engineering. New York: John Wiley & Sons, 1980.
[6.6] Lenk, J. D. L enks RF Handbook. New York: McGraw-Hill,
1992.
[6.7] Matthaei, G., L. Young, and E. M. Jones Microwave Filters,
Impedance-Matching Networks, and Coupling Structures.
Dedham: Artech House, 1980.
[6.8] Scott, A. W. Understanding Microwaves. New York: John
Wiley & Sons, 1993. [6.9] Vizmuller, P. RF Design Guide.
Norwood: Artech House, 1995.

92

CHAPTER 7

The Wave Fundamentals

7.1 INTRODUCTION
The subject of "RF/Microwaves" primarily deals with electrical energy at
high frequencies. Therefore to know microwaves, one needs to know the
three qualities of energy in general.

7.2 Qualities of Energy

The following qualities apply to any and all types of energy whether
electrical, mechanical, chemical, etc. at high or low frequencies.
However, since we are dealing with electronics, we will narrow the
following discussion to electrical energy and waves only.
A. QUALITY #1: EXISTING CHARACTERISTICS
These characteristics can be divided into three classes:
1. A Flow: is the transfer of energy from one point to another. The
energy in a flow can have any type of waveform. So a flow is a transfer.
This is shown in Figure 7.1.

93

2. A divergence (also referred to as a "dispersal"): is a generalized

case of a "flow" where a number of flows travel from or to a common
center as shown in Figures 7.2 (a) and (b).

NOTE: "A divergence" is similar in concept but different (in definition)

from "divergence of a vector quantity" which is an exact mathematical
operation measuring the net outflux (or influx) of a vector quantity.
3. A standing wave (also called a ridge of energy): is energy suspended
in space and comes about when two flows or divergences of
approximately equal magnitude and exact frequency impinge against one

94

another with sufficient amplitude to cause an enduring state of energy,

which may last after the flow itself has ceased. For example, a resonator
or a cavity oscillator falls into the category of devices that generate this
type of wave characteristic. A few examples are shown in Figures 7.3 (a)
and (b).

B. QUALITY #2: WAVELENGTH

Wavelength is a characteristic of an orderly flow of motion and describes
its regular and repeated pattern by the distance between its peaks. Many
motions are too random and too chaotic to have an orderly flow and thus
have no wavelength.
DEFINITION- WAVELENGTH: Is defined to be the physical distance
between two points having the same phase in two consecutive cycles of a
periodic wave along a line in the direction of propagation as shown in
Figure 7.4.
As frequency increases, the wavelength () decreases as can be observed.
Thus higher frequency waves have shorter wavelengths as already
discussed in Chapter 5 (see Figure 7.4).

95

Wavelength has no bearing on the wave characteristics (quality #1) but

applies to the repetition property of the wave flow. A standing wave has
a potential flow when released, therefore may be considered to have a
wavelength even though it is not a flow or a wave in the truest sense of
the word.
If a random wave is periodic, it can be considered to have a wavelength
using "Fourier theorem". It can be proven mathematically that through
the use of Fourier analysis, any wave can be decomposed into its Fourier
harmonics, provided that the wave is continuously flowing and periodic
as shown in Figure 7.5.

C. QUALITY #3: A FLOW'S DIRECTION (OR ABSENCE THEREOF)

This quality describes the direction or the absence of direction of flow. A
few examples are shown in Figure 7.6. This quality is an important one,

96

since energy can have a flow with no net transfer of energy i.e. absence
of the direction of flow.

For example, a wave traveling from a transmitter to a receiver, or

electrons moving in a wire under the influence of an electric field is said
to have a "direction of flow".
Examples of absence of direction include a) a free electron moving in the
lattice of a solid at equilibrium (i.e. when no external field is applied)
which is a flow with an absence of direction or b) an electron in an atom
moving in an orbit around the nucleus. Both are flows without a net
transfer of energy.

7.3 DEFINITION OF A WAVE

So far we have loosely used "a wave" to mean a special case of a flow of
energy. Now we need to define it exactly:
DEFINITION- A WAVE: Is a disturbance that propagates from one point
in a medium to other points without giving the medium, as a whole, any
permanent displacement.
This general definition of a wave includes any and all disturbances that
could be of electrical or non-electrical origins. However, now we further
restrict our definition to a special class of waves which are of electrical
origin. These waves are called electromagnetic (EM) waves. Now, we
need to define an important term:

97

DEFINITION-AN ELECTROMAGNETIC (EM) WAVE: Is a radiant energy

flow produced by oscillation of an electric charge. In free space and
away from the source (which is moving electric charges), EM rays of
waves consist of vibrating electric and magnetic fields which move at the
speed of light (in vacuum), are at right angles to each other and to the
direction of motion.
The propagation of a simple electromagnetic wave in free space is shown
in Figure 7.7. EM waves propagate with no actual transport of matter and
grow weaker in amplitude as they travel farther in space.

EM waves include Radio, microwaves, infrared, visible/ultraviolet light

waves, X-, gamma- and cosmic- rays. (See Electromagnetic spectrum in
Chapter 6).
These are all different types of electrical energy and all follow the same
principles that we have discussed so far in this chapter (see Figures 7.8
and 7.9).

98

On a larger view of things, we can observe that RF and Microwaves

are a special case of EM waves, which itself is a subset of larger
field of study, i.e., waves. Of course, this last itself is a subset of a
much larger sphere of existence known as "energy", as shown in
Figure 7.10.

7.4 MATHEMATICAL FORM OF PROPAGATING

WAVES
We know that f(x-xo) is the same function as f(x) except shifted to the
right a distance "xo" along +x axis. If instead, we consider f(x-vt)

99

then the function f(x) is shifted to the right a distance xo=vt, where
"v" and "t" can be considered to be the velocity of motion and the
elapsed time, respectively. The distance (xo) increases as time
elapses, therefore the function is displaced continuously farther out
along the +x axis as time elapses.

7.4.1 An Important Special Case: Sinusoidal Waves

Assume f(x) is a sinusoidal function:
f(x)=A cos
x
(7.1)
Where A is the amplitude and is the phase constant. Then a sinusoidal
wave propagating in +x direction would be represented in time and
phasor domain by:
a. Time domain form:
f(x,t)= Acos
(x-vt)
= Acos(
x-
t)
(7.2)
Or,
b. in Phasor form:
F=Ae-jx
(7.3)
Where (=
v) is the angular frequency.
To find the wavelength (), we know that it is defined to be the physical
distance between two peaks (or valleys). We note that at t=0, the wave's
peak is at x = 0. The next peak is at x = and the sinusoidal wave has a
phase of 2, thus:
= 2
= 2
/

(7.4)
For the wave propagating in the "-x" direction, the following can be
written:
c. Time domain: f(x,t)=Acos(
x+
t)
(7.5)
Or,
d. Phasor domain: F=Aejx
(7.6)
The phase velocity (VP), which is defined to be the velocity at which the
plane of the constant phase propagates, can be obtained from:
x-
t = k,
(7.7)
Where k is an arbitrary constant.

100

Differentiating Equation (7.7) with respect to time gives the phase

velocity:
dx/dt-
=0 Vp=dx/dt=
/

(7.8)
In an unrestricted or "free" space, a plane wave travels at velocity VP
which is given by:
=1/ 0 0
Where 0 and 0 are the permeability and permittivity of free space.
Equations (7.2) and (7.5) show a simple wave that keeps its size and
shape while propagating at a constant velocity VP. This type of
propagation is said to be undistorted and unattenuated since it is
propagating in free space (or vacuum) which is a non-dispersive
medium.
DEFINITION-A DISPERSIVE MEDIUM: is a medium in which the phase
velocity (VP) of a wave is a function of its frequency.
This means that a complex wave, consisting of several frequencies,
travels through a dispersing medium at different velocities i.e. each
frequency component travels at Vp=/ with different time delays. This
would cause the wave to be distorted at the exit point.
For example a square-pulse waveform entering and traveling through a
dispersive medium will lose its shape and will appear rounded at both of
its edges when exiting the medium.
EM waves can have a "rise and fall" as well as an "advance and retreat"
type of oscillation of the field quantity, as they propagate (see Figure
7.11).

7.4.2 Types of Waves

Waves are like fluids and propagate according to the medium in which
they find themselves. If the medium is unrestricted, then it would be
called "Free space wave propagation". When the source is a point and the
medium of propagation is free space, waves have spherical wavefronts as
shown in Figure 7.12.

101

7.4.3 A Special Case: Plane Waves

When waves under consideration are at an infinite distance away from
the source of disturbance, then the wavefront of each wave is a plane
surface and these waves are called plane waves (see Figure 7.13). These
plane waves are in the TEM mode of propagation.

102

DEFINITION-TEM (TRANSVERSE ELECTRO-MAGNETIC) MODE: Is

defined to be waves having the electric and magnetic fields
perpendicular to each other and to the direction of propagation. These
waves have no field components in the direction of propagation.
A typical TEM wave in free space is shown in Figure 7.14.

Figure 7.14 A typical TEM wave in free space.

The mathematical expression, a(x,t), for the plane wave propagation is

defined below.
a. General time domain form is given by:
a(x,t)=Ao f(
t-
x)
(7.9)
For a time harmonic wave, we can write:
a(x,t)=Aocos(
t-
x)
-j
x j
=Re(Aoe e t)
(7.10)
b. In phasor domain, we have:
A(x)= Aoe-jx,
(7.11)
Which is a plane wave propagating in +x direction as shown in Figure
7.12 and 7.13.

103

7.5 PROPERTIES OF WAVES

There are several properties of waves that are worthy of consideration at
the outset of this section:
PROPERTY #1: FLOW PROPERTY
This property is in common with quality #1 for energy. A wave is a flow.
It goes from point "A" to point "B" and in doing so a transfer of energy
takes place, of course with a reduced amplitude at the destination, as
shown in Figure 7.15.

PROPERTY #2: WAVELENGTH PROPERTY

This property was discussed as quality #2 of energy. A wave with a
regular and periodic (or repeating) waveform has a wavelength which is
the physical distance between two peaks (or valleys) in two consecutive
cycles as defined earlier in a more precise way. This concept is shown in
Figure 7. 16.

104

In order to derive wavelength () we know that the speed of propagation

(v) is uniform thus the distance () traveled in one wave period (T=1/f)
is:
= vT= v/f
(7.12)
At high RF and microwaves the wavelength ranges from one meter to
one millimeter corresponding to a frequency of 300 MHz to 300 GHz.
PROPERTY #3: REFLECTION AND TRANSMISSION PROPERTY
When a wave encounters an obstacle or a different medium, some of it
reflects back (called a reflected wave) and the rest of it transmits through
(called a transmitted wave). This is true for any and all types of waves.
_________________________________________________________
EXAMPLE 7.1a: PERFECT REFLECTION
How does a perfect mirror behave for an incident wave?
Solution:
For a perfect mirror we have perfect reflection, i.e. 100% of the incident
wave reflects back and zero transmission takes place as shown in Figure
7.17.

_________________________________________________________
Example 7.1b: Perfect transmission
What would constitute a perfect transmission condition?
Solution:
For a perfect transmission, the two media have to be identical in their
electrical properties (such as permittivity, permeability, etc.) as shown in

105

Figure 7.18. This means that for this condition to occur, the second
medium has to continue to behave electrically the same as the first.

__________________________________________________________
Property #4) Standing-wave property
When two waves of exactly the same magnitude and frequency travel
opposite to each other, the result is not a wave but an "Oscillation with
no propagation" called a "Standing wave" which has a fixed location, as
shown in Figure 7.19.

The standing wave can be written mathematically in:

a. Phasor domain:
Ae-jx + Ae+jx= 2Acos
x
b. Time domain:
2Acos(
x)cos(
t)

(7.13a)
(7.13b)

106

Since Equation (7.13b) is not of the form f(x-t), thus it is not a wave
but a pure oscillation at a fixed location!
NOTE: A definite pre-requisite for a standing wave is two opposite waves
of exact frequency. However, their amplitudes should be comparable, if
not equal. The result would be a standing wave plus a traveling wave
and not a pure standing wave as described above.

7.6 TRANSMISSION MEDIA

When waves are constricted to a limited transmission space (also called a
line, guide, channel, etc.), then the waves take on different forms and
patterns according to the shape of the guide, just like fluid flow in a pipe.

7.6.1 Types of Transmission Media

A few examples of the wave patterns in different transmission media are:
a) Coaxial line, b) two-wire transmission line, c) a waveguide, d) a
microstrip line, e) a parallel plate waveguide, and f) a stripline, as shown
in Figure 7.20.
Generally, any and all of these five transmission media could be called
"transmission lines", but the terminology has been made more specific to
convey more exact concepts, thus:
1. (a), (b), (e) and (f) are generally labeled as transmission lines (TLs).
2. (c) Is labeled a waveguide and,
3. (d) Is labeled a Microstrip line.
(a), (b), (e) and (f) all will support propagation of Transverse Electromagnetic (TEM) waves and will be used specifically in this book. An
example of a TEM wave was shown earlier in Figure 7.14, where the
direction of propagation is perpendicular to the oscillating electric and
magnetic fields.
NOTE 1: Structure (d), a microstrip line, supports a quasi-TEM wave
which is a wave with a small axial field. This type of transmission line
has gained tremendous popularity in Microwave integrated circuits due
to its planar structure and ease of fabrication using printed circuit
technology. Microstrip lines will be discussed in detail in a later section.

107

108

NOTE 2: Structures (a) and (c), a coaxial line and a waveguide, are
closed structures and are preferred since they have much less radiation
losses than the other three open structures.
NOTE 3: Structure (f), a stripline transmission line, can be thought of as
a "flattened out" coaxial line, where both have a center conductor which
is enclosed by an outer ground conductor with a uniform dielectric
material filling the space between the two.
NOTE 4: Other types of transmission media include: Slotline, Dielectric
waveguide, Coplanar waveguide and Ridge waveguide. These
transmission media have non-TEM modes of propagation and are
beyond the scope of this book and can be found in advanced texts.
A summary of Transmission media and their different characteristics is
shown in table 7.1. The comparison made in this table can be roughly
divided into two important general areas:
a. Electrical considerations: mode of propagation, dispersion,
Bandwidth, Power loss and power capacity (items 1 through 5), and
b. Mechanical considerations: Physical size, ease of fabrication and ease
of integration with other elements and components (items 6 through
8).

109

7.6.2 A Short History of Transmission Media

Waveguides were used for most microwave systems during 1930s and
1940s but they have a limited bandwidth, are bulky and expensive, even
though they have the advantage of being able to handle high powers
much needed for Radar applications. During this same period, coaxial
lines were also developed as a broadband and medium power
transmission line, but they are difficult to integrate into or fabricate in the
integrated circuit technology which is suited to planar type transmission
lines.
Planar transmission lines received attention in 1950s. They are low cost
compact and capable of being integrated with planar microwave
integrated devices and circuits. Therefore, they play an important role in
planar microwave technology for transmission of signals between
devices, circuits and networks. Examples of planar transmission lines
include microstrip line (developed in 1952), stripline (developed circa
1955), slotline (developed in 1969).
Other planar transmission lines (e.g. coplanar waveguides, finlines, etc.)
have also been developed through time. Overall and amongst all planar
transmission lines, none have proven as popular as microstrip line
technology, which has gained tremendous interest in planar circuit
applications. For this reason microstrip lines are discussed and analyzed
in depth in a later section in this chapter.

7.6.3 Waves on a Transmission Line (TEM Mode)

When we mention a "transmission line", it is commonly understood to be
any system of conductors suitable for conducting TEM-mode
electromagnetic waves efficiently between two or more terminals.
Common examples of TEM-mode transmissions lines are telephone
lines, power lines, coaxial lines, parallel plate lines, etc.
At lower frequencies the length of the line is much smaller than the
signal wavelength and thus the transmission line can be treated as a
"lumped element" with almost zero loss and no time delay for signal
propagation between two points.

110

However, at high RF/microwave frequencies the length of the line is

comparable to the signal wavelength and the time delay of propagation
(and the corresponding signal phase shift) can no longer be ignored.
Under these conditions, the "distributed circuit model" is used to analyze
a transmission line. Such a model provides the governing differential
equations for voltage and current waves propagating along a transmission
line without a need to resort to Maxwell's equations to solve for the
electromagnetic field quantities.

7.6.4 The Governing Equations

At high frequencies, an infinitesimal length of a transmission line
can be modeled by two series elements (R, L) in conjunction with
two shunt elements (G, C) as shown in Figure 7.21a.

FIGURE 7.21a

An infinitesimal portion of a transmission line (TL).

Juxtaposing an infinite number of this infinitesimal model into a

long chain, will create a workable model for a transmission line as
shown in Figure 7.21b.

111

FIGURE 7.21b The equivalent circuit of a TL at high frequencies.

To develop the governing differential equations, we will examine

one x section of a transmission line as shown in Figure 7.21c.

FIGURE 7.21c Voltage and current in an infinitesimal length of TL.

Using KVL for the x section, we can write:

v(x,t)= i(x,t) Rx+ Lx i(x,t)/t +v(x+x,t)
Upon rearranging terms and dividing both sides by x, we obtain:
v ( x + x ) v ( x , t )
i ( x , t )
= Ri( x, t ) + L
x
t
Letting x0, yields:
v ( x , t )
i ( x , t )
(7.14a)
= Ri( x, t ) + L
x
t
Similarly, using KCL we can write:
v ( x + x , t )
i(x,t)= v(x+x,t) Gx+ Cx
+i(x+x,t)
t
Upon rearranging terms, dividing by x and letting x0, we have:

112

i ( x , t )
v ( x , t )
= Gv(x,t) + C
(7.14b)
x
t
Equations (7.14a) and (7.14b) are two cross-coupled equations in
terms of v and i. These two equations can be separated by first
differentiating both equations with respect to "x" and then properly
substituting for the terms, which leads to:
2 v( x, t )
i ( x , t )
2 i( x , t )
R
L

=
+
x
xt
x 2
v( x, t )
v ( x , t )
2 v( x, t )

+C
= R Gv( x, t ) + C
L G
t
t
t 2

Or,
2 v ( x, t )
2 v( x, t )
v ( x , t )
= LC
+ ( RC + LG )
+ RGv( x, t )
2
2
x
x
t
(7.14c)

Similarly for "i," we can write:

2 i ( x, t )
2 i( x, t )
i( x , t )
=
LC
+ ( RC + LG )
+ RGi( x, t )
2
2
x
x
t
(7.14d)
For sinusoidal signal variation for "v" and "i", we can write the
corresponding Phasors as follows:
v(x,t)=Re[V(x)ejt]
i(x,t)=Re[I(x)ejt]
Using phasor differentiation results from Chapter 3, Equations
(7.14c) and (7.14d) can be written as:
d 2 V( x) 2
V(x)=0
(7.14e)
dx 2
d 2 I( x ) 2
I(x)=0
(7.14f)
dx 2
Where
=+j=

(R + jL )(G + jC)

113

is the propagation constant, with real part () and imaginary part

(), called the attenuation constant (Np/m) and phase constant
(rad/m), respectively.
The solution to the second order differential equations as given by
Equations (7.14e) and (7.14f), can be observed to be of exponential
type format (ex ). Thus we can write the general solutions for V(x)
as follows:
V(x)=Vo+ e-x +Vo- ex

(7.14g)

Where the complex constants Vo+ and Vo- are determined from the
boundary conditions imposed by the source voltage and the load
value.
Similarly, I(x) can be obtained from V(x) (see Equation 7.14a) as:
+

1 dV ( x ) Vo e x Vo e x

I(x)=
=
(7.14h)
Zo
R + jL dx
Where,
R + jL
Zo=
(7.14i)
G + jC
is the characteristic impedance of the transmission line.

SPECIAL CASE: A LOSSLESS TRANSMISSION LINE

For this case, we have R=G=0. This yields the following
simplifications:
= j
LC =j,

L
C
Where = LC is the phase constant.

Zo=

In this case Equations 7.14e-f can be written as:

d 2 V(x)
+ 2V(x)=0
2
dx

(7.14j)

114

d 2 I( x )
+ 2I(x)=0
dx 2

(7.14k)

Similar to Equations 7.14g-h, the solutions to Equations 7.14j-k are

given by:
V(x)=Vo+ e-jx +Vo- ejx
(7.14l)
+ j x
j x
Vo e
V e
I(x)= o
(7.14m)
Zo

NOTE 1: Transmission line Equations 7.14e-f and 7.14j-k could

have all been derived using the Maxwell's equations directly from
the field quantities E and H as delineated in an appendix presented
at the end of the book entitled "Laws of electricity and Magnetism",
under items 14 and 19 .
It will be seen shortly in Chapter 7, that the term e-x [ or e-jx]
represents a propagating wave in "+x" direction while ex [ or ejx]
represents a propagating wave in "-x" direction on a transmission
line. The combination of the two waves propagating in opposite
directions to each other, forms a standing wave on the transmission
line (as discussed earlier in Chapter 2), which will be later explored
further in Chapter 7.
NOTE 2: Based on a given set of boundary conditions for the source
and the load, we can find the constants in the equations 7.14l-m. For
example, if the source voltage (at x=0) is known to be V=Vg and the
load voltage (at x= l ) is V=VL, then the constants Vo+ and Vo- can
easily be found from the following two equations:
x=0, Vg=Vo+ +Vo(7.14n)
+ -j
l
- j
l
x=l, VL=Vo e +Vo e
(7.14o)
_______________________________________________________
EXERCISE 7.1
a. Derive expressions for Vo+ and Vo- from equations 7.14n-o in
terms of Vg and VL.
b. Given the load value as Z=ZL, find Vo+ and Vo- in terms of Vg
and ZL [as in part (a)].

115

Hint: Use V2= ZL [Vo+ e-jl -Vo- ejl]/Zo

__________________________________________________________

7.6.5 Sinusoidal Wave Propagation on a Transmission Line

Consider a transmission line as shown in Figure 7.21d. Assuming a
sinusoidal signal excitation, the propagating voltage and current waves
on a transmission line are also sinusoidal and can be expressed as:
v(x,t)=Re[V(x)ejt]
(7.14p)
i(x,t)=Re[I(x)ejt]
(7.14q)
Where complex quantities V(x) and I(x) are phasor quantities.

FIGURE 7.21d Incident and reflected waves on a transmission line.

Using the distributed circuit model of a transmission line and its

corresponding equivalent circuit, the following differential equations for
I(x) and V (X) can be derived (see Chapter 5, example 5.5) for an
infinitesimal length of a transmission line:
d 2 V(x) 2
- V(x)= 0
(7.15a)
dx 2
d 2 I( x ) 2
- I(x) =0
(7.15b)
dx 2
Where is the complex propagation constants given by:
(7.16)
= +j = [(R+jL)(G+jC)]1/2
Where:
= attenuation constant (in Nepers/m)
= phase constant (in radian/m)
R = resistance per unit length in /m

116

L = inductance per unit length in H/m

G = conductance per unit length in S/m
C = capacitance per unit length in F/m
By observation, we notice that the general solution to the problem is of
exponential form, therefore we can write:
+
V
(7.17a)
V1(x)= Vo+e-x I1(x)= o e-x
ZO

Vo x
e
(7.17b)
ZO
Where Z0 is the characteristic impedance of the transmission line; Vo+
and Vo- are complex constants in general, whose values depend upon the
source and the transmission line characteristics, as will be seen shortly.

V2(x)= Vo-ex,

I2(x)= -

Since we are dealing with a linear system, the general solution for
voltage and current is obtained using the superposition theorem as
follows:
V(x)=V1(x)+V2(x)= Vo+e-x + Vo-ex
(7.17c)
+

V
V
I(x)=I1(x)+I2(x)= o e-x - o ex
(7.17d)
ZO
ZO
From Equation (7.17), we observe that voltage and current are a pair of
waves co-existing and are inseparable for a distributed circuit. Each
solution for voltage or current consists of two waves which will be
labeled as follows:
a. An incident wave: e-x =e-x e-jx
b. A reflected wave: ex =ex ejx
Where "x" is referred to as the electrical length.

(7.18a)
(7.18b)

As already mentioned, each wave travels at the phase velocity (VP) given
by:
VP= /
= c (air),
(7.19)
Where c is the speed of light in vacuum given by:
c = 1/(
oo)1/2 =2.9988 x 108 3 x 108 m/s.

117

The time-average incident power propagating along a transmission line is

given by (assuming ZO is a real number):

+2

V
1
P ( x ) = Re V + ( x)I + ( x) * = o e 2x
(7.20)
2
2Z O
The same can be written for the reflected power propagating back to the
source.
+

The law of conservation of energy requires that the rate of decrease of

propagating power P(x) along the line should equal the average power
loss per unit length (Ploss). Thus we can write:
P ( x )
= Ploss = 2P( x )
x
P
P / x [P( x + x ) P( x )] / x
= loss
=
2P( x )
2P
2P( x )

(7.21)

Equation (7.21) shows an interesting and yet very practical way to

measure the actual attenuation constant (). This method is particularly
helpful if one is trying to establish the integrity of a faulty line, since a
simple comparison of the measured () with the nominal () would
reveal the needed information. The following example elucidates this
point further.
_________________________________________________________
EXAMPLE 7.3
The microwave power at one point (P1) on a transmission line is
measured to be 10 mW. At a distance of d=50 cm away, another power
measurement (P2) indicates a power of 7 mW. Determine the attenuation
constant of the transmission line.
Solution:
P
P / x [P( x + x ) P( x )] / x
= loss
=
2P( x )
2P
2P( x )
Where
P/
x=(P2-P1)/d
Thus we have:

118

(P2 P1 ) / d (7 10 ) / 0.5
=
= 0.3 1/m
2P1
2x10
(Np/m)=ln(1/0.3)= 1.20 Np/m
__________________________________________________________
=

7.6.6 The Concept of the Reflection Coefficient

Any time an incident wave encounters a second medium different than
the first, it is partly reflected (creating a reflected wave) while the
remaining is transmitted through (creating a transmitted wave). The
reflected wave encountering the incident wave forms a standing wave as
described earlier. Thus we can see that there are four possible waves in a
transmission line:
a. An incident wave,
b. A reflected wave,
c. A transmitted wave, and
d. A standing wave.
Let us now define an important term:
DEFINITION- REFLECTION COEFFICIENT: Is defined to be the ratio of the
reflected wave phasor to the incident wave phasor.
In the special case of a uniform transmission line when the incident wave
encounters a second medium such as a termination (load) or a
discontinuity, then under these conditions, the ratio of the reflected wave
phasor to the incident wave phasor is "The reflection coefficient".
To illustrate this concept, consider a transmission line circuit with a load
(ZL) located at x=ll, as shown in Figure 7. 21d.
V+(x)= Vo e-x

(7.22a)

V-(x)= Vo ex

(7.22b)

At the load end (x =l), V+(l) is given by:

+
V+(l)= Vo e-l

119

However to find V-(l), we need to realize that the reflected wave reflects
from the load by a factor of L (i.e., L is the load reflection coefficient
at x=l):
V-(ll)= L Vo+e-l
(7.23d)
The reflected wave travels back a distance of x= l-x towards the source
as:
V-(x)= V-(l-x)= L Vo+e-l e-x = L Vo+e-l e-(l-x)
(7.23b)
where x is an imaginary reference frame set up at the load (x=0)
and is directed toward the source (x= l). Thus V-(x) can be written
as:
(7.23c)
V-(x)= L Vo+e-2l ex=Vo-ex
The reflection coefficient can now be defined as:

Vo 2 x
V ( x ) Vo e x
= + x = + e = L e 2 l e 2 x
(x)= +
V ( x) Vo e
Vo

(7.23e)

Thus the total voltage and current phasors [V(x), I(x)] along the
transmission line can now be written as:
V(x)= V+(x)+V-(x)= Vo+e-x + Vo-ex
Or,

V
V(x)= Vo+ (e-x + o + ex)= Vo+ (e-x +
L e-2l ex)
(7.24)
Vo
Similarly,
I(x)= I+(x)-I-(x)=V+(x)/Zo-V-(x)/Zo
Or,
+
V
I(x)= o (e-x -L e-2lex).
(7.25)
ZO
The input impedance, ZIN(x), at any point along the transmission line is
obtained through dividing equation 7.24 over 7.25 and is given by:
e x + L e 2 l e x
V( x )
Z IN ( x) =
= Z O x
(7.26)
I( x )
e L e 2 l e x

120

A SPECIAL CASE:
At the load end (where x = l), the following is observed:
1 + L
Zin(l)= ZL = Z O
1 L
Z ZO
L = L
(7.27a)
ZL + ZO
We can generalize equation (7.27a) for any arbitrary point along the
transmission line with an input impedance (Zin), and write the reflection
coefficient (
IN) at that point as:
Z IN Z O
IN =
(7.27b)
Z IN + Z O
Using Equation (7.27a) and letting the distance from the load as d=l-x,
Equation (7.26) can be written as:
Z + Z O tanh d
(7.28)
Z IN (d ) = Z O L
Z O + Z L tanh d

7.6.7 Lossless Transmission Lines

Since most of the transmission lines at RF/microwave frequencies have
negligible losses, we will focus exclusively on lossless transmission
lines.
In a lossless transmission line, there is no series resistance (R) or shunt
leakage conductance (G). Thus the energy propagating on the line does
not get attenuated in strength (or power). Considering Figure 7.22a, the
following simplifications can be made:
=0,
=j,
L
Zo=
(7.29a)
C
VP=/ =1/(LC)1/2
(7.29b)
=VP/f=2/
(7.29c)

121

FIGURE 7.22a A lossless Transmission line.

Using Equation (7.23e), we can write:

V
V ( x ) Vo e jx
= + jx = o + e j2x =
(x)= = +
L e-j2l ej2x
V ( x ) Vo e
Vo

(7.30a)

Using (7.24) and (7.25), we can write the voltage and current on a
lossless transmission line as:
V(x)= V+(x)+V-(x)= Vo+e-jx + Vo- ejx
(7.30b)

V
V(x)= Vo+e-jx 1 + o + e j2x

Vo

Or,
V(x)= Vo+e-jx[1+
(x)]
(7.31a)
Similarly, I(x) can be written as:
+

Vo -jx Vo jx
+
I(x)= I (x)-I (x)=
e e
ZO
ZO
+

V
I(x)= o e-jx[1-
(x)]
ZO

(7.31b)

The input impedance at any point (x) on the transmission line from
Equations 7.31 can now be written as:
1 + ( x )
ZIN(x)=ZO
(7.32)
1 ( x )
Using Equation (7.30) and letting the distance from the load as d=l-x,
Equation (7.32) can be written as:

122

Z IN (d) = Z O

Z L + jZ O tan d
Z O + jZ L tan d

(7.33)

The time-average incident (Pi or P+), reflected (Pr or P-) and transmitted
(Pt or PL) powers propagating along a transmission line are given by
(assuming ZO is a real number, =0):

+ 2

Vo
1
P + ( x ) = Re V + ( x)I + ( x ) * =
2
2Z O

(7.34a)
+ 2

Vo
1
2
P ( x ) = Re V ( x)I ( x) * =
,
2
2Z O

Pt ( x) = P + ( x) P ( x) = (1 )
2

Vo

(7.34b)

+ 2

2Z O

=PL,

Where PL is given by:

1
1
*
*
*
PL = Re(VL I L ) = Re( VL VL / Z L ) = VL
2
2

(7.34c)

Re(Z L )
2 ZL

(7.34d)

7.6.8 Determination of Vo+ and VoFrom the earlier discussion, we know that the incident wave is given by:
V+(x)= Vo+e-jx
(7.34e)
To find Vo+ we need to look at the source end (x =0), where we have:
V(0)=Vg -ZoI(0)
(7.34f)
Where,
V(0)=Vo+ejl [1+(0)]
+
Vo jl
e [1-(0)]
I(0)=
ZO
Upon substitution for V(0) and I(0) in Equation 7.34f, we have:
Vo+=(Vg/2)
(for x=0 at source)
(7.34g)
Equation (7.34g) is simply stating a voltage division of the source voltage
between the source impedance (Zg) and the characteristic impedance of

123

the line (Zo), where Zg=Zo. This can be easily visualized by noting that
the incident wave does not see the load at first but only Zo, thus the
voltage division!
Moreover, we know that the reflected wave is given by:
(7.34h)
V-(x)= Vo-ejx
To find Vo- we need to visualize the incident wave traveling toward the
load with a magnitude of Vg/2 given by:
V+(x)= (Vg/2)e-jx
At the load (x=l), the incident wave reflects back to the source by a
factor of L. This is called the reflected wave and is given by:
V+(l)= (Vg/2)e-jl
V-(l)= L (Vg/2) e-jl
The reflected wave travels back a distance of x= l-x and arrives at the
source as:
V-(ll)= L (Vg/2) e-jl e-jx= L (Vg/2) e-jl e-j(l-x)= (Vg/2) L e-j2l ejx
(7.34i)
where x is an imaginary reference frame set up at the load (x=0) and is
directed toward the source (x= l). Comparing 7.34i with 7.34h, we
obtain:
Vo- = (Vg/2) L e-j2l (for x=0 at source)
(7.34j)
Therefore using Equations (7.34g) and (7.34j) in Equation (7.30b), the
total voltage at each point (x) along the transmission line can be written
as:
V(x)= Vo+e-jx + Vo- ejx =(Vg/2) e-jx + (Vg/2)
L e-j2l ejx
Or,
(x)]
(7.34k)
V(x)= (Vg/2) e-jx[1 + L e-j2l ejx]= (Vg/2)e-jx[1+
It should be noted that Equation (7.34k) is the same as Equation (7.31a)
derived earlier.

7.6.9 A Summary of Analysis

Let us now recapitulate what we have developed for the incident and
reflected waves mathematically and write:

124

V+(x)= (Vg/2)e-jx =(Vg/2)

-
x
- j
x
-j2
L e l ejx=(Vg/2)
L -2
l+
x
V (x)= Vo e =(Vg/2)

(7.34m)
(7.34n)

At the source end (x=0) we have:

V+(0)= (Vg/2) =(Vg/2)
0
V (x)= (Vg/2)
L-2
l

(7.34o)
(7.34p)

At the load end (x=l) we have:

V+(x)= (Vg/2)e-jl =(Vg/2)
-
l
l
V-(x)= (Vg/2)L e-j2l ejx = (Vg/2)L-

(7.34q)
(7.34r)

What we have presented so far can be summarized in one diagram, which

reveals all of the complexities of transmission line analysis and makes
them into great simplicities. Figure 7.22b shows these simplicities clearly
and makes the analysis or design of any complex transmission line an
expedient task!

FIGURE 7.22b

A summary of transmission line key points.

125

7.6.10 Voltage Standing Wave Ratio (VSWR)

As described earlier, a standing wave results from two waves having the
same frequency traveling in opposite directions on a transmission line.
The meeting of these two waves produces a standing wave pattern of
voltage and current on a transmission line.
The maximum value of voltage anywhere along the transmission line is
given by:
L|)
(7.35)
Vmax=|V(x)|max=| Vo+|+| Vo-|=| Vo+|(1+|
The minimum value of the voltage is given by:
Vmin=|V(x)|min=| Vo+|-| Vo-|=| Vo+|(1-|
L|)
(7.36)
Similarly, for the current standing wave we have:
+

Imax=|I(x)|max=|I |+|I |=

Vo

ZO

ZO

ZO

Vo

(1 )
L

ZO

(7.37)

Imin=|I(x)|min=|I |-|I |=
+

(1 + )
+

Vo

Vo

Vo

ZO

Vo

ZO

(7.38)

Equations (7.35) to (7.38) are used to define the standing wave ratio
(often referred to as voltage standing wave ratio "VSWR") as follows:
V
I
1+ | L |
VSWR= max = min =
(7.39)
Vmin I max 1 | L |
Or,
VSWR 1
|
L|=
(7.40)
VSWR + 1
_____________________________________________________
EXAMPLE 7.4
What is the VSWR for a matched transmission line (Zo=50)?
Solution:

126

ZL=Zo L=0
VSWR=(1+0)/(1-0)=1
_____________________________________________________
EXAMPLE 7.5
What is the VSWR for:
a. An open load (ZL=),
b. A short load (ZL=0)? Assume Zo=50 .
Solution:
a) ZL= L=limZL(ZL-50)/(ZL+50)=1
VSWR=(1+1)/(1-1)=
b) ZL=0
L=(0-50)/(0+50)=-1 |L|=1
VSWR=(1+1)/(1-1)=
______________________________________________________
Conclusion: From examples 7.2 and 7.3, we can see that:
1
VSWR
(7.41)
_____________________________________________________
EXAMPLE 7.6
What is the ZIN of a TL at x=0 (d=l-x = l) for an open circuit load?
Solution:
For ZL=, (7.34) can be written as:
Z + jZ O tan l
Z OC = Z IN (l ) = lim Z L Z O L
Z O + jZ L tan l

ZOC = -jZo cot

l
(7.42)
_____________________________________________________
EXAMPLE 7.7
What is the ZIN of a TL at x=0 (d=l-x = l) for an short circuit load?
Solution:
For ZL=0, (7.34) can be written as:
0 + jZ O tan l
ZSC = Z IN (l ) = Z O
ZO + 0
ZSC = jZotan
l
(7.43)
_______________________________________________________

127

7.6.11 Quarter-Wave Transformers

The two main functions of any transmission line at any frequency, are
two-folded as follows:
a. Transmission of power, and/or
b. Transmission of information.
At RF/microwave frequencies, it becomes essential that all lines be
matched to each other, to the source and finally to the load. This is due to
the obvious fact that reflections due to mismatch or discontinuities (e.g.
at a connection, at a junction, etc.) will result in echoes and will reduce
the transmitted power and will distort the information carrying signal.
A simple method for matching a resistive load ZL to a lossless feed line
(having a real characteristic impedance ZO) is the use of a quarter-wave
transformer which is a piece of a transmission line having a /4 length
and a characteristic impedance of (ZO')/4.
The characteristic impedance of the quarter-wave transformer (ZO')/4
terminated in a real load ZL can be derived as follows:
DERIVATION OF (ZO')/4
x=0 d=ll-x = l
d=
/4 d=(2
/
)(
/4)= /2 tan
d=

Thus the input impedance of the quarter-wave transformer (Equation

7.33b) terminated in a real load ZL can be written as:
d=
/4
'
'
tan l
' Z + jZ O tan l
' jZ
Z in ( / 4) = lim l / 4 ( Z O ) L '
= (Z O ) O
jZ L tan l
Z O + jZ L tan l
2

Z'
Z in = O
ZL
Or,
'
( Z O ) / 4 = Z in Z L

(7.44)

(7.45)

POINT OF CAUTION: This simple method of matching is applicable on

when both of the following conditions are met:

128

a) The feed transmission line is lossless (this leads to a characteristic

impedance value which is a real number), and
b) The load is resistive
There are cases where the load is a complex number, and thus at first
glance a quarter-wave transformer does not seem to lend itself for
matching purposes.
However, in such a case the load should first be converted into a real
number by adding a reactance having the same value as the load's
reactance but with the opposite sign. The resultant load is resistive and
can then be transformed to the feed line's characteristic impedance
through the use of a quarter-wave transformer as described above (For
further details on matching techniques, please see Chapter 10). The
following example further elucidates this-simple method of matching.
__________________________________________________________
EXAMPLE 7.8
What is (ZT )/4 of a quarter-wave transformer to transform a load of 100
to a 50 feed line as shown in Figure7.23?
Solution:
To create a match, we require that Zin to be the same as the characteristic
impedance of the feed line, i.e., Zin=50 . Using (7.45), we can write:
(ZT)/4=(ZLZin)1/2
Thus we obtain:
(ZT )/4=(100 x 50)1/2=70.7

__________________________________________________________

129

NOTE: Example 7.8 clearly shows why these types of

shorted transformers are ideal for electrically isolating the
RF circuitry from the DC bias source in an amplifier circuit
as will be discussed later in Chapter 15, RF/Microwave
Amplifiers. This is because the RF circuitry is connected at the
input side of the transformer while DC bias source is at the
short-circuited end of the transformer (of course, the short
circuit is created by the use of a high-value capacitor to
ground). In this fashion, the RF signals see an open circuit at
the RF side and would not be able to travel to the DC bias
source, while the DC bias sees a direct connection (i.e. a
short circuit) to the RF circuitry.

7.6.12 A Generalized Lossless Transmission Line Circuit

In the previous examples, the main focus has been on the effects of a
load on the current and voltage waves traveling on a transmission line.
However the source of the waves, which is the generator located at the
other end, plays an important role in the propagation of the waves along
the transmission line.
Up to this point in our discussion, the generator's internal impedance has
been a real number equal to the characteristic of the transmission line. In
effect, the generator was matched to the line and only the effects of the
mismatch of the load was studied so far. Obviously, this is a special case.
The most general case is having mismatches at both ends (i.e., at the
generator and at the load ends), which will now be discussed in detail.

7.6.13 Analysis
Consider a finite lossless transmission line (T.L.) of length (l) with a
characteristic impedance (ZO) driven by a generator (Vg) with an internal
impedance (Zg) at x=0 and terminated in a load (ZL) at x=l as shown in
Figure 7.24.
The boundary condition (B.C.) at each end can be written as:
a. B.C. #1-Voltage and current at x=0 is given by:
Vi=Vg-ZgIi
b. B.C. #2-Voltage and current at x=l is given by:
VL=ZLIL

130

FIGURE 7.24

A general transmission line circuit.

c. Voltage and current on the T.L. for 0 x l, from (7.31) and (7.32) is
given by:
V(x)=V+(x)+V-(x)
(x)]
(7.46a)
V(x)= |Vo+|e-jx[1+
+

I(x)= I (x)-I (x)=

Vo

e-jx[1-(x)]

(7.46b)
ZO
Where, from (7.30), (x) is given by:

V
V ( x ) Vo e jx
= + jx = o + e j2x =
(7.46c)
(x)= +
L ej2(x-l)
V ( x ) Vo e
Vo
Applying the boundary condition given by (a), we can solve for |V+| as
follows:
+
V
Vi=V(0)=Vo +e-jx[1+
(0)]= Vg- Zg o [1-
(0)]
(7.47)
ZO
Where
(0)=
L e-j2l

131

From Equation 7.47, we can solve for Vo + in terms of Vg to obtain:

Z O Vg
e j l
+

Vo =
(7.48a)

j
2

Z O + Z g 1 L g e

where
Zg ZO
g =
(7.48b)
Zg + ZO
Thus V(x) and I(x) under this general condition may be obtained by
substituting for Vo + from Equation (7.48) in (7.46) as follows:
Z O Vg jx 1 + L e j2 ( x l )

V(x) =
e
(7.49a)
1 e j2l
ZO + Zg
L
g

j
2

(
x

l
)
1 L e

Vg

I( x ) =
e jx
(7.49b)
1 e j 2 l
ZO + Zg
L g

NOTE: Equations 7.49a, b show phasor expressions for the

voltage and current due to a sinusoidal voltage source (Vg)
feeding a finite transmission line, which is terminated in a
general load(ZL).

These equations represent the summation of infinite number

of reflections from both ends of the transmission line, i.e.,

V ( x ) =V1 + +V1 - +V2 + +V2 - +...= (Vi + +Vi- )

(7.50a)

i=1

where
|V1 + |= |Vo + |,
|V1 - |=|L| |Vo + |,
|V2 + |=| g|| L| |Vo + |,
|V2 - |=|g||L|2 |Vo + |, etc.
and,
Vo + = V g [Z o /(Z o +Z g )]
___________________________________________________
EXERCISE 7.1
Prove that the summation of infinite number of voltage

132

reflections as shown by Equation 7.50a converges to Equation

7.49a.
HINT: Note that:

xn=1+x+x2++xn+=1/(1-x),

|x|<1

(7.50b)

n =0

_______________________________________________________________

SPECIAL CASES
From Equation (7.49) we can derive several special useful cases as
shown in Figures 7.25a, b,c.

Case I. Matched at Both Ends

Zg=ZL=ZO (see Figure 7.25a)
L=0
g=
Vo+=(Vg/2)
(for x=0 at source)

FIGURE 7.25a

Matched at both ends.

Since L=0, there is no reflected wave from the load, and thus Equation
7.49 can be written as:
V(x) =
I( x ) =

Vg
2
Vg

2Z O

e j x

(7.51a)

e j x

(7.51b)

At the source end (x=0), we have:

133

Vg

V (0) = Vi =

I ( 0) = I i =

2
Vg

2Z O

(7.51c)
(7.51d)

and at the load end (x=l), we can write:

Vg j l
V(l) = VL =
e
(7.51e)
2
Vg jl
I(l) = I L =
e
(7.51f)
2Z O
In this case there are no standing waves on the transmission line and
magnitude of the voltage and current is the same everywhere on the line,
that is,
Vg
|Vi|=|VL|=|V(x)|=
(7.51g)
2
Vg
|Ii|=|IL|=|I(x)|=
(7.51h)
2Z O
NOTE: In some texts, the reference for length is located at the load end
rather than the generator end. This axis is designated by the x-axis in
figure 7.24. This means that there is a shift in the x-axis (by +l), i.e.,
x=x+l
(7.51i)
Thus Equations (7.51a,b) can now be written in terms x' as:
V(x)= (Vg/2)e-j(x+l) (for x=0 at load)
Or,
V ( x' ) =
I( x' ) =

Vg
2
Vg

e j l e j x'

2Z O

e j l e j x '

Case II. Matched at the Source End

Zg= ZO, ZL ZO
g =0 (see Figure 7.25b)

(7.51j)
(7.51k)

134

FIGURE 7.25b

Matched at source end only.

Since g=0, there is no reflected wave from the source, but there is a
reflected wave from the load. Thus Equation 7.49 can be written as:
Vg j x
V(x) =
e
1 + L e 2 j ( x l )
(7.52a)
2
Vg jx
I( x ) =
e
1 L e j2 ( x l )
(7.52b)
2Z O

In terms of the shifted axis (x=x+l), we can write Equations (7.52) as:
Vg j ( x'+ l )
V( x' ) =
e
1 + L e j2x'
(7.52c)
2
Vg j ( x' + l )
I( x' ) =
e
1 L e j2x'
(7.52d)
2Z O

At the generator end (x=-l), we have:

Vg
V ( l) = Vi =
1 + L e j2 l
2
Vg
I( l) = I i =
1 L e j2l
2Z O

(7.52e)

(7.52f)

At the load end (x=0) we have:

Vg jl
V (0) = VL =
e (1 + L )
2
Vg jl
I ( 0) = I L =
e (1 L )
2Z O

(7.52g)
(7.52h)

135

It should be noted that Equation (7.52g) is the same as Equation (7.34k)

when x=l.

Case III. Matched at the Load End

Zg ZO, ZL=ZO (see Figure 7.25c)

L =0

(c)
FIGURE 7.25c

Matched at load end only.

Using Equation 7.49, we can write:

Z O Vg jx
V(x) =
e
ZO + Zg
Vg

(7.53a)

e j x
(7.53b)
ZO + Zg
At the generator end (x=0), we have:
Z O Vg
V(0) = Vi =
(7.53c)
ZO + Zg
This result could have easily be written using the voltage division
principle, without using Equation (7.49). In other words, this case
reduces to a simple voltage division between the source internal
impedance (Zg) and the transmission line presenting a constant input
impedance of (Zo).
Vg
I ( 0) = I i =
(7.53d)
ZO + Zg
I( x ) =

And at the load end (x=l) we have:

136

V(0) = VL =

I ( 0) = I L =

Z O Vg
ZO + Zg

Vg

e j l

(7.53e)

e j l

(7.53f)
ZO + Zg
The voltage and current on the transmission line have the same
magnitude except for a phase shift with length:
VL = Vi e jl for x=ll
(7.53g)
Or in general,
(7.53h)
V ( x ) = Vi e jx
Similarly, we can write for current:
I L = I i e jl for x=l
(7.53i)
Or in general,
I ( x ) = I i e j x
(7.53j)
Equations (7.53a) and (7.53b) can be written in terms of the shifted axis
(x=x+l) as:
V( x' ) = Vi e j ( x'+ l )
(7.53k)
I( x' ) = I i e j ( x'+ l )
(7.53l)
__________________________________________________________
EXAMPLE 7.9
Consider a 50 lossless transmission line of length l=1 m, connected to
a generator operating at f=1 GHz and having Vg=10 V and Zg=50 at
one end and to a load ZL=100 at the other (see Figure 7.26).

FIGURE 7.26 Circuit for Example 7.10.

137

Determine:
a. The Voltage and current at any point on the transmission line.
b. The voltage at the generator (Vi) and load (VL) ends.
c. The reflection coefficient and VSWR at any point on the line.
d. The average power delivered to the load.
Solution:
a. Since Zg=ZO=50 g=0
Since g=0, special case (b) above applies here. Thus we can write:
=/c=2x109/3x108=20/3
Z Z O 100 50 1
L = L
=
=
Z L + Z O 100 + 50 3
V g j x
1

V ( x) =
e
1 + L e j 2 ( x l ) = 5e j 20x / 3 1 + e j 40 ( x 1) / 3
2
3

V g j x
1

I ( x) =
e (1 L e j 2 ( x l ) ) = 0.1e j 20x / 3 1 e j 40 ( x 1) / 3
2Z O
3

b. At the generator end (x=0 m), we have:

1

Vi = V (1) = 51 + e j 40 / 3 = 4.16 + j1.44 V

3

At the load end (x=1 m), we have:

1 20 j 20 / 3
VL = V (1) = 5e j 20 / 3 1 + =
e
3 3
c. The reflection coefficient and VSWR are as follows:
1
(x) =L ej2x= e j40 x / 3
3
1 + L 1 + 1 / 3
VSWR =
=
=2
1 L 1 1 / 3
d. The average power delivered to the load is:
=0

2
20e j 20 / 3 / 3
VL
1
2
*
P( x) = Re V L ( x) I L ( x) =
=
= = 0.22 W
2
2Z L
2 x100
9

138

NOTE: If the load was completely matched to the line the power
delivered to the load would have been:
ZL=50
|Vi|=|VL|=Vg/2=5 V

(Pav )max =

VL

2Z L

52
=
= 0.25 W
2x50

Since there is no reflected power, (pav)max is also the incident power (Pi)
which is higher than the (Pav) calculated earlier under unmatched
conditions. The difference in the two powers is due to the reflected
power back to the source:
Pr=|L|2Pi=(1/9)(0.25)=0.03 W
__________________________________________________________

7.7 MICROSTRIP LINE

Amongst all planar transmission lines, microstrip line has gained much
popularity and importance in microwave planar circuit technology, and
thus will be considered and analyzed in this section.
A microstrip line is a transmission line consisting of a strip of conductor
of thickness (t), width (w) and a ground plane separated by a dielectric
medium of thickness (h) as shown in Figure 7. 27.
Since it is an open conduit for wave transmission, not all of the electric or
magnetic fields will be confined in the structure. This fact along with the
existence of a small axial E-field, leads to a not-purely TEM wave
propagation, but a quasi-TEM mode of propagation.
These types of transmission lines are very popular and are used
extensively in microwave planar circuit design and microwave integrated
circuit (MIC) technology. Use of printed circuit board technology and its
simplicity of fabrication, along with ease of placement and
interconnection of lumped elements and components has made this type
of transmission line very popular and much superior to other types of
planar transmission lines

139

7.7.1 Wave Propagation in Microstrip Lines

The dielectrics used in the fabrication of the Microstrip line are
characterized by a dielectric constant (r) defined by:
r=/o,
(7.54a)
-12
o=8.854 x 10 F/m
where and o are the dielectric's and vacuum's permittivity,
respectively. Some of the most popular dielectrics are: Duroid (r=2.23,
6, 10.5), alumina (r=9.5-10), Quartz (r=3.7), silicon (r=11.9), etc.
The EM-wave propagation in a microstrip line is approximately nondispersive below a cut-off frequency (fo), which is given by:
ZO
f O (GHz ) = 0.3
(7.54b)
h r 1
Where h is in centimeters.

140

The phase velocity for a quasi-TEM is given by:

VP=c/
ff
Where c is the speed of light and ff is the effective relative dielectric
constant.
Since the field lines are not contained in the structure and some exist in
the air (see Figure 7.27b), the effective dielectric constant satisfies the
following relation:
1< ff <r
In general, the effective dielectric constant (ff ) is a function of not only
the substrate material (r) but also of the dielectric thickness (h) and
conductor width (W).
The characteristic impedance (Zo) is given by:
1
ZO =
VP C O
Where Co is the capacitance per unit length.

(7.55)

The wavelength () of a propagating wave in the microstrip line is given

by:
= VP/f=
o/
ff
(7.56a)
where o=c/f is the wavelength in free space.
NOTE: The wavelength of a TEM wave (TEM) propagating in the
dielectric material is different than the wavelength (o) of a propagating
wave in free space as follows:
TEM=
o/
r
(7.56b)

As can be seen from these equations the characteristic impedance (Zo)

and the wavelength () both are functions of the geometry (w, h) of the
microstrip line. This variation is shown in Figures 7.28 and 7.29.

141

142

7.7.2 Empirical Formulas

The essential empirical formulas for a microstrip line can be categorized
as follows (assuming zero or negligible thickness of the strip of metal on
top of the dielectric, i.e., t/h<0.005):
a. ff FORMULA
The effective dielectric constant (ff) is given by [assuming that the
dimensions of the microstrip line (W, h) are known]:
For W/h 1:
1 / 2
2
r + 1 r 1
h
W

ff =
+
+ 0.04 1
,
1 + 12
2
2
W
h

(7.57)

For W/h 1:
1 / 2
+ 1 r 1
h
(7.58)
1
12
ff = r
+
+

2
2
W
The effective dielectric constant (ff) can be thought of as the dielectric
constant of a homogeneous medium that would fill the entire space,
replacing air and dielectric regions.
b. Zo FORMULA
The Characteristic impedance is given by [assuming that the dimensions
of the microstrip line (W, h) are given or known]:
For W/h 1:
60 8h W
ZO =
ln
+

ff W 4h

For W/h 1:
ZO =

(7.59)

120

ff [W / h + 1.393 + 0.667 ln( W / h + 1.444)]

(7.60)

c. W/h FORMULA
Assuming (ff) and Zo are given, then the microstrip dimensions
(W/h) can be found as follows (a design problem):

143

For W/h 2:
W
8e A
=
h e 2A 2

(7.61)

For W/h 2:

1
W 2
0.61
= B 1 ln( 2B 1) + r
ln( B 1) + 0.39

h
2 r
r

(7.62)
Where

A=

ZO
60

r + 1 r 1
0.11

0.23 +
+
2
r + 1
r

(7.63)

and

B=

377

(7.64)

2Z O r

d. FORMULA
The wavelength in the microstrip line () is given by:
For W/h < 0.6:

r
= O
0.0297
r 1 + 0.6( r 1)( W / h )

1/ 2

(7.65)

For W/h 0.6:

r
= O
0.1255
r 1 + 0.63( r 1)( W / h )

1/ 2

(7.66)

e. FORMULAS (Attenuation factors)

Another characteristic of the microstrip line is its attenuation when
signals travel on it. There are two types of losses in a microstrip line:

144

Dielectric substrate loss due to dielectric conductivity

Conductor Ohmic loss due to skin effect
The loss factor () can be found by noting that the power carried
along a transmission line in (+x direction) in a quasi-TEM mode can
be written as:

] [

1
V + (x)
P ( x ) = V + ( x )I + ( x )* =
2
ZO
Where
V+(x)=|V+|e -x e -jx
Thus we have:
+

P + (x) =

V+

(7.67)
(7.68)

2Z O

e 2 x e j 2 x = P + e j 2 x

(7.69)

Where
+

P =

V+

e 2 x

2Z O
and is the total attenuation factor which is composed of two
components:
=
d +
c
(7.70)
where
d =Dielectric loss factor, and
c= Conductor loss factor
These two loss factors are discussed next:
e1. d FORMULA
Attenuation Due to Dielectric Loss
Attenuation due to dielectric loss identified by "dielectric loss factor
(d)" using the quasi-TEM mode of propagation, is given by:
For Low-loss dielectric

d = 27.3

tan r ff 1

O r 1 ff

(dB/cm)

Where (tan) is the loss tangent given by:

(7.71)

145

tan =

For high-loss dielectric

1 ff 1

d = 4.34 O
(dB/cm)
(7.72)
O r 1 ff
Where is the conductivity of the dielectric and o=4 x 10-7 (H/m)
is the permittivity of the free space.
1/ 2

e2. c FORMULA
Attenuation Due to Conductor Loss
Attenuation due to dielectric loss identified by "conductor loss factor
(c)" using the quasi-TEM mode of propagation (for W/h ), is
given approximately by:
W/h ,
RS
c =
(Np/m)
(7.73)
ZO W
Where

f o
(7.74)

is the surface resistivity of the conductor. Usually conductor loss is

more dominant than the dielectric loss in most microstrip lines, i.e.,
c>>d; =c+dc
RS =

However, there are some cases (such as in silicon substrates) where

the dielectric loss factor (d) is of the same order or larger than the
conductor loss Factor (c).
_______________________________________________________
EXAMPLE 7.10
A 50 microstrip transmission line needs to be designed using a
sheet of Epsilam-10 (r=10) with h=1.02 mm. Determine W, and
ff.by:
a. An exact method

146

b. An approximate method
Solution:
a. Exact method
we will design a microstrip line with W/h2. Thus from (7.61) we
have:
W
8e A
= 2A
h e 2
Where
Z + 1 r 1
0.11 50 10 + 1 10 1
0.11
=
0.23 +
A= O r
+
+
0.23 +

60
2
r + 1
r 60
2
10 + 1
10
A=2.152
Thus (W/h) is obtained to be:
W
= 0.96
h
and
W=1.02x0.96=0.98 mm
Since W/h>0.6, we use (7.66) to find and then use (7.56a) to find
ff as follows:
1/ 2

r
= O
0.1255
r 1 + 0.63(r 1)(W/ h)

1/ 2

r
= O
0.1255
10 1 + 0.63(101)(0.96)

=0.39o
and
=0.39o=
o/
ff
ff=(1/0.39)2=6.6
b. Approximate method
Using Figure 7.25, we obtain W/h for ZO=50 to be:
ZO=50 W/h1 W=h=1.02 mm
From Figure 7.26 for W/h=1, we obtain:
/TEM=1.23
From (7.56b) we have;
TEM=
o/
r=
o/
10=0.316
o
Thus is found to be:
=1.23x0.316
o=0.39
o
and from (7.56a) we have:

147

=
o/
ff ff=(
o/
)2
2
ff=(1/0.39) =6.6
_______________________________________________________
EXAMPLE 7.11
Design a 50 transmission line that provides 90 phase shift at 2.5
GHz. Assume h=1.27 mm and r=2.2.
Solution:
To find "W", we assume that W/h2 and will verify this assumption
later. From Equation (7.62) and (7.64), we find:
377
B=
2Z O r
B=7.985,
And
1
W 2
0.61
= B 1 ln( 2B 1) + r
ln( B 1) + 0.39

2 r
r
Yielding:
W/h=3.08 W=3.08x1.27 =3.91 mm
The value of W/h=3.08 is obviously greater than 2, which justifies
our earlier assumption.
So far we have found the width of the line, now we need to know the
length of the line. Using the given phase shift of 90 yields:
=l=l/Vp=2fl/(c/ff)=2flff/c=90=/2
l=c/(4fff)
From the above equation we can see that in order to find l, we need
to find ff. Using Equation (7.58), we obtain:
For W/h 1:
1 / 2
r + 1 r 1
h
ff =
+
1 + 12
2
2
W
ff=1.87
Thus the length of the transmission line is given by:

148

l=3x108/(4x2.5x109x1.87)=0.0219 m=2.19 cm
_______________________________________________________

Chapter 7- Symbol List

A symbol will not be repeated again, once it has been identified and
defined in an earlier chapter, with its definition remaining
unchanged.
C0 - Capacitance per unit length
EM Electro-Magnetic
k Arbitrary constant
TEM Transverse Electro-Magnetic
TL Transmission Line
v Velocity of motion
VP - Phase velocity
V+ - Incident voltage
V- - Reflected voltage
Z0 Characteristic Impedance
ZOC - Open circuit impedance
ZSC - Short circuit impedance
Z/4 - Impedance at the location /4.
- Phase constant
- Reflection coefficient
L - Reflection coefficient at the load
(x) - Reflection coefficient at location x
Dielectric permittivity
ff Effective relative dielectric constant
o Permittivity of vacuum (8.85x10-12 F/m)
r Dielectric constant of a material
- Propagation constant
o - Wavelength in free space
- Angular frequency (=v)

CHAPTER -7 PROBLEMS

149

7.1)

In the two-port network shown in Figure P7.1, assume that

(VS)RMS=200 V, and ZL=50+j50 .
a) Find V+(0), V+(/8), V-(0), V-(/8).
b) Calculate net voltages: V(0), V(/8), I(0) and I(/8).
c) Calculate the input powers at x=0, /8 and show:
P(0)=P(/8).
d) Find ZIN(0)

Figure P7.1

7.2)

Find the input impedance and the reflected power at Port(1)

and the power delivered to the load at port(2) for the circuit
shown in Figure P7.2. Assume Vs=cos 2x109t Volts.

7.3)

In the lossless transmission line circuit shown in Figure P7.3,

calculate
the incident power, the reflected power and the power
transmitted into the 75 line. Show that: PINC=PREF+PTRANS
A lossless transmission line (l=0.6) is terminated in a load
impedance (ZL=40+j20 ). Find the reflection coefficient at
the load, the input impedance of the line and the VSWR on
the line.

Figure P7.2

7.4)

150

Figure P7.3

7.5) In the circuit shown in Figure P7.5, calculate the reflection

coefficient at the load, the VSWR on the line and the power to load.

Figure P7.5

7.6)

Consider the lossless transmission line circuit shown in

Figure P7.6. Calculate:
a) The load impedance (ZL)?
b) The reflection coefficient at the input of the line.
c) The VSWR on the line.

7.7)

A lossless transmission line is terminated with a 200 load

if the VSWR on the line is 2.0, find the possible values for
the line's characteristic impedance.

Figure P7.6

151

7.8)

For a lossless transmission line, terminated in a reactive load

(ZL=jX), find the reflection coefficient and the VSWR. What
is ||?

REFERENCES
[7.1] Cheng, D. K. Fundamentals of Engineering
Electromagnetics. Reading: Addison Wesley, 1993.
[7.2] Cheung, W. S. and F. H. Levien. Microwave Made Simple,
Principles and Applications. Dedham: Artech House,
1985.
[7.3] Collin, R. E. Foundation For Microwave Engineering, 2nd
Ed., New York: McGraw-Hill, 1992.
[7.4] Edwards, T. C. Foundations for Microstrip Circuit
Design. New York: John Wiley & Sons, 1981.
[7.5] Gardiol, F. Microstrip Circuits. New York: John Wiley &
Sons, 1994.
[7.6] Gonzalez, G. Microwave Transistor Amplifiers,
Analysis and Design, 2 nd ed. Upper Saddle River:
Prentice Hall, 1997.
[7.7] Kraus, J. D. Electrom agnetics, 3rd Ed., New York:
McGraw-Hill, 1984.
[7.8] Plonsey, R. and R. E. Collin. Principles and
Applications of Electromagnetic Fields, 2nd Ed., New
York: McGraw-Hill, 1982.
[7.9] Radmanesh, M. M. and B. W. Arnold, Generalized
Microstrip-Slotline Transitions, Theory and Simulation
Vs. Experiment, Microwave Journal, Vol. 36, No. 6, pp.
8895, June 1993.
[7.10] Radmanesh, M. M. and B. W. Arnold, Microstrip-Slotline
Transitions: Simulation Versus Experiment, EESof Users
Group, IEEE MTT-S International Microwave
Symposium, Albuquerque, New Mexico, June 1992.
[7.11] Radmanesh, M. M. The Gateway to Understanding:
Electrons to Waves and Beyond, AuthorHouse, 2005.
[7.12] Radmanesh, M. M. Cracking the Code of Our Physical
Universe, AuthorHouse, 2006.

152

CHAPTER 8

Two-Port Network
Representations
8.1 INTRODUCTION
RF/microwaves devices, circuits and components can be classified
as one-, two-, three- or N-port networks. However, a majority of
circuits under analysis are two-port networks. Therefore, we will
focus primarily on two-port network characterization and will study
its representation in terms of a set of parameters that can be cast into
a matrix format.
DEFINITION- A TWO PORT NETWORK: is a network which has only
two access ports, one for input or excitation and one for output or
response.
The description of two-port networks from a circuit viewpoint can
best be achieved both at low and high frequencies through the use of
network parameters. These parameters are discussed in the
upcoming sections.

8.2 LOW-FREQUENCY PARAMETERS

To characterize a linear network at low frequencies, several different
sets of parameters are available, where one may be selected to fit the
application to obtain the most optimum results.

153

Voltages and currents at each port provide us with four variables of

interest: v1, v2, i1 and i2. There are six ways of picking two out of a
set of four variables, but only four combinations (or sets) will yield
non-trivial and unique parameters. These are called Z-, Y-, h-, and
ABCD-parameters. A two-port network with four voltage and
current parameters is shown in Figure 8.1.

These parameters are defined next.

A. IMPEDANCE OR Z-PARAMETERS

v1=z11i1 + z12i2
v2=z21i1 + z22i2
Or, in matrix form:
[V]=[Z][I]
v
[V ] = 1
v 2

[I ] = 1
i

Z 11

Z 21

B.

(8.2)
(8.3a)
(8.3b)

i 2

[Z] =

(8.1a)
(8.1b)

Z 12
Z 22

(8.4)

ADMITTANCE OR Y-PARAMETERS
Similarly, we can write the Y-parameters in matrix form as:
[I]=[Y][V]
(8.5)
where[I] and [V] are defined as before and [Y] as follows:

154

[Y] =

Y11

Y21

C.

Y12
Y22

(8.6)

HYBRID OR H-PARAMETERS

v 1 h 11
i = h
2 21

h 12 i 1
.
h 22 v 2

(8.7)

D.

TRANSMISSION OR ABCD PARAMETERS

v 1 A B v 2
(8.8)
i =

.
1 C D i 2
_______________________________________________________
EXAMPLE 8.1
Find the [ABCD] matrix for a series impedance element (Z) as
shown in Figure 8.2.

Solution:
Using KVL and KCL, the following can be written:
v1=v2-Zi2=Av2-Bi2
i1=-i2=0-i2=Cv2-Di2
Thus the [ABCD] matrix is given by:
A B 1 Z
C D = 0 1

______________________________________________________
EXAMPLE 8.2
Find the [ABCD] matrix for a shunt element (Y) as shown in Figure
8.3.

155

Solution:
Using KVL and KCL, the following can be written:
v1=v2=v2+0=Av2 - Bi2
i1=Yv2-i2 =Cv2 - Di2
Thus the [ABCD] matrix is given by:
A B 1 0
C D = Y 1

_______________________________________________________
EXAMPLE 8.3
Find the [ABCD] matrix for a circuit consisting of a series element
(Z) and a shunt element (Y) as shown in Figure 8.4.

Solution:
The [ABCD] matrix for the whole circuit which is a cascade of a
series and
a shunt element, is a multiplication of the two matrices as follows:
A B A1 B1 A 2 B 2
C D = C D . C D

1
1 2
2
1 Z 1 0
=
.

0 1 Y 0
Thus the [ABCD] matrix is given by:
A B 1 + ZY Z
C D = Y
1

156

_____________________________________________________
Example 8.4
Find the [ABCD] matrix for a transformer as shown in Figure 8.5.

Solution:
Using the transformer voltage and current rule, which states that if
the voltage is stepped down then in order to preserve the power flow
the current must be proportionately stepped up, we have:
v1=nv2=Av2-Bi2
1
i1= - i2 =Cv2-Di2
n
Thus the [ABCD] matrix is given by:
A B n 0
C D = 0 1

n
_______________________________________________________
Example 8.5
Find the [ABCD] matrix for a lossless transmission line of length (l)
and characteristic impedance (ZO) as shown in Figure 8.6.

Solution:

157

Using results obtained in Chapter 7, we know that the output voltage

or current have the same magnitude as the input but lag behind in
phase by e-jl. Thus the following can be written:
v2=v1 e-jl v1=v2 e+jl = v2 cosl+jv2 sinl
i2= -i1 e-jl i1= -i2 e+jl = -i2 cosl-ji2 sinl
Since the load end is considered to be matched to the transmission
line, we can write:
v2= -Zoi2
v1= (cosl)v2 - (jZo sinl)i2= Av2 -Bi2
i1= (jYosinl)v2 -( cosl) i2= Cv2 -Di2
Thus the [ABCD] matrix can be written as:
jZ o sin l
A B cos l
C D = jY sin l
cos l

o
_______________________________________________________

8.3 HIGH-FREQUENCY PARAMETERS

We note that Z-, Y-, h- and ABCD-parameters are based upon the
following considerations at each of the network ports:
Net voltage (v) and net current (i)
Short and open circuit terminations
Simple observations at high RF/microwave frequencies reveal that:
a. Shorts and open circuit terminations are difficult to implement
over a broad range of frequencies and thus cannot be used to
characterize networks, and
b. At high RF/microwave frequencies, the net voltage (or net
current) are a combination of two or more voltage (or current)
traveling waves.
Based on these observations, the Z-, Y-, h- and ABCD-parameters
cannot be accurately measured at these higher frequencies and
therefore we have to use the concept of propagating or traveling
waves to define the network parameters.

158

The network representation of a two-port network at high

RF/microwave frequencies is called "scattering parameters" (or "Sparameters" for short).
When cascading networks, a variation of S-parameters called chain
scattering parameters (or T-parameters) are used to simplify the
analysis.
These two types of high frequency parameters are very popular and
are primarily used at the high RF/microwave frequencies.

8.4 FORMULATION OF THE S-PARAMETERS

The high frequency S- and T-parameters are used to characterize
high RF/microwave two-port networks (or N-Port networks, in
general). These parameters are based on the concept of traveling
waves and provide a complete characterization of any two-port
network under analysis or test at high RF/microwave frequencies.
In view of the linearity of the Electromagnetic field equations and
the linearity displayed by most microwave components and
networks, the "scattered waves" (i.e. the reflected and transmitted
wave amplitudes) are linearly related to the incident wave amplitude.
The matrix describing this linear relationship is called the "scattering
matrix," or [S].
While the lower frequency network parameters (such as Z-or Ymatrices, etc.) are defined in terms of net (or total) voltage and
currents at the ports, these concepts are not practical at high
RF/microwaves frequencies where it is found that any set of
parameters to be meaningful, must be defined in terms of a
combination of traveling waves.
To characterize a two-Port network, which has identical
characteristic impedances at both the input and output ports, let us
consider the incident and reflected voltage waves at each port as
shown in Figure 8.7.

159

To accurately define the S-parameters, we will consider a voltage

[Vi+] incident on and a voltage [Vi-] reflected from the terminals of a
two-port network (i=1,2) as shown in Figure 8.7.

The scattering matrix, [S], is now defined to describe the linear

relationship between the incident voltage wave phasor matrix [Vi+]
and the reflected or transmitted wave phasor matrix [Vi-] at any of
the two ports as follows:
V1- =S11V1+ + S12V2+
V2- =S21V1+ + S22V2+
Or, in matrix form we can write:
V1 S 11 S 12 V1+
=
. +
V2 S 21 S 22 V2
Or,
[V-]=[S][V+]
Where,
V1

[V ] = ,
V2
V +
[V + ] = 1 +
V2
and,
S
S
[S] = 11 12
S 21 S 22

(8.9)

(8.10)

This linear relationship is expressed in terms of a ratio of two

phasors which are complex numbers with the magnitude of the ratio

160

always less than or equal to 1. Each specific element of the [S]

matrix is defined as:
S11=

V1
V1

| V + =0 =in (Input reflection coefficient when Output

2

port is terminated in a matched load.) (8.11)

S21=

V2
V1

|V + =0

(Forward transmission coefficient when

output port is terminated in a matched load.) (8.12)

S12=

V1
V2

|V + =0

(Reverse transmission coefficient when

input port is terminated in a matched load.) (8.13)

S22=

V2
V2

| V + =0 =out (output reflection coefficient when input

1

port is terminated in a matched load.) (8.14)

S-parameters as defined above, have many advantages at high
RF/microwave frequencies which can be briefly stated as:
a. S-parameters provide a complete characterization of a network, as
seen at its two ports.
b. S-parameters make the use of short or open (as prescribed at lower
frequencies) completely unnecessary at higher frequencies. It is a
known fact that the impedance of a short or an open varies with
frequency which is one reason why they are not useful for device
characterization at high RF/Microwave frequencies. Furthermore,
the presence of a short or open in a circuit can cause strong
reflections (since |L|=1), which usually lead to oscillations or
damages to the transistor circuitry.
c. S-parameters require the use of matched loads for termination and
since the loads absorb all the incident energy, the possibility of
serious reflections back to the device or source is eliminated.

161

______________________________________________________
EXAMPLE 8.6
Given the [ABCD] matrix for a two-port network, derive its [S]
matrix (see Figure 8.8).

Solution:
To obtain S11, we terminate port 2 in a matched load and find the
input reflection coefficient (in) as follows:
Z ZO
S 11 = in = in
Z in + Z O
Where Zin=v1/i1. Substituting for v1 and i1 in terms of [ABCD] we
have:
v1=Av2 -Bi2
i1=Cv2 -Di2
v2= -Zoi2
v
Av 2 Bi 2 AZ O + B
Z in = 1 =
=
i1
Cv 2 Di 2 CZ O + D
Now, Substituting for Zin, we can write S11 as:
A + BYO CZ O D
S 11 = in =

where is given by:

=A+BYo+CZo+D
Similarly, S12, S21 and S22 can be found as follows:
2( AD BC)
S 12 =

2
S 21 =

A + BYO CZ O + D
S 22 = out =

_______________________________________________________

162

In general, using the same technique as demonstrated in example 8.6,

any set of network parameters can be converted into another set of
parameters. Appendix H shows the conversion relation between the
z-, y-, h-, ABCD- and the S-parameters.
The conversion among the three transistor configurations is an
important relation which becomes useful in many practical design
situations. Appendix I shows the conversion relation between the yparameters of a transistor in common-emitter, common-base and
common-collector configurations. If parameters other than Yparameters (e.g. S-parameters) are needed, then appendix H can be
used effectively to convert Y- to S-parameters.

8.5 PROPERTIES OF S-PARAMETERS

The S-parameters of an N-port network, in general, has certain
properties and inter-relationships amongst the parameters themselves
which are worth considering. In the following discussion, due to the
popularity and frequent use, we limit our discussion solely to twoport networks. Depending upon whether the network is reciprocal or
lossless the S-parameters will have different properties which are
discussed below.

8.5.1 Reciprocal Networks

A reciprocal network is defined to be a network that satisfies the
reciprocity theorem which is defined as:
RECIPROCITY THEOREM: Is a theorem stating that the interchange
of electromotive force at one point (e.g. in branch k, vk) in a passive
linear network, with the current produced at any other point (e.g.
branch m, im) results in the same current (in branch k, ik) when the
same electromotive force is applied in the new location (branch m,
vm), that is,
vk/im=vm/ik
(8.15a)
or,
Zkm=Zmk
(8.15b)

163

As observed, this theorem only applies to passive networks having

linear bilateral impedances. Networks that satisfy this condition
include all passive networks that contain linear passive elements
including resistors, capacitors, inductors and transformers except:
a. Independent or dependent sources,
b. Non-linear elements and/or active solid state devices such as
diodes, transistors, etc.
It can be shown that for all reciprocal networks, the [S] matrix is
symmetrical, i.e.
S12 = S21
(8.16a)
Generalizing the above, it can be shown that for an N-Port network:
Sij = Sji for i
j
(8.16b)
Where,
i=1, N
j=1, N.
A SPECIAL CASE: A SYMMETRICAL RECIPROCAL NETWORK
A special case of a reciprocal network is a symmetrical network.
Symmetrical networks are defined as Networks that have identical
size, element values and arrangement for corresponding electrical
elements in reference to a plane or line of symmetry.
Due to the symmetry of a network topology and by observation, the
input impedance obtained by looking into the input port is equal to
the impedance looking into the output port. The equality of input and
output impedances leads to the equality of input and output
reflection coefficients in addition to equality of S12 and S21 as
required by the reciprocity theorem stated earlier. Therefore for
symmetrical passive networks we can always write:
S11 = S22
(8.17a)
S12 = S21
(8.17b)
Or, in general for any symmetrical passive N-port network we can
write:S-parameters as:
Sii = Sjj
(8.18a)
Sij = Sji
(8.18b)

164

Where, ij and,
i=1,......N
j=1,......N.

8.5.2 Lossless Networks

For a lossless passive network (i.e. one containing no resistive
elements) the power entering the circuit will always equal to the
power leaving the network, i.e. the power is conserved. This
condition will impose a number of restrictions on the s-parameters
which gives rise to the unity and zero property as follows:
A. THE UNITY PROPERTY OF [S] MATRIX

This property states that for a passive lossless two-port (or in general
an N-Port network), the sum of the products of each term of any one
row (or any one column) multiplied by its own complex conjugate is
unity, i.e.
N

S
i =1

ij

S *ij = 1 ,

j=1,2,....,N

(8.19)

Where i and j are row and column numbers respectively.

For a two-port network, Equation (8.19) yields two equations:
S11S11*+S12S12*=1
(8.20a)
S22S22*+S21S21*=1
(8.20b)
Furthermore, if the lossless network is also reciprocal
S21), these two equations are greatly simplified as follows:
S12=S21
|S11|=|S22|
|S11|2+|S21|2=1

(i.e. S12=
(8.21a)
(8.21b)
(8.21c)

B. THE ZERO PROPERTY OF [S] MATRIX

This property states that for a passive lossless N-port network, the
sum of the products of each term of any row (or any column)
multiplied by the complex conjugate of the corresponding terms of
any other row (or column) is zero:

165
N

S
k =1

ki

S *kj = 0,

for i
j, & i,j=1,2,....,N

(8.22)

where i and j are row and column numbers, respectively.

For a two-Port network this equation simplifies into two equations:
S11S12*+S21S22*=0
(8.23a)
S11S21*+S12S22*=0
(8.23b)
Furthermore, if the lossless network is also reciprocal (i.e. S12 = S21),
then the above two equations simplify into one equation:
S12=S21
(8.24)
S11S21*+S21S22*=0
Note: A matrix satisfying the zero and unity property is called a
unitary matrix.
C. ANALYSIS OF RECIPROCAL LOSSLESS NETWORKS
From the zero and unity properties of the S-matrix, The S-parameters
of a reciprocal lossless network are constrained by Equations
(8.20), (8.21) and (8.24) as follows:
S21=S12
(8.25a)
|S11|=|S22|
(8.25b)
|S11|2+|S21|2=1
(8.25c)
(8.25d)
S11S21*+S21S22*=0
If we let:
S 11 = S 11 e j11 ,
S 22 = S 22 e j 22
and,
S 21 = S 21 e j 21

Then Equations (8.25c) and (8.25d) give:

|S21|=(1-|S11|2)11/2
2 /2
S 11 1 S 11
e j( 11 21 ) + e j( 21 22 ) = 0

) (

which
yields:
e j( 11 21 ) + e j( 21 22 ) = 0 e j( 11 21 ) = e j e j( 21 22 )

(8.26a)

166

22=2 21 - 2n

11+
Or,
+ 22
1

21 = 11
+ m n
2
2

For n=0,1,2,...

(8.26b)

Equations (8.26a) and (8.26b) provide the magnitude and phase of

S21 (or S12) in terms of magnitude and phase of S11 and S22.
Therefore from a measurement knowledge of S11 and S22, one can
completely describe and specify a reciprocal lossless two-port
network. This use of S-parameters in specifying a lossless and
reciprocal two-port network, shows its usefulness and versatility.
The following will illustrate the concept of S-Parameters further.
_______________________________________________________
EXAMPLE 8.7
What are the s-parameters of a series element (Z) as shown in
Figure 8.9?

Solution:
Since this is a reciprocal and symmetrical network, we have:
S11=S22,
S12=S21
So we only need to find S11 and S21.

167

NOTE: This is not a lossless network since Z=R+jX has a lossy

component!

V1

| +
(8.27)
+
V1 V2 = 0
According to Equation (8.27) S11 is the input reflection coefficient
when the output is matched (see Equation 8.27); that is,
S11=IN=(ZIN-Zo)/(ZIN+Zo),
Where ZIN=Z+Zo, thus we have:
(8.28)
S11=Z/(Z+2Zo)
a. S11=

V2

| +
(8.29)
+
V1 V2 =0
From Equation (8.29) we can see that S21 is the voltage gain (or loss)
when the output is matched. Thus by applying a source voltage (E1)
at port 1, the voltage gain is found as follows:
I=E1/(Zo+ZIN)
V2=V2-+V2+

b. S21=

Since the load is matched, v2+ =0. Thus we have:

V2-= ZoI
V1=V1++V1-= V1+(1+S11)=ZINI
V1+= ZINI/(1+S11)

(8.30a)

(8.30b)

Dividing (8.30a) by (8.30b), we have:

S21= V2-/V1+=Zo(1+S11)/ZIN
S21=2Zo/(Z+2Zo)
(8.31)
Observation: For a "series Z" network, from Equations (8.29) and
(8.31) we can see that:
S21 = 1-S11
(8.32)
Therefore the whole S-matrix can be written as:

168

2Z O
Z

Z + 2Z
Z + 2Z O
O

S=
Z
2Z O

Z + 2Z O Z + 2Z O
_______________________________________________________
EXAMPLE 8.8
What is the s-parameters of a shunt element (Y) as shown in Figure
8.10?

Solution:
Similar to example 8.1, this is a reciprocal and symmetrical network,
thus:
S11=S22,
And,
S12=S21
So we only need to find S11 and S21.

V1
a. S11= + | V + =0
V1 2
S11=(ZIN-Zo)/(ZIN+Zo)
ZIN= (1/Y||Zo)
Substituting for ZIN in S11 above, we obtain:
(8.33)
S11=-ZoY/(2+ZoY)

169

V2

| +
+
V1 V2 =0
By applying a source voltage E1 to port 1, we obtain:
I=E1/(Zo+ZIN)
b. S21=

Since port 2 is terminated in a matched load (i.e. V2+=0), we can

write:
V2=V2-=(1/Y||Zo)I
(8.34a)
V1=V1++V1-= V1+(1+S11)=ZINI
V1+= ZINI/(1+S11)
(8.34b)
Dividing (8.34a) by (8.34b), we obtain:
S21= v2-/v1+=(1/Y||Zo)(1+S11)/ZIN
S21=2/(2+ZoY)

(8.35)

Observation: For a "shunt Y" network, from Equation (8.33) and

(8.35) we can see that:
S21=1+S11
(8.36)
Therefore the whole S-matrix can be written as:
2
ZOY

2 + Z Y 2 + Z Y
O
O

S=

Z
Y
2
O

2 + Z O Y 2 + Z O Y

8.6 SHIFTING REFERENCE PLANES

The S-parameters relate amplitude and phase of traveling waves,
which are incident on, transmitted through, or reflected from a
network terminal. Therefore the location of the reference plane must
be known precisely to calculate or measure the exact phase of the Sparameters.
Consider a two-Port network in which the reference plane at port 1
has moved a distance l1 to port 1'. Similarly, the reference plane at
port 2 has moved a distance l2 to port 2' as shown in Figure 8.11.

170

The voltage waves at each new port (i.e. 1' and 2') can now be
written as:
Vi+'=Vi+ eji, i=1,2
(8.37a)
Vi-'=Vi- e-ji, i=1,2
(8.37b)
where i=li (i=1,2) is the electrical length corresponding to the
reference plane shift at each port.
Inverting Equation (8.37), we can write:
Vi+=Vi+'e-ji, i=1,2
Vi-=Vi-'eji, i=1,2

(8.38a)
(8.38b)

Upon substitution of Equation (8.38) in:

[V-]=[S][V+]
and further mathematical manipulation, we obtain [S'] which is the
shifted S-parameters as:
j 2 1
11
j ( 1 + 2 )
21

[S ] = S e
S e
'

or conversely,

'
S 11
e j 2 1

[S] =

'

S 21e

j( 1 + 2 )

S 12 e j( 1 + 2 )

S 22 e j2 2

'
S 12
e j( 1 + 2 )

S '22 e j2 2

To summarize this analysis, we note that:

S'ii=Sii e-j2i i=1,2
S'ij=Sij e-j(i+j) i
j, i=1,2

(8.39)

(8.40)

(8.41)
(8.42)

Equation (8.41) shows that the phase of Sii is shifted by twice the
electrical length, because the incident wave travels twice over this

171

length upon reflection. On the other hand, at port i (i = 1,2), (8.42)

shows that Sij (i j) is shifted by the sum of the electrical lengths
because the incident wave must pass through both lengths in order to
travel from one shifted port to the other.

8.7 TRANSMISSION MATRIX

The following discussion in general, applies to a cascade of N-port
networks. However, for the sake of simplicity, we limit our analysis
to two-port networks only. When cascading a number of two-port
networks in series, a more useful network representation is needed in
order to facilitate the calculation of the overall network parameters.
This new representation should relate the output quantities in terms
of input quantities. Using such a representation will enable one to
obtain a description of the complete cascade by simply multiplying
together the matrices describing each network.
At low frequencies, the transmission matrix (also known as ABCD
matrix) is defined in terms of the net input voltage and current as the
independent variables and output net voltage and current as the
dependent variables.
However, at high RF and microwave frequencies, the transmission
matrix is expressed in terms of the input incident and reflected
waves as the independent variables and the output incident and
reflected waves as the dependent variables.
Using the latter definition at RF/microwave frequencies, the
transmission matrix formulation becomes very useful when dealing
with multi-stage circuits (such as filters, amplifiers, etc.) or infinitely
long periodic structures such as those used in small-wave circuits for
traveling wave tubes, etc.
The transmission matrix (or T-matrix) for a two port network as
shown in Figure 8.12, is defined as:

172

V1+ T11
=
V1 T21

T12 V2

T22 V2+

(8.43)

The relationship between S- and T-parameters can be derived using

the above basic definition as follows:

T11
T
21

1
T12 S 21
=
T22 S 11
S 21

S 22

S 21

S 11S 22
S 12
S 21

(8.44)

The reverse relationship expressing [S] in terms of [T] matrix can

also be derived with the following result:

S 11
S
21

T21
S 12 T11
=
S 22 1
T11

T21 T12
T11

12

T11

T22

(8.45)

For a cascade connection of two port networks as shown in Figure

8.13, the overall T-matrix can be obtained as follows:
V1+ T11 T12 V2
(8.46a)
=
. +
V1 T21 T22 V2

V ' 1+ T11'
' = '
V 1 T21

T12' V ' 2
. +
T22' V ' 2

(8.46b)

173

But we note that:

V2+=V'1-,
V2-=V'1+
Therefore combining (8.46) and (8.47), yields:
V1+ T11 T12 T11' T12' V ' 2
.
=
. '
' '+
V1 T21 T22 T21 T22 V 2

(8.47a)
(8.47b)

(8.48)

Thus the total T-matrix is multiplication of the two T-matrices:

[T]tot=[T][T ']
(8.49)

8.8 GENERALIZED SCATTERING PARAMETERS

The scattering matrix defined earlier was based on the assumption
that all ports have the same characteristic impedances (usually
Zo=50 ). Even though this is the case in many practical situations,
however, there are cases where this may not apply and each port has
a non-identical characteristic impedance (see Figure 8.14). Thus a
need to generalize the scattering parameters arises.

In this case we have to modify our ordinary definition for the [S]
matrix to include the characteristic impedances at each port. Taking
each port's characteristic impedances into account, we need to define
two normalized voltage waves as follows:
ai=Vi+/Zoi, i=1,2
(8.50)
bi=Vi-/Zoi, i=1,2
(8.51)

174

where "i" is the port number, "ai" represents the normalized incident
voltage , "bi" represents the normalized reflected voltage wave from
the ith port and "Zoi" is the characteristic impedance at the ith port
(Note: Zoi is a real number for lossless lines).
Thus the total voltage and current at each port can now be written as:
Vi=Vi++Vi-=
Zoi(ai+bi)
(8.52)
Ii=Ii+-Ii-= Vi+/Zoi - Vi-/Zoi =(ai-bi)/
Zoi
(8.53)
The average net power delivered to the ith port can now be expressed
in terms of ai and bi with no further concern about different Zoi at
each port:
Pi=Re[ViIi*]/2
=Re[|ai|2-|bi|2+(ai*bi-aibi*)]/2
(8.54)
Noticing that (ai*bi-aibi*) term is purely imaginary leads the
expression for the net real power:
Pi= (|ai|2-|bi|2)/2
(8.55)
This equation is meaningful since it is clearly showing that the net
power delivered to each port is equal to the normalized incident
power less the normalized reflected power.
The generalized [s] matrix can now be defined in terms of the
normalized voltage waves as
follows:
[b] = [S][a]
(8.56)
where each element of the generalized [S] matrix is now defined as:
b
s11= 1 |a 2 =0 =in (reflection coefficient at port 1 with port 2
a2
matched)
(8.57)

s12=

b1
|a = 0
a2 1

(transmission coefficient from port 1 to port 2 with

port 2 matched)

(8.58)

175

s21=

b2
|a = 0
a1 2

(transmission coefficient from port 2 to port 2

with port 1 matched)

s22=

b2
|a = 0 =out
a2 1

(8.59a)

(reflection coefficient at port 2 with port 1

matched)
(8.59b)
Clearly these definitions are very similar to the earlier ones for the
[S] matrix except that "Vi+" and "Vi-" are replaced by "ai" and "bi",
respectively. Alternately, each element can be expressed as a general
equation by:
b
Sij= i |ak = 0
for i,j,k=1,2 and kj
(8.60a)
aj

A SPECIAL CASE
Consider a network, with identical characteristic impedances at all
ports, having a known [S] matrix. If transmission lines of unequal
characteristic impedances (Zoi) are connected to each port, the new
[s] matrix for the entire network with the help of (8.60a), can now be
written as :

Vi Z oj
(Sij)new= +
| +
for i,j,k=1,2 and kj
(8.60b)
Vj Z oi Vk = 0
we can simplify Equation (8.59) to yield:
Z oj
(Sij)new =(Sij)old
for i,j,k=1,2 and kj
(8.60c)
Z oi
In the next section we will discuss the subject of "signal flow
graphs" whereby any complex circuit can be analyzed in terms of a
simple diagram which can yield the relation between desired
variables.

8.9 Signal Flow Graphs (SFGs)

Any linear system, or more specifically any linear electrical network,
can be described by a set of simultaneous linear equations. Solutions
to these equations can be obtained by the following methods:

176

a. The elimination theory which is a method of successive

substitutions,
b. The Cramer's rule which is a method of solving by using
determinants, and
c. Any of the topological techniques such as the "flow graph
techniques" represented by the works of Mason.
Although the algebraic manipulation methods (a) and (b) above can
be executed by a computer with relative ease and speed, however,
they do not allow a pictorial analysis or perspective on the physical
nature and the signal flows taking place inside the linear system.
The signal flow graphs, through the use of graphical diagrams,
provide a physical insight into the cause-effect relationships between
the variables of the system. This method of analysis enables the
circuit analyst to gain an intuitive understanding about the system
(or network) operation.
In the previous sections, we have seen how the incident, reflected
and transmitted waves are inter-related through a series of linear
equations expressed concisely by the S-parameter matrix. This fact
indicates that we are dealing with a linear system to which the signal
flow graph can directly be applied.
In essence, a signal flow graph is an alternate and yet a simpler
method to the "block diagram method" in representing a complicated
linear system.
The advantage of a signal flow graph method over a block diagram
method is the availability of a flow-graph gain formula which
provides the relation between system variables without requiring any
detailed procedure for manipulation or reduction of the flow graph.
In This section we will present a detailed discussion about the
application and construction of signal flow graphs for analysis of
any linear RF/MW network or system. Before we proceed into a
detailed discussion of signal flow graphs, let us first define it.

177

DEFINITION-SIGNAL FLOW GRAPH: is defined to be an abbreviated

block diagram consisting of small circles (called nodes)
representing the variables that are connected by several directed
lines (called branches) representing one-way signal multipliers; an
arrow on the line indicates direction of signal flow, a letter near the
arrow indicates the multiplication factor.
Having defined what a signal flow graph is, we will now discuss the
main features and its construction as well as a reduction technique to
generate ratios of any set of desired variables.

8.9.1 Main Features of a Signal Flow Graph

The main components of a signal flow graph are nodes and branches
which are defined as follows:
a. Node: Each port of a microwave network (e.g. the ith port) can be
represented by two nodes:
Node ai- representing a wave entering port i (an independent
variable)
Node bi- representing a wave reflected from port i (a
dependent variable)
b. Branch: A branch is a directed path between an "a-node" (an
independent variable) and a "b-node" (a dependent variable). A
branch represents a signal flow from node a to b. The multiplication
factor placed near the arrow is the associated S-parameter.
A signal flow graph is a convenient technique to represent and
subsequently analyze the flow of waves in a Microwave network.
There are certain rules one needs to follow in constructing one:
1. Each variable is shown as a node.
2. S-parameters are shown by branches.
3. Branches enter dependent variable nodes (reflected wave
variables).
4. Branches emanate from independent variable nodes (incident
wave variables)

178

5. A node is equal to the sum of the branches entering it. For

example consider Figure 8.15, where the dependent variable b1 can
be written as: b1=S11a1+S12a2

______________________________________________________
Example 8.9
Draw the signal flow graph for a linear two-port microwave network
as shown in Figure 8.16.

Solution:
A two-port network at microwave frequencies can be characterized
by S-parameters given by:
b1=S11a1+S12a2
b2=S21a1+S22a2
Therefore using the rules set forth (1-5) set forth above, the signal
flow graph for a two-port network consists of two a-nodes (a1,a2),
two b-nodes (b1,b2) and four branches (S11, S12,S21,S22) as shown in
Figure 8.11)
_______________________________________________________
EXAMPLE 8.10
Find the signal flow graph (SFG) of a microwave amplifier shown in
Figure 8.17.

179

Solution:
In order to obtain the SFG for the microwave amplifier, we apply
axiom #4 and dissect or compartmentalize the problem into three
separate areas, obtain SFG for each part and then create a final SFG
by combining these three SFGs into one (see Chapter 1, Solutions
to Problems section):
a. Source SFG
The SFG of a signal generator with an internal impedance (ZS) is
obtained from Figure 8.18 as follows:
Vg + Vg
+

Vg=ES+ ZSIg Vg + Vg =ES+ Z S

ZO
ZO

Vg

ZO

ZO
ZS ZO

+ E S
ZO + ZS
ZO ZS + ZO

Vg

Thus Equation (8.61) can be written as:

bg=Sag+bS
Where

(8.61)

(8.62)

180

bg =

Vg

ag =

Vg

S =

ZO
+

ZO

ZS ZO
ZS + ZO

b S = ES

ZO

ZO + ZS
The source SFG is shown in Figure 8.19.

(8.63)
(8.64)

b. Load SFG
The SFG of a load impedance (ZL) is obtained from Figure 8.20 as
follows:
V + V
VL= ZLIL VL+ + VL- = Z L L L
Z
Z O
O

+
V
V Z ZO

L = L L
(8.65)
ZO
Z O Z L + Z O

Thus Equation (8.65) can be written as:

bL=LaL
Where

181

bL =

VL

aL =

VL

ZO
+

ZO

ZL ZO
ZL + ZO
The load SFG is shown in Figure 8.21.
L =

(8.66)

c. Linear Two-Port Network SFG

The SFG of a linear two-port network has already been obtained in
example 8.9 and is shown in Figure 8.16.
d. Final SFG
Combine the SFG for part (a), (b) and (c) to obtain the final SFG as
shown in Figure 8.22.

______________________________________________________

8.9.2 Signal Flow Graph Reduction

Once a microwave network has been represented in terms of a signal
flow graph, the wave amplitude ratio of any two variables can be
obtained by using the following two techniques:
1. Mason's Rule (from Control System Theory)

182

This method has been well documented in any "control System" text
and will not be repeated here.
2. Signal Flow Graph Reduction Technique
This method is worth presentation and will be further discussed in
this work. In its simplest form, it consists of reduction of a signal
flow graph to a single branch using four basic decomposition rules.
These rules are briefly summarized herein but each can easily be
obtained by simple observation and basic application of the basics of
signal flow graphs set forth earlier.
RULE #1-SERIES RULE: Two branches in series whose common
node has only one incoming and outgoing wave may be combined
into a single branch with a coefficient (or multiplication factor)
equal to the product of the two coefficients (see Figure 8.23). That is
Sc=SaSb
(8.67)

RULE #2- PARALLEL RULE: Two branches in parallel both going

from one node to another may be combined into a single branch
whose coefficient is the sum of the two coefficients (see Figure
8.24). That is
Sc=Sa+Sb
(8.68)

RULE #3- SELF LOOP RULE: A branch beginning and ending on the
same node (called a self loop) with a coefficient Sl (see Figure 8.25)

183

can be eliminated by multiplying the coefficients of the branches

1
.
(feeding that node) by
1

S
l

RULE #4- SPLITTING RULE: Any node can be split into two separate
nodes where each of the two nodes are connected only once to the
incoming and the outgoing nodes as shown in Figure 8.26.

_______________________________________________________
EXAMPLE 8.11
Consider a two-port network characterized by a scattering matrix
[S] driven by a source with an internal impedance of (ZS) and
terminated in a load impedance (ZL) as shown earlier in Figure
8.27. Calculate:
a. The input reflection coefficient (in), and
b. The output reflection coefficient (out);
(a) and (b) are to be calculated in terms [S], the source reflection
coefficient (S), and the load reflection coefficient (L).
Solution:
The overall signal flow graph was shown earlier in Figure 8.22, from
which we can derive the following for each of the two cases:

184

a. The input reflection coefficient (

in)
The signal flow graph (SFG) for this network is shown in Figure
8.28.

Using the 4 decomposition rules stated above, we can reduce the

signal flow graph step by step as shown in Figure 8.29. Using Figure
8.29(d), we can write in as:
S S
in = S 11 + 12 21 L
(8.69)
1 S 22 L

185

b. The output reflection coefficient (

out)
The signal flow graph for this part is shown in Figure 8.30. Very
similar to part (a) for in, we can reduce the signal flow graph step
by step and write a similar equation for out as:
S S
out = S 22 + 12 21 S
(8.70)
1 S 11 S

8.9.3 Applications of Signal Flow Graphs (SFGs)

Consider a two-port network and its corresponding signal flow graph
as shown earlier in Figures 8.17 and 8.22, respectively. We know
that the square of the incident normalized wave represents the
incident power while the reflected power is represented by the

186

reflected normalized wave. Using SFG, we can find the relationship

between the various wave variables and therefore the several types
of power as follows:

a. PIN
PIN is the input power to the network given by:
1 2 1
1 2
2
2
PIN = a 1 b 1 = a 1 1 IN
2
2
2

(8.71)

b. PAVS
PAVS is the power available from the source, defined as the input
power (PIN) delivered by the source to a conjugately matched input
impedance:
2
2
1
1
(8.72)
PAVS = PIN |IN =S * = b g a g
2
2
Where
Sbg
ag=
bS
bg=bs+
Sag=
(8.73)
2
1 S
Thus we have:

PAVS =

bS

1 S

(8.74)

c. PL
PL is the power delivered to the load given by:
1
1
1
2
2
2
2
PL = b 2 a 2 = b 2 1 L
2
2
2

(8.75)

d. PAVN
PAVN is the power available from the network (defined as the power
delivered to the load when the load is conjugately matched to the
network is given by:
1
2
2
1
PAVN = PIN |L = OUT * = b 2 a 2
2
2
L = OUT *

187

PAVN =

1
2
b 2 1 OUT
2

(8.76)

8.9.4 Power Gain Expressions

Using the results obtained in the previous section for various powers,
we can now obtain several power gain expressions as follows:

a. TRANSDUCER POWER GAIN (GT)

b2
P
GT L =
PAVS
bS

(1 )(1 )
2

(8.77)

Using the SFG technique, the ratio b2/bS is obtained to be:

b2
S 21
=
(8.78)
b S (1 S 11 S )(1 S 22 L ) S 12 S 21 S L
Thus GT can be written as:
GT =

1 S

1 IN S

S 21

1 L

1 S 22 L

(8.79)

Or, we can write an alternate form for GT as:

GT =

1 S

1 S 11 S

S 21

1 L

1 OUT L

(8.80)

b. POWER GAIN (GP)

b2 / bS
P
GP L =
PIN
a1 / b S

2
2

1 L 2

1 2
IN

(8.81)

Using the SFG technique, the ratio a1/bS is obtained to be:

a1
1 S 22 L
=
b S 1 (S 11 S + S 22 L + S 12 S 21 S L )(1 S 11 S ) + S 11S 22 S L
(8.82)
Thus GP can be written as:

188

GP =

1
1 IN

S 21

1 L

1 S 22 L

(8.83)

c. AVAILABLE POWER GAIN (GA)

L= out*

b2
P
G A AVN =
PAVS
bS

(1 )(1
2

OUT

Using (8.78), the ratio b2/bS when L= out* we obtain:

b2
S 21
=
b S (1 S 11 S ) 1 OUT 2

(8.84)

(8.85)

Thus GA can be written as:

GA =

1 S

1 S 11 S

S 21

1
1 OUT

(8.86)

Note: So far we have discussed the various power gains which is

mostly encountered in Microwave amplifier design. However, in
most audio and RF amplifier design we use voltage gain. The
voltage gain of an amplifier is defined to be the ratio of total output
voltage to the total input voltage as follows:
V
a + b2
A V = OUT = 2
(8.87)
VIN
a1 + b1
Dividing Equation (8.87) by bS, we obtain:
a / b + b2 / bS
AV = 2 S
(8.88)
a1 / b S + b1 / b S
Using Equations (8.78) and (8.82) for b2/bS and a1/bS and similar
derivations for b1/bS and a2/bS, we obtain AV to be given by:
S 21 (1 + L )
AV =
(8.89)
(1 S 22 L ) + S 11 (1 S 22 L ) + S 12S 21L

189

8.10 SUMMARY
Having defined the S-parameters and derived power gain
expressions in this chapter, we will present important concepts about
design of matching networks as well as stability of two-ports in the
next chapters which will lay the foundation for active circuit design.

Chapter 8- Symbol List

A symbol will not be repeated again, once it has been identified and
defined in an earlier chapter, with its definition remaining
unchanged.
A, B, C, D - ABCD parameters
h11, h12, h21, h22 Hybrid or h-parameters
i1 - Current into port 1 of a network
i2 - Current into port 2 of a network
ik Current at branch k
SFG Signal Flow Graph
S11, S12, S21, S22 Scattering parameters (or S-parameters)
T11, T12, T21, T22 Transmission parameters (or T-parameters)
v1 - Voltage at port 1 of a network
v2 - Voltage at port 2 of a network
vk - voltage at branch k
Y11, Y12, Y21, Y22 Admittance parameters (or Y-parameters)
Z11, Z12, Z21, Z22 Impedance parameters (or Z-parameters)

CHAPTER-8 PROBLEMS
8.1)
P8.1

Determine the S-parameters of the circuit shown in Figure

Figure P8.1

190

8.2)

8.3)

Find the scattering matrix and the transmission matrix of a

loss-less transmission line of length "l" in a 50 system
when:
a. Characteristic impedance of the line is (Zo=50 ),.
b. Characteristic impedance of the line is (Zo=100 ).
Find the generalized scattering matrix of a two-port
consisting of a junction of two lossless transmission lines as
shown in Figure P8.3.

Figure P8.3

8.4)

Find the transmission matrix of the circuit shown in Figure

P8.4

Figure P8.4

8.5)

Derive the [ABCD] matrix for the step-down transformer

circuit shown in Figure P8.5.

Figure P8.5

191

8.6)

A two-Port network has the scattering matrix as shown

below. From
this data:
a) Determine whether the network is reciprocal or lossless,
b) If the output terminals are shorted together, what will the
input reflection coefficient be?
0.2 j0.9
[S] =

j0.9 0.4

REFERENCES
[8.1] Chang K. Microwave and Optical Component. Vols I,
II. New York: John Wiley & Sons, 1989.
[8.2] Cheung, W. S. and F. H. Levien. Microwave Made Simple.
Dedham: Artech House, 1985.
[8.3] Desor, C. A. and E. S. Kuh. Basic Circuit Theory.
Tokyo: McGraw-Hill, 1969.
[8.4] Gonzalez, G. Microwave Transistor Amplifiers,
Analysis, and Design, 2nd ed. Upper Saddle River:
Prentice Hall, 1997.
[8.5] Gardiol, F. E. Introduction to Microwaves. Dedham:
Artech House, 1984.
[8.6] Ishii, T. K. Microwave Engineering. 2nd ed., Orlando:
Harcourt Brace Jovanovich Publishers, 1989.
[8.7] Laverghetta, T. Practical Microwaves. Indianapolis:
Howard Sams, 1984.
[8.8] Pozar, D. M. Microwave Engineering, 2nd ed. New York:
John Wiley & Sons, 1998.
[8.9] Saad, T. Microwave Engineers Handbook. Vols I, II.
Dedham: Artech House, 1988.
[8.10] Scott, A. W. Understanding Microwaves. New York: John
Wiley & Sons, 1993.
[8.11] Vendelin, G. D. Design of Amplifiers and
Oscillators by the S-Parameter Method. New York: John
Wiley & Sons, 1981.
[8.12] Vendelin, G. D., A. M. Pavio, and Ulrich L. Rhode.
Microwave Circuit Design. New York: John Wiley, 1990.

192

RF & Microwave Design Essentials

CHAPTER 9

The Smith Chart

9.1 INTRODUCTION
One of the most valuable and yet pervasive graphical tools in all of
microwave engineering is the "Smith chart", originally developed in
1939 by P. Smith at the Bell Telephone Laboratories. This chart is
the reflection coefficient-to-impedance/admittance converter or vice
versa and can greatly simplify the analysis of complex design
problems involving transmission lines or lumped elements.
Furthermore, the smith chart provides valuable information about
the circuit's performance when line lengths change or new elements
are added to the circuit, particularly where obtaining the same
amount of information through mathematical models and
calculations would be very tedious and time consuming.
Over the years, it has proven itself to be a most useful tool and is
thus employed frequently in all stages of circuit analysis or design
whether done through manual methods or computer-aided-design
(CAD) Software techniques.

Matthew M. Radmanesh, Ph.D. 193

9.2 A VALUABLE GRAPHICAL AID

Considering the equation for the reflection coefficient (as given
earlier in Chapter 8) we have:
Z ZO ZN 1
=
=
(9.1)
Z + ZO ZN + 1
Z
where Z N =
=r+jx is the normalized impedance and ZO is the
ZO
characteristic impedance of the transmission line or a reference
impedance value.
Based on (9.1), the Smith chart can be derived mathematically as
discussed in the next section. This chart is a plot of for different
normalized resistance and reactance values, where the circuit is
assumed to be passive i.e. Re(Z)0.
It can be shown that the loci of constant resistance values are circles
centered on the horizontal (or real) axis while the loci of constant
reactance values are circles centered on the vertical (or imaginary)
axis offset by one unit.

9.3 DERIVATION OF SMITH CHART

The Smith chart is a plot of
Z 1
= N
(9.2)
ZN + 1
in the -plane as a function of r and x. Using (9.2) and separating
in terms of its real part (U) and imaginary part (V) we obtain:
ZN=r+jx
r + jx 1
(9.3)
=
= U + jV
r + jx + 1
r2 1 + x2
(r + 1) 2 + x 2
2x
V=
(r + 1) 2 + x 2
U=

(9.4)
(9.5)

194

RF & Microwave Design Essentials

At this juncture we note that by using (9.4) and (9.5), we can obtain
two families of circles which when superimposed on each other will
make up the entire Smith chart. The procedures to obtain these two
families of circles are described next.
a. Constant-r circles
The first family of circles is obtained by eliminating "x" from (9.4)
and (9.5) which gives:
2

1
2
U
+V =

r +1
r +1

(9.6)

Equation (9.6) represents a family of circles with a center located at

r

(Uo,Vo)=
,0 ,
(9.7a)
r +1
with a radius of
1
R=
(9.7b)
.
r + 1
From Equations (9.7) we can observe that all constant-r circles are
centered on the real axis with a shrinking size as "r" is increased. In
this regard, we note that r=0 circle is the most-outer circle of the
Smith chart while r= circle is reduced to a point at (0,1). Figure 9.1
depicts this concept further.
b. Constant-x circle
The second family of circles is obtained by eliminating "r" from
(9.4) and (9.5) which gives:

(U 1)2 + V 1

1
=
x
x

(9.8)

Matthew M. Radmanesh, Ph.D. 195

Equation (9.8) represents a family of circles with a center located at

1
(Uo',Vo' )= 1,
(9.9a)
x
with a radius of
1
R'=
(9.9b)
x
From Equations (9.9) we can observe that all constant-x circles are
centered on a shifted line parallel to the imaginary axis (by +1 unit
to the right), with a shrinking size as "x" increases. In this regard, we
note that x=0 circle is the real axis of the Smith chart while x=
circles are reduced to a point at (1,0). This case is shown in Figure
9.2.

RF & Microwave Design Essentials

196

As described earlier, plotting the two families of circles as

represented by (9.6) and (9.8) for all values of (r,x) creates a circular
chart commonly known as the Smith chart.
The standard Smith chart is a one-to-one correspondence between
points in the normalized impedance (ZN) plane [where r=Re(ZN)0]
and points in the reflection coefficient () plane. The upper half of
the chart represents normalized impedance values with positive
reactances while the lower half corresponds to negative reactances
(x 0)
NOTE: The Smith chart could have also been developed based on
normalized admittance (YN) as follows:
Y
YN =
= g + jb
(9.10a)
YO
Where YO=1/ZO is the normalized characteristic admittance or a
reference admittance value.
Thus we can write Equation (9.2) as:

Matthew M. Radmanesh, Ph.D. 197

1
1
Y 1
Z N 1 YN

=
=
= N
1
ZN + 1
YN + 1

+1
YN
Now using the transformation:
Y 1

' = N
YN + 1
we obtain the same results as for impedance except
transformation will be from YN-plane into '-plane where
'= -= ej180

(9.10b)

(9.10c)
that the
(9.11)

Equation (9.11) indicates that ' and are only 180 apart but have
the same magnitude, which means that when dealing with
admittances and impedances on the same chart we need to keep in
mind the 180 phase adjustment every time we convert ZN to YN or
vice versa. Therefore a Smith chart can be used as an impedance
chart (Z-Smith chart) or equally as an admittance chart (Y-Smith
chart).
SUMMARY: In summary, using a Smith chart requires awareness
and an understanding of the following transformations:
ZN
YN '
' ej180

The magic of the Smith chart lies in the fact that through the use of
the above transformation, a semi-infinite and an unbounded region
(i.e. 0r, -x+) is transformed into a finite and workable
region (i.e., 01) which creates easily understood graphical
solutions to many complex microwave problems.

9.4 DESCRIPTION OF THE SMITH CHART

As discussed in the previous sections, instead of plotting contours of
constant reflection coefficient, contours of constant normalized
resistance and reactance are plotted in the -plane. A selected

RF & Microwave Design Essentials

198

collection of these contours (which are circles), plotted in the plane, comprise the entire smith chart (commonly known as the
"compressed smith chart") which includes impedances with both
positive and negative real parts (see Figure 9.1). The "compressed
smith chart" is obtained when the entire impedance plane is mapped
on a one-to-one basis onto the reflection coefficient plane as shown
in Figure 9.2.
The "compressed smith chart", even though very general and applies
to both active and passive circuits, is yet impractical and is seldom
used in design. Instead, a more useful part of this chart (called a
standard smith chart) is used in practice for all passive networks
where Re(Z)0, which corresponds to mapping only the right-hand
side half of the impedance plane into a circle in the reflection
coefficient plane with radius ||1 as shown in Figure 9.3.

Standard Smith chart represents a graphical display of impedance-toreflection coefficient transformation, in which all values of
impedance with Re(Z)0 (representing a semi-infinite region of the
resistance-reactance rectangular plane) is mapped one-to-one into a
circle with the radius of one unit in the reflection coefficient plane.

Matthew M. Radmanesh, Ph.D. 199

A full blown-out version of the standard smith chart is shown in

Figure 9.4 where each circle is marked with its corresponding
resistance or reactance value.

Thus the Smith chart is comprised of many circles either fully or

partially enclosed within the outermost circle (||=1) of the standard
Smith chart.

RF & Microwave Design Essentials

200

The set of circles centered on the horizontal (or real) axis are circles
of constant normalized resistance with values ranging from zero
(extreme far left) to infinity (extreme far right) on the chart with
each circle having a variable reactance.
On the other hand, the set of circles centered on the vertical axis
which is offset by one unit from the center, represents circles of
constant normalized reactances with values ranging from - to +
with each circle having a variable positive resistance. These are
shown as partial-circles starting from the right-hand side of the chart
and going above the real axis (representing normalized positive
reactances) and below (representing normalized negative
reactances); the center real axis (horizontal line) represents the zero
reactance circle with an infinite radius.
The markings for the positive and negative normalized reactances
can be seen on the Smith chart close to the outermost circle.
The key to understanding the Smith chart is realizing that it is a
polar plot of the reflection coefficient:
=|
|ej, 0 180
(9.12)
with the reference of zero degrees at the right side of horizontal
semi-axis.
All passive networks (|| 1) have impedance values with Re(Z) 0
which when normalized by the characteristic impedance of the
transmission line (to which they are connected) can be represented
uniquely on the Smith chart.
The real usefulness of the Smith chart lies in its ability to provide a
one-to-one correspondence between reflection coefficient and input
normalized impedance (or admittance) values.
Furthermore, moving a distance "l" toward the load along a lossless
transmission line corresponds to a change in the reflection
coefficient by a factor of e-2jl which corresponds to a counterclockwise rotation of 2l on the Smith chart as shown in Figure 9.5.

Matthew M. Radmanesh, Ph.D. 201

9.5 SMITH CHART'S CIRCULAR SCALES

Consider a standard smith chart as shown in Figure 9.4. Any specific
normalized impedance (z=r+jx) value can be uniquely located on
this chart for r0. The r-values would be on the resistance circles
and x-values on the partial circles for reactance. The positive
reactance would be on the upper half of the chart whereas negative
reactance values are plotted on the lower half of the chart.
Traveling on a transmission line toward the generator corresponds to
moving clockwise on the Smith chart, whereas traveling toward the
load corresponds to moving in a counter-clockwise direction as
indicated by arrows on the left hand side outer-edge of the Smith
chart.
The phase relationship and electrical length along a transmission line
are shown on the outer edge of the Smith chart in terms of two

RF & Microwave Design Essentials

202

secondary scales: One is graduated in fractional wavelength (l/)

and the other in degrees.

9.5.1 Wavelength Scale

Since the impedance value on a transmission line repeats itself every
half wavelength, a complete revolution on the wavelength scale is
equivalent to a half wavelength on the Smith chart.

9.5.2 Degree Scale

The degree scale goes through 180 degrees positive and 180 degrees
negative with 0 degrees being on the right-hand semi-axis. This
scale shows that in a complete revolution of the chart, the reflection
coefficient's phase changes 360 degrees corresponding to a half
wavelength on the wavelength scale.
These two scales (i.e. the wavelength & the degree scales) are
important because they show that when a transmission line is
terminated in a load impedance not equal to the line's characteristic
impedance, the resulting impedance value on the line varies
cyclically every half wavelength (this is of course due to the periodic
nature of the standing wave pattern on the line). This fact is built
into the Smith chart through these two scales and thus facilitates
impedance calculations at various points along a line after the chart
has been entered for a specific impedance value.

9.6 SMITH CHART'S RADIAL SCALES

A number of radially marked scales, at the bottom of the Smith
chart, are placed in such a manner that they can be radially set off
and their values read off from the center of the Smith chart by using
a pair of dividers or compass. These scales are described as follows:

9.6.1 Reflection
Starting with the scale on the right hand side, this scale is designed
to show the ratio of the reflected wave to the incident wave and is
further sub-divided into four scales in the following manner:

Matthew M. Radmanesh, Ph.D. 203

1. REFL. COEF. is the reflection coefficient and has the following

two sub-scales:
a.

b.

VOL: is the voltage reflection coefficient magnitude and is

defined as:
|
|=|V/V+|
(9.13)
This scale starts from 0 at the center and ends at 1 at the outer
rim of the chart.
PWR:

is the power reflection coefficient and is defined to be:

|
| =|V/V+|2
(9.14)
Similar to 1(a) above, this scale starts at 0 at the center and
ends at 1 at the outer edge of the chart.
2

2. LOSS IN DB is the loss due to reflection and is expressed in dB

with the following two sub-scales:
a.

RETN:

is the return loss and is defined as the ratio of the

incident power to the reflected power at any point on the
transmission line, expressed in dB and is equal to:
Rloss(dB)= 10 log10(Pi/PO)=10 log10 (1/|
|2)
|
(9.15)
Rloss(dB)= 20 log10|
This scale starts from 0 (corresponding to ||= 1) at the outer
edge of the chart and approaches infinity at the center of the
chart (where ||=0) which indicates that the more perfect the
load, the less reflection from the load and higher the return
loss.

b.

REFL:

(Reflected loss or Mismatch loss) is the loss caused by

reflection and is equal to the ratio of incident power to the
difference between incident and reflected power expressed in
decibels as follows:
Mloss(dB)=10 log10[Pi/(Pi-PO)]= -10 log10(1-|
|2) (9.16)
This scale starts from zero at the center and approaches to
infinity as || approaches unity at the outer edge of the chart.

9.6.2 TRANSM Loss

TRANSM Loss: is the transmission loss and is used primarily for
lossy transmission lines and has two scales:

RF & Microwave Design Essentials

204

a. LOSS COEF: is the transmission loss coefficient and is used as a

correction factors for the additional line losses created in a lossy
transmission line due to high VSWR. A high VSWR on a line
creates peaks of high current densities alternated with high voltage
density peaks. Since resistive losses are proportional to the current
value squared and dielectric losses are proportional to voltage value
squared, the locale where these peaks of energy lie create additional
losses on the line which is not accounted for through ordinary
calculations.
Thus a correction factor is needed to provide a more accurate
estimation of line losses when a high VSWR exists on the lossy
transmission line. The "LOSS COEF" scale provides the much
needed correction factor when the VSWR on the line is greater than
unity. The correction factor provided by this scale would increase
the calculated line losses which will affect the attenuation factor
calculations.
For example, when the VSWR is 1 (i.e. A matched case) the
correction factor from this scale is read to be one. On the other hand,
when the VSWR of the line is increased to 4 (due to a load
mismatch), then the correction factor is read off to be approximately
2.1, which means that the line losses have more than doubled due to
this high VSWR.
b. 1 DB STEPS: is the transmission loss in 1-dB steps and is used to
calculate VSWR on a lossy transmission line. Graphically, a lossy
line can no longer be represented by a constant VSWR circle,
instead by a spiral on the Smith chart due to the attenuation of both
the incident wave's amplitude traveling "toward the load" and the
reflected wave's amplitude back to the generator. This power loss is
shown in Figure 9.6. From Figure 9.6, it can be seen that:
|
g|2=Pr/Pi (at the source end)

(9.17)

|
L|2=Pr'/Pi' (at the load end)
|
g|<|
L|,

(9.18)

This would hold true as long as the line remains lossy.

Matthew M. Radmanesh, Ph.D. 205

It is important to note that as the measurement plane moves

toward the generator and away from the load, the reflection
coefficient becomes smaller and thus VSWR is reduced as illustrated
in the next example.
_______________________________________________________
EXAMPLE 9.1
Consider an unknown load connected to a lossy 50 cable (with 2
dB of insertion loss) connected to a generator. The VSWR at the
generator end is measured to be 2.0. What is the VSWR at the load?
Solution:
We first plot the VSWR circles at the generator end by dropping a
vertical line from the constant-VSWR circle to intersect the "1 dB
Step" scale at point A (see Figure 9.7). Now we add 2 dB correction
to this value "toward load" as indicated on the scale to obtain point
B. The radius related to point B is that of the load VSWR and is
found by drawing a vertical line from point B to intersect the left
hand semi-axis on the Smith chart. By swinging this radius around,
it is seen that the new VSWR circle has a VSWR=3.2 at the load
end.

RF & Microwave Design Essentials

206

______________________________________________________
As can be seen from Example 9.1, a lossy line can improve the
VSWR at the generator end at the expense of power loss, which may
not always be desirable.
We also note from this example that moving away from the center
(i.e. higher ||) on this scale is labeled as "Toward Load" while
moving toward the center of the chart (i.e. lower ||) is labeled as
"Toward generator".
NOTE: Incidentally, it is interesting to note that the values on "the
transmission loss (in one dB steps)" are one-half of the values on the
"Return loss in dB" scales. This factor of "One-half" is caused by
the fact that the "return loss scale" indicates two-way power
attenuation through a given piece of cable, whereas the transmission
loss is defined as merely a one-way attenuation loss.

9.6.3 Standing Wave

This scale shows the voltage standing wave ratio (VSWR) as
follows:
1. VOL. RATIO: (voltage ratio) this scale plots the VSWR as a
ratio of maximum voltage to minimum voltage as given by the
following equation:

Matthew M. Radmanesh, Ph.D. 207

VSWR =

Vmax 1 +
=
Vmin 1

(9.19)

The "VOL. RATIO" scale progresses from 1 at the center of the

chart (||= 0) to infinity at the left-hand margin (||= 1).

2. IN DB: This scale expresses VSWR in dB by the relation:

(VSWR)dB=20 log10 (VSWR)ratio
(9.20)
_____________________________________________________
EXAMPLE 9.2
What does VSWR= 2.0 on the "Voltage ratio" scale correspond in
dB?
Solution:
Using the adjacent dB scale, we read a value of 6.0 dB on it.
______________________________________________________
In the next chapter we will discuss the applications of the smith chart
which are of great importance to the design of RF and microwave
circuits.

9.7 THE NORMALIZED ZY SMITH CHART

By superimposing two Smith charts, with one 180 rotated, we
obtain a normalized impedance-admittance Smith chart (also known
as a ZY Smith chart) as shown in Figure 9.8. The rotated represents
the admittance, whereas the other chart represents impedance. The
proof for 180 chart rotation to obtain admittance values, is
presented in Chapter 10 (see application #4).
The ZY Smith chart has therefore two markings: one for impedance
chart and another for the admittance chart. Symbols XS and +XS
are used on the left hand side for the impedance chart and BP and
+BP are used for admittance chart, respectively. From These
markings, we can see that positive reactances (+XS) are on the upper
half of the chart while negative reactances (XS) are on the lower
half of the chart, respectively. This situation is reversed for the

RF & Microwave Design Essentials

208

admittance chart, where positive susceptances (+BP) are located on

the lower half and the negative susceptances (BP) are on the upper
half of the chart.

Each point on a ZY Smith chart represents the impedance and the

corresponding admittance value simultaneously, whereby one can
read off these values by a simple glance at the chart. This means that
given an impedance (or admittance) value, its corresponding
admittance (or impedance) value can readily be read off from the
chart without any resort to calculations. This is an important feature

Matthew M. Radmanesh, Ph.D. 209

and a major improvement over a standard Smith chart, since it

greatly facilitates the circuit design process, particularly where
complicated designs are desired. As will be seen in Chapter 11, the
ZY Smith chart is an essential analytical tool and will be extensively
used for RF/Microwave circuit design applications

Chapter 9- Symbol List

A symbol will not be repeated again, once it has been identified and
defined in an earlier chapter, with its definition remaining
unchanged.
Pi Power incident
Pr power reflected
PWR Power reflection coefficient
REFL Reflection loss or mismatch loss
REFL COEF Reflection coefficient
RETN Return loss
Rloss - Return loss
RN - Normalized resistance
VOL Voltage reflection coefficient magnitude
VSWR Voltage Standing Wave Ratio
XN - Normalize reactance
Z0 Characteristic impedance
ZN Normalized impedance
g - Reflection coefficient at the generator/source
L - Reflection coefficient at the load

CHAPTER-9 PROBLEMS
9.1)

9.2)
9.3)

What is a standard Smith chart? What range of resistor and

reactive
values is mapped into a standard Smith chart?
What resistor values get mapped into a compressed Smith
chart? Show by drawing a diagram.
A lossless transmission line is connected to a load
ZL=100+j100 . Using a Smith chart:
a) Determine the reflection coefficient at the load

210

RF & Microwave Design Essentials

9.4)

9.5)

b) Calculate the return loss.

c) Find the VSWR on the line.
d) Determine the reflection coefficient and the input
impedance /8 away from the load.
Using a smith chart find ZL for:
a) =0.6 ej45
b) =-0.3
c) =0.4+j0.4
VSWR on a lossless transmission line (ZO=50 ) is
measured to be 3.0. Using a Smith chart determine:
a) The magnitude of the reflection coefficient in "ratio" and
in "dB".
b) The return loss in dB.
c) The mismatch loss in dB
d) If the load is resistive(R>ZO) and is located 3/8 away
from the source, determine the load impedance value, the
input impedance of the transmission line and the reflection
coefficients at the load and at the source (see Figure P9.5).

Figure P9.5

REFERENCES
[9.1] Cheng, D. K. Field and Wave Electromagnetics, 2 nd
ed., Reading: Addison Wesley, 1989.
[9.2] Cheung, W. S. and F. H. Levien. Microwave Made Simple,
Principles and Applications. Norwood: Artech House,
1985.

Matthew M. Radmanesh, Ph.D. 211

[9.3] Ginzton, E. L. Microwave Measurements, New York:

McGraw-Hill, 1957.
[9.4] Gonzalez, G. Microwave Transistor Amplifiers,
Analysis and Design, 2 nd ed. Upper Saddle River:
Prentice Hall, 1997.
[9.5] Kosow, I. W. and Hewlett-Packard Engineering Staff,
Microwave Theory and Measurement. Englewood Cliffs:
Prentice Hall, 1962.
[9.6] Reich, H. J., F. O. Phillip, H. L. Krauss, and J. G. Skalnik,
Microwave Theory and Techniques, New York: D. Van
Norstrand Company, Inc., 1953.
[9.7] Schwarz, S. E. Electromagnetics for Engineers, Orlando:
Saunders College Publishing, 1990.
[9.8] Smith, P. H. Transmission-Line Calculator, Electronics, 12,
pp 29-31, Jan. 1939.
[9.9] Smith, P. H. An Improved Transmission-Line Calculator,
Electronics, 17, pp 130, Jan. 1944.

RF & Microwave Design Essentials

212

CHAPTER 10

Smith Chart Applications

10.1 INTRODUCTION
The Smith chart applications in the analysis or design of RF and
microwave circuits can be subdivided into three categories:
a. Circuits containing primarily "distributed elements", particularly
transmission lines (TLs).
b. Circuits containing "lumped elements".
c. Circuits containing "Distributed and lumped elements" in
combination.

10.2 DISTRIBUTED CIRCUIT APPLICATIONS

The most common distributed circuit element is a transmission line
(TL) and the Smith chart can be used effectively for calculation of
values of its different parameters. Before we proceed into different
Smith chart applications, it would serve us well, at the outset, if we
define, the following notations which will be used throughout this
book:
Impedance: Z=R+jX (
)
(10.1)
Admittance: Y=G+jB (S)
(10.2)

Matthew M. Radmanesh, Ph.D. 213

The normalized values are given by:

(10.3)
(Z)N=Z/ZO= R/ZO+jX/ZO =r+jx
(Y)N=Y/YO=G/YO +jB/YO=g+jb
(10.4)
Where,
(10.5)
r=R/ZO,
x=X/ZO,
(10.6)
(10.7)
g=G/ZO,
b=B/ZO
(10.8)
(10.9)
YO=1/ZO
And "ZO" is the characteristic impedance of the transmission line or
a reference impedance value.

10.2.1 Application #1:

INPUT IMPEDANCE (ZIN) DETERMINATION USING A KNOWN LOAD
(ZL)
The input impedance (Zin) at any point on a transmission line, a
distance "l" away from the load (ZL), can be calculated by the
following procedure:
a. Plot the normalized load impedance [(ZL)N=ZL/Z0)] on the
Smith chart,
b. Draw the constant VSWR circle that goes through (ZL)N,
c. Starting from (ZL)N, move "toward generator" on the constant
VSWR circle a distance "l/",
d. Read off the normalized input impedance value (Zin/Zo) from
the chart as shown in Figure 10.1.
This process can be reversed easily when the input impedance (Zin)
is known and the load impedance is unknown (ZL). In this case,
starting from (Zin)N, one moves "Toward load" a distance "l/" on
the constant VSWR circle to arrive at (ZL)N.

RF & Microwave Design Essentials

214

______________________________________________________
EXAMPLE 10.1
Find the input impedance of a transmission line (ZO=50 ) that has
a length of /8 and is connected to a load impedance of
ZL=50+j50 ?
Solution:
a. Locate (ZL)N=ZL/ZO=1+j1 on the smith chart.
b. Draw the constant VSWR circle as shown in Figure 10.2.
c. Now move "toward Generator" on the constant VSWR
Circle a distance of /8 (or 90) to obtain:
(Zin)N=2-j1 Zin=ZO(Zin)N=100-j50 .

Matthew M. Radmanesh, Ph.D. 215

_______________________________________________
10.2.2 Application #2
INPUT

IMPEDANCE

DETERMINATION

USING

THE

INPUT

IN|
1)
REFLECTION COEFFICIENT (|
When the reflection coefficient at any point on a transmission line is
known, the input impedance at that point can be calculated as
follows:
a. Locate IN=|IN|ej on the Smith chart; The magnitude of || can
be read off the "Reflection coefficient voltage" radial scale at the
bottom of the chart while "" is read off the circular scale (See
Figure 10.3).
b. Normalized values of resistance and reactance (r,x) can be read
off the Smith chart at point "A", giving Zin as:
Zin =Zo(r+jx)
NOTE 1: If conversely, the input impedance (Zin) is known and the
corresponding reflection coefficient is desired to be found, the
procedure would be as follows:
a. Plot the normalized input impedance (Zin)N on the Smith chart and
read off the angle "" on the circular scale.
b. Draw the constant VSWR circle,
c. The intersection of this circle with the right-hand horizontal axis
is found and dropped off onto the "reflection coef." radial scale at
the bottom and the || value is read off as shown in Figure 10.3.

RF & Microwave Design Essentials

216

NOTE 2: If the value of Zin at a distance l from the reflection

coefficient location is sought, one needs to use the procedure
described in application #1.

10.2.3 Application #3
IMPEDANCE

DETERMINATION USING REFLECTION COEFFICIENT

WHEN |
| > 1

When the magnitude of the reflection coefficient is greater than

unity, the corresponding impedance has a negative resistance value
and thus maps outside the standard Smith chart. In this case, another
type of chart called a compressed Smith chart (as discussed earlier)
should be used. This chart includes the standard Smith chart (|| 1)
and (||>1) region which corresponds to the negative resistance
region.
An alternate way of determining an impedance (Z) having (||>1), is
by using a standard smith chart with the help of the following
procedure:
a. Obtain the complex conjugate of the reflection coefficient at point
"B", (*=||-).
b. Plot 1/* on the standard Smith chart (see point "C" in Figure
10.4)

Matthew M. Radmanesh, Ph.D. 217

c. Read off the normalized impedance value (r+jx, corresponding to

1/*) on the Smith chart.
d. The impedance (Z) value corresponding to is obtained by
negating "r" and keeping "x" intact, i.e.,
Z=ZO(-r+jx)
This procedure can be proven as shown below.
PROOF:
Assuming ZO is a real number, the normalized impedance (Z/ZO)
corresponding to is given by (see point A in Figure 10.4):
Z/ZO= -r+jx, r>0
(10.10)
Where "r" and "x" are normalized values of resistance and reactance,
respectively.
Knowing that =||, we can write:
-r+jx
=(Z-Zo)/(Z+ZO)=(-r+jx-1)/(-r+jx+1)
* =|
|
-

=(r+jx+1)/(r+jx-1)
Thus we have:
1
1/
* =

||
1/
* =(r+jx-1)/(r+jx+1) =(Z'/Zo-1)/(Z'/Zo+1)

(10.11)
(10.12)

(10.13a)
(10.13b)

RF & Microwave Design Essentials

218

Where Z'/Zo=r+jx is the impedance corresponding to 1/*, that is,

1/
* r+jx
(10.14)
From Equation (10.13a) we can see that 1/* (shown at point B in
Figure 10.4) has the same angle as , namely, they are on the same
vector. Therefore from Equations (10.11) and (10.14) we conclude
that the impedance corresponding to is simply obtained by
reversing the sign of the real part of Z', i.e.,
(10.15)
Z=Z'(r,x) (-r,x)
_______________________________________________________
EXAMPLE 10.2
What is the impedance (ZD) of a device having D=2.2326.5 ?
Assume ZO=50 .
Solution:
a. We find D*=2.23-26.5
b. Plot 1/D*=0.44726.5 on the smith chart. From the chart we
obtain:
ZD'=50(2+j1)=100+j50
c. Using ZD' from step (b), we can write ZD as:
ZD= -100+j50
______________________________________________________

10.2.4 Application #4
DETERMINATION OF ADMITTANCE (Y) FROM IMPEDANCE (Z)
As discussed in Chapter 7, we know that the reflection coefficient
[(x)], the normalized input impedance [ZN(x)] and the normalized
input admittance [YN(x)] at any point on the line are given by:
ZN(x)=[1+
(x)]/[1-
(x)]
(10.16)
and,
YN(x)= 1/ZN(x)=[1-
(x)]/[1+
(x)]
(10.17)
Where,
(x)= L ej2X
(10.18)
is the reflection coefficient at any point (x) on the transmission line.
From the expression for (x), we note that for every phase change of
2l= (i.e. every l =/4), (x) changes sign which leads to the

Matthew M. Radmanesh, Ph.D. 219

inversion of the expressions given in (10.17) and (10.18) causing

ZN(x) to become YN(x) and vice versa. This observation indicates
that YN is located 180 degrees opposite to ZN on the VSWR circle as
shown below in Figure 10.5.

______________________________________________________
Example 10.3
Find the admittance value for an impedance value of Z= 50+j50 ,
in a 50 system.
Solution:
ZO=50 YO=1/50=0.02 S
ZN=Z/ZO=1+j1
Using the smith chart, YN can be read off at 180 away on the
constant VSWR circle:
YN=0.5-j0.5
Y=YOYN Y=0.01-j0.01 S
_______________________________________________________
NOTE 1: Z to Y conversion can also be obtained by rotating the Zchart by 180 and super-imposing it on the original chart, which
will give a ZY-Smith chart. The Y-chart has negative susceptance on
the upper half and positive susceptance on the lower half, exactly
opposite of the Z-chart. The Z-Y chart is shown in Figure 10.6.

RF & Microwave Design Essentials

220

NOTE 2: The standard Smith chart may be considered to be a Y- or

Z-chart depending on the first time of entrance of values in it, being
either admittance (Y) or impedance (Z).
NOTE 3: When working with series elements, the concept of
impedance becomes important and we need to use the Z-chart. On
the other hand when working with parallel (or shunt) elements, the
concept of admittance becomes paramount and therefore we switch
to the Y-chart.

10.2.8 Application #8
INPUT IMPEDANCE DETERMINATION USING SINGLE STUBS
DEFINITION-STUB: A stub is defined to be a short section of a
transmission line (usually terminated in either an open or a short)
often connected in parallel and sometimes in series with a feed
transmission line in order to transform the load to a desired value.
In general, the stub can have any general termination (ZL'), however
in practice as explained above, ZL' is either a short or an open circuit
as shown in Figures 10.22, 10.23 and 10.24.

Matthew M. Radmanesh, Ph.D. 221

There are two cases that will be considered separately as follows:

a. Parallel stubs, and
b. Series stubs
a. Parallel (or Shunt) Stubs
Consider the stub located a distance "d" away from a load (ZL) as
shown in Figure 10.22a. We would like to determine the input
impedance of the combination.
Before we proceed to find the input impedance, we need to
determine the stub's susceptance. Since the stub is connected in
parallel, we use the smith chart as a Y-chart. The stub has a length
(l) which can be used to determine its input admittance (or
susceptance).
If the stub is terminated in a short, we use the Y-chart and start from
Y = (see point "A" in Figure 10.23c) and travel "l" toward
generator to arrive at point "B". We read off the stub's susceptance
from the chart. In a similar fashion, an open stub's susceptance can
be found except we should start at Y=0 on the opposite side (see
Figure 10.23 and 10.24).

RF & Microwave Design Essentials

222

Matthew M. Radmanesh, Ph.D. 223

To find the input impedance, the following steps are carried out:
1. Locate ZL on the Smith chart (use a ZY-chart)at point "A" in
Figure 10.25.

2. Draw the constant VSWR circle.

3. Travel a distance (d) toward the generator on the VSWR circle to
arrive at point "B",
4. Now since we are adding the parallel stub, we must switch to the
Y-chart and travel on a constant conductance circle an amount equal
to the susceptance of the stub to arrive at point "C", as shown in
Figure 10.25.
5. To find the input impedance, we switch back to the Z-chart and
read off the normalized values (r,x) at point "C" corresponding to
(Zin)N. The total input impedance is given by:
Zin=ZO(Zin)N
b. Series Stubs
Consider a series stub located a distance (d) away from the load (ZL)
as shown in Figure 10.26. Similar to the parallel stub case, we need
to know the series stub's reactance (jX) based on its electrical length
(l).

RF & Microwave Design Essentials

224

Since the stub is in series, we use the smith chart as a Z-chart. If the
stub is terminated in a short, start from Z = 0 (point "A" in Figure
10.27c) and travel a distance of l/ "toward generator" to arrive at
point "B". Read off the normalized stub's reactance (jx=jX/ZO) from
the chart as shown in Figure 10.27b. Similarly, an open stub's
reactance can be determined by following the above procedure
except by starting from Z= on the chart as shown in Figure 10.27c.

To find the input impedance, the following steps are carried out:

Matthew M. Radmanesh, Ph.D. 225

1. Locate (ZL)N on the Smith chart at point "A" as shown in Figure

10.28 (use a Z-chart).
2. Draw the constant VSWR circle,
3. From (ZL)N, travel a distance (d) toward the generator on the
VSWR circle to arrive at point "B",
4. Now, since we are adding the series stub, we travel on a constant
resistance circle an amount equal to the reactance of the stub, jx, to
arrive at point "C".
5. The input impedance is read off at point "C" in Figure 10.28

_______________________________________________________
EXAMPLE 10.7
Consider a transmission line (ZO=50 ) terminated in a load
ZL=15+j10 as shown in Figure 10.29. Calculate the input
impedance of the line where the shunt open stub is located a
distance of d=0.044 From the load and has a length of l=0.147.

RF & Microwave Design Essentials

226

Solution:
a. The susceptance of the open stub is first calculated by
moving on a smith chart from Y=0 and moving a distance of
0.147 toward generator to arrive at (YOC)N=j1.33 as shown in
Figure 10.30.

Next, the input impedance is found by :

b. Locate(ZL)N=(15+j10)/50=0.3+j0.2 on the smith chart (see
point "A" in Figure 10.31):
c. Draw the constant VSWR circle.
d. From ZL, travel a distance of 0.044 to arrive at point "B".
The admittance is read off to be:
(YB)N=1-j1.33 (point "B" in Figure 10.31)

Matthew M. Radmanesh, Ph.D. 227

e.

Adding an open shunt stub of length l=0.147 with

(YOC)N=j1.33 gives:
(Yin)N=(YB)N+(YOC)N=(1-j1.33)+j1.33=1
(Zin)N=1/(Yin)N=1 Zin=ZO=50
Adding the shunt stub on the smith chart results in arriving
at point "O", which is obtained by moving on r=1 constant
resistance circle by -j1.33.

NOTE: Use of Application #8 in the design of circuits to bring about

reflection-less loads are widely explored in the next chapter, where
matching circuits are treated in depth.
_______________________________________________________
EXAMPLE 10.8
Consider a transmission line (ZO=50 ) with ZL=100 as shown in
Figure 10.32. Calculate the input impedance of the line where the
shorted series stub is located a distance of d=/4 from the load and
has a length l=/8.

Solution:
a. The reactance of the series shorted stub is first calculated by
moving on a smith chart from Z=0 a distance of 0.125 toward

RF & Microwave Design Essentials

228

generator to arrive at (ZSC)N=j1 as shown in Figure 10.33. Next, to

find the input impedance we perform the following steps.

b. Locate(ZL)N=100/50=2 on the smith chart (see point "A" in

Figure 10.34):

c. Draw the constant VSWR circle. From ZL, travel a distance of

0.25 to arrive at point "B". The impedance is read off to be:
(ZB)N=0.5 (at point "B")

Matthew M. Radmanesh, Ph.D. 229

NOTE: Since the load is resistive and has a value more than ZO, the
(ZL)N value and location corresponds to (Zmax)N (at point "A") and ZB
corresponds to (Zmin)N (for more details, see application #5).
d. From point "B", move toward generator on a constant resistance
circle to 0.5+j1 (point "C" in Figure10.34) which corresponds to
adding a series stub of length l=0.125 or (ZSC)N=j1, giving:
(Zin)N=(ZB)N+(ZSC)N=0.5+j1
Zin=ZO(Zin)N =25+j50

10.3 LUMPED ELEMENT CIRCUIT

APPLICATIONS
Lumped elements are usually employed in the design of microwave
circuits. These elements are mostly lossless reactive elements (such
as inductors or capacitors ) and are added either in series or in
parallel in the circuit.

10.3.1 Application #9
INPUT IMPEDANCE FOR A SERIES LUMPED ELEMENT
Consider the circuit shown in Figure 10.35 where a load (ZL) is in
series with a series element (ZS). The lumped element can be
reactive (lossless), resistive (lossy) or a combination of both. In this
application we consider a very general lumped element consisting of
both resistive and reactive components.

Since the lumped element is in series with the load, we need to

consider only the Z-chart markings of the ZY-Smith chart (or only a
Z-chart), in order to determine Zin. We know mathematically that:
Zin=ZL+ZS
Thus:

230

RF & Microwave Design Essentials

(Zin)N=(rL+rS)+j(xL+xS)

(10.48)

The purpose of this application is to show how to achieve this result

graphically where the exact steps are delineated below:
a. Locate (ZL)N on the Smith chart(see point "A" in Figure 10.36)
b. Moving on the constant resistance circle that passes through ZL,
add a reactance of jxS to arrive at point "B".
c. Now moving on a constant reactance circle that passes through
point "B", add a resistance of rS to arrive at point "C".
d. The input impedance value is read off at point "C", using the Zchart markings.
ALTERNATE PROCEDURE:
Point "C" could have equally been reached by the following steps
(see Figure 10.36):
a. Move on a constant reactance circle (that passes through "A"
and add the resistance of rS to arrive at point "B'" (see Figure
10.36)
b. Now moving on a constant resistance circle (that passes
through point "B'"), add the reactance of jxS to arrive at point
"C".

Matthew M. Radmanesh, Ph.D. 231

10.3.2 Application #10

INPUT ADMITTANCE FOR A SHUNT LUMPED ELEMENT
Consider the circuit shown in Figure 10.37, where a load (YL) is in
parallel with a shunt element (YP). In general, The lumped element
is considered to have both resistive and reactive components (similar
to Application #9).

Since the lumped element is in parallel with the load, only the Ychart markings of the ZY-Smith chart need be considered.
The total admittance is given mathematically by:
Yin=YL+YP
(Yin)N=(gL+gP)+j(bL+bP)
Similar to the application #9, we now present the procedure to
determine (Yin)N graphically:
a. Locate (YL)N on the Y-chart at point "A" in Figure 10.38.

b. Move on the constant conductance circle [that passes through

(YL)N] and add a susceptance of "jbP" to arrive at point "B".
c. Move on the constant susceptance circle (passing through

RF & Microwave Design Essentials

232

"B") by adding a conductance of "gP" to arrive at point "C".

d. The input admittance is read off at point "C" using the Y-chart
markings.
ALTERNATE PROCEDURE:
Similar to Application #9, the input admittance equally could have
been determined by:
a. Moving on a constant susceptance circle and adding gP to arrive at
point "B" as shown in Figure 10.38.
b. Now add jbP on a constant conductance circle to arrive at point
"C".

10.3. 3 Application #11

INPUT IMPEDANCE OF SINGLE SHUNT/SERIES
REACTIVE ELEMENTS
This is a special case of applications #9 and #10 where the series or
the shunt elements are lossless (i.e. purely reactive). In this case
there are 4 possible combinations (see Fig. 10.39) as follows:
1. Series L
2. Series C
3. Shunt L
4. Shunt C

Matthew M. Radmanesh, Ph.D. 233

This case has already been discussed in Chapter 6 under the heading
of "L-network matching" which is now revisited and treated with the
help of the smith chart.
To find the input impedance, we first calculate the normalized series
reactance (jx=jX/ZO) or normalized shunt susceptance value
(jb=jB/YO) of the lumped element before entering the Smith chart.
Next, we locate (ZL)N on the chart as point "A"(see fig. 10.40). Now
starting from point "A", the following steps are applied:
1. To add a series L: on a constant resistance circle, move up by
jxS=jL/ZO.
2. To add a Series C: on a constant resistance circle, move down by
jxS=-j/CZO.
3. To add a shunt L: on a constant conductance circle, move up by
jbP=-j/LYO.
4. To add a shunt C: on a constant conductance circle, move down
by jbP=jC/YO.
These are all shown in Fig. 10.40.

RULE OF THUMB
Upon close observation of these four cases, it appears that for the
majority of load values, adding series (or shunt) inductor would
move point "A" upward on the constant resistance (or conductance)
circle while adding a series (or shunt) capacitance would move point
A downward.

RF & Microwave Design Essentials

234

However, it should be noted that the above is a good rule of thumb

to follow when dealing with purely reactive elements, but should
never be generalized outside the scope of this discussion. This rule
of thumb is limited but workable and will never actually replace the
reasoning and the understanding that goes into making it.
_____________________________________________________
Example 10.9
Calculate the total input admittance of a combination of a load
ZL=50+j50 with a shunt inductor of L=8 nH at fO=1 GHz as
shown in Fig. 10.41. Assume a 50 system.
Solution:
ZO=50 YO=0.02 S
a. We first find the susceptance of the shunt inductor:
jBP=-j/(OL)=-j0.02 S jbP=jBP/YO=-j1

b. Locate (ZL)N=ZL/ZO=1+j1 on the smith chart at point "A" in Fig.

10.42.

c. Since this is an inductor, we need to move upwards from point

"A" on a constant conductance circle by -j1 to arrive at point "B".

Matthew M. Radmanesh, Ph.D. 235

d. The normalized input admittance is read off at point "B" as:

(Yin)N=0.5-j1.5
Yin=YO(Yin)N=0.01-j0.03 S
Or,
Zin=1/Yin=10+j30

10.3.4 Application #12

INPUT IMPEDANCE DETERMINATION FOR ANY
COMBINATION OF SERIES AND SHUNT REACTIVE
ELEMENTS
In this application, we will consider the case where there are several
series and shunt elements in combination with the load (as shown in
Figure 10.43).

Application #11 can be used repeatedly to arrive at the total input

impedance as described in the following steps:
a. Since the first element adjacent to the load is
connected in series, we start with (ZL)N and locate it on the Zchart (see point "A" in Figure 10.44),
b. On the constant resistance circle passing through (ZL)N,
a reactance of jxS1= jXS1/ZO is added to arrive at point "B".
c. Now switching to the Y-chart, we move on the constant
conductance circle and add a susceptance of jbP=jBP1/YO to
arrive at point "C".
d. Since the next element is in series, we switch back to
the Z-chart and move on a constant resistance circle by
adding a reactance of jxS2=jXS2/ZO to arrive at point "D".
e. The final element is in parallel, so we switch to the Y
chart and add a susceptance of jbP2=jBp2/YO to arrive at "E".

RF & Microwave Design Essentials

f.

236

The total impedance is now read off on the Z-chart at

point "E" as shown in Figure 10.44.

_____________________________________________________
Example 10.10
Find the input impedance at f=100 MHz for the circuit shown in
Figure 10.45.

Solution:
First, we choose the normalizing factor arbitrarily to be:
ZO=50 ,
And,
YO=0.02 S.
Then we normalize all impedance and admittance values:
jxS1=(jX1)N=jL1/ZO=j0.36
jbP1(jB1)N=jC1/YO=j0.27
jxS2=(jX2)N=jL2/ZO=j1.0
jbP2=(jB2)N=jC2/YO=j0.38
jxS3=(jX3)N=jL3/ZO=j0.48

Matthew M. Radmanesh, Ph.D. 237

(ZL)N=300/50=6
a) Locate (ZL)N on the smith chart(point "A" in Figure 10.46).

b) Since the first element (L3) adjacent to the load is a series

inductor, we move upward from point "A" on a constant
resistance circle by a reactance of j0.36 to arrive at point "B".
c) Now switch to constant conductance circle, add the next shunt
element by moving downward by j0.27 to arrive at point "C".
d) For the next series inductor, switch to the constant resistance
circle and move upward by j1.0 to arrive at point "D".
e) Next, for the shunt capacitor, switch to a constant conductance
circle and move downward by j0.38 to arrive at point "E".
f) Finally, for the series inductor, switch to the constant
conductance circle and move upward by j0.48 to arrive at point
"F".
g) Now we read off the value of the normalized input impedance at
point "F" as:
(Zin)N=Zin/ZO=0.4-j1.0 Zin=20-j50

10.4.1 Application #14

INPUT IMPEDANCE (OR ADMITTANCE) DETERMINATION FOR A COMBINATION OF DISTRIBUTED AND
LOSSLESS LUMPED ELEMENTS

RF & Microwave Design Essentials

238

This final Application deals with circuits having distributed elements

(such as transmission lines) and lossless lumped elements (such as
capacitors and inductors).
To obtain the input impedance (or admittance) we use the following
two rules:
a. When dealing with distributed elements, for ease and
convenience we start from the load end. Then we travel on a
constant VSWR circle a length (l) towards the generator.
b. When dealing with lossless lumped elements, we also start
from the load end but move on a constant resistance (or
conductance circle) depending on whether the lumped
element is in series (or shunt) with the rest of the circuit.
The overall procedure is the same as delineated in the previous
applications. The example below will illustrate this concept further.
_______________________________________________________
Example 10.13
In the circuit shown below (Figure 10.61), determine the input
impedance at f = 10 GHz.

Solution:
To find Zin we perform the following steps:
a) Locate (ZL)N=100/50=2 on the smith chart (see Figure 10.62).
b) Since the first element adjacent to the load is a series transmission
line we draw the constant VSWR transmission line.
c) Starting from (ZL)N, at point "A", we move on this circle a length
of /4 "toward generator" to arrive at point "B".
d) Now since the next element is a shunt capacitor, we switch to the
Y-chart and move on the constant conductance circle to arrive at
point "C". The shunt capacitor has a susceptance of:
jB=j2 x 1010 x 0.318 x 10-12=j0.02

Matthew M. Radmanesh, Ph.D. 239

e) The next element is a series transmission line, so we switch back

to the Z-chart and draw the constant VSWR circle that passes
through "C".
f) Now from point "C" we move a distance of /8 "toward
generator" to arrive at point "D" as shown in Figure 10.62.
g) The value of the input impedance is read off at point "D" as:
(Zin)N=0.4+j0.55 Zin=20+j27.5

______________________________________________

Chapter 10- Symbol List

A symbol will not be repeated again, once it has been identified and
defined in an earlier chapter, with its definition remaining
unchanged.
b - Normalized susceptance, b=B/Z0
bL - Normalized susceptance at the load
bP - Normalized susceptance at the parallel element
g - Normalized conductance, g=G/Z0
gL - Load Normalized conductance
gP - Shunt Normalized conductance
Imax - Maximum current on a transmission line.
Imin - Minimum current on a transmission line.
lmax location of Zmax on the transmission line
lmin location of Zmin on the transmission line
r - Normalized resitance, r=R/Z0

RF & Microwave Design Essentials

240

x- Normalized reactance, x=X/Z0

Vmax - Maximum voltage on a transmission line.
Vmin - Minimum voltage on a transmission line
Y0 - Characteristic admittance
Z0 - Characteristic impedance
ZD - Device impedance
Zin - Input impedance
(Zin)N - Normalized input impedance
Zmax - Maximum impedance, corresponding to the location of the
peak of the voltage and the valley of the
current in a standing wave pattern on a transmission line.
Zmin - Minimum impedance, corresponding to the location of the
valley of voltage and the peak
of the current in a standing wave pattern on a transmission line.
D -Device reflection coefficient

CHAPTER-10 PROBLEMS
10.1) The normalized impedance of an unknown device is measured
(from 1 to 2 GHz) to have a frequency response as plotted in
Figure P10.1. Determine an equivalent circuit for the
unknown device with all element values correctly calculated
(assume ZO=50 , f=1 GHz).

Figure P10.1

10.2) Using a Smith chart, determine Zin for the circuit shown in
Figure P10.2. Assume fo=1 GHz.

Matthew M. Radmanesh, Ph.D. 241

10.3) A lossless coaxial line (ZO=50 ) is terminated in a

100 load. If the incident voltage wave has an rms
magnitude of 10 V, determine:
a. The reflection coefficient and the VSWR on the line
b. The magnitude and location of VMAX, VMIN, ZMAX and
ZMIN on the line.
c. The magnitude and location of IMAX and IMIN on the line.
d. Determine the power absorbed by the load.
e. Plot the voltage standing wave pattern on the line for both
voltage and current.
10.4) A lossless transmission line (ZO=50 ) is terminated in a load
(ZL=100+j100 ). A single shorted stub (l=/8, 50) is
inserted /4 away from the load as shown in Figure P10.4.
Using a Smith chart, determine the line's input impedance
(ZIN).

10.5) A lossless transmission line (ZO=75 ) is terminated in an

unknown load. Determine the load if the VSWR on the line is
found to be 2 and the adjacent voltage maxima are at x=-15
cm and -35 cm where the load is located at x=0.

RF & Microwave Design Essentials

242

10.6) Determine the input impedance of a transmission line at a

distance of 2m from the load impedance if the wavelength on
the transmission line is found to be 16 cm. What is the
VSWR on the line?
10.7) Determine the input impedance of the lumped-element
network shown in Figure P10.7 (all values are in ).

10.8) Find the input impedance of the transmission line circuit in a

50 system (as shown in Figure P10.8) for ZL=25+j25 ,
d1=3/8, d2=/4 and l=/8. What is the VSWR at the input
terminals?

10.9) Find the input impedance of a double-stub shunt tuner as

shown in Figure P10.9. Assume that the stubs are short circuited
and l1=0.23, l2=0.1, d=/8 and ZO=50 . What is the reflection
coefficient at the input terminals when f=1 GHz?

Matthew M. Radmanesh, Ph.D. 243

REFERENCES
[10.1] Anderson, E. M. Electric Transmission Line Fundamentals.
Reston: Prentice Hall, 1985.
[10.2] Cheng, D. K. Field and Wave Electromagnetics, 2 nd ed.,
Reading: Addison Wesley, 1989.
[10.3] Cheung, W. S. and F. H. Levien. Microwave Made Simple,
Principles and Applications. Norwood: Artech House, 1985.
[10.4] Ginzton, E. L. Microwave Measurements, New York:
McGraw-Hill, 1957.
[10.5] Gonzalez, G. Microwave Transistor Amplifiers, Analysis
and Design, 2 nd ed. Upper Saddle River: P. Hall, 1997.
[10.6] Kosow, I. W. and Hewlett-Packard Engineering Staff.
Microwave Theory and Measurement. Englewood Cliffs:
Prentice Hall, 1962.
[10.7] Liao, S. Y. Microwave Circuit Analysis and Amplifier
Design. Upper Saddle River: Prentice Hall, 1987.
[10.8] Pozar, D. M. Microwave Engineering, 2nd ed. New York:
John Wiley & Sons, 1998.
[10.9] Radmanesh, M. M. The Gateway to Understanding: Electrons to
Waves and Beyond, AuthorHouse, 2005.
[10.10] Radmanesh, M. M. Cracking the Code of Our Physical
Universe, AuthorHouse, 2006.
[10.11] Reich, H. J., F. O. Phillip, H. L. Krauss, and J. G. Skalnik.
Microwave Theory and Techniques, New York: D. Van
Norstrand Company, Inc., 1953.
[10.12] Schwarz, S. E. Electromagnetics for Engineers. New York:
Saunders College Publishing, 1990.
[10.13] Smith, P. H. Transmission-Line Calculator, Electronics, 12, pp.
29-31, Jan. 1939.
[10.14] Smith, P. H. An Improved Transmission-Line Calculator,
Electronics, 17, pp. 130, Jan. 1944.

RF & Microwave Design Essentials

244

PART III
____________________________________

CIRCUIT DESIGN
ESSENTIALS
CHAPTER 11

DESIGN OF MATCHING
NETWORKS

CHAPTER 12

STABILITY IN ACTIVE
NETWORKS

CHAPTER 13

GAIN CONCEPTS IN
AMPLIFIERS

CHAPTER 14

NOISE IN ACTIVE
NETWORKS

CHAPTER 15

AMPLIFIER DESIGN

Matthew M. Radmanesh, Ph.D. 245

RF & Microwave Design Essentials

246

CHAPTER 11

Design of Matching
Networks
11.1 INTRODUCTION
Having studied the Smith chart in full detail and seen the ease and
simplicity that it brings to the analysis of distributed or lumped
element circuits, we now turn to the design of matching networks.
Applications #1 through #14 in Chapter 10 have in reality set the
stage for most of the possible ways a Smith chart could be used as an
essential tool in RF/microwave circuit analysis and more importantly
in network design.
Many of these applications will be cited as reference throughout the
rest of this chapter in order to simplify and further speed up the
process of the design of a matching network, which is an essential
part of any modern active circuit.

11.2 DEFINITION OF IMPEDANCE MATCHING

At the outset of this section, we will define an important
nomenclature:

Matthew M. Radmanesh, Ph.D. 247

DEFINITION- MATCHING: is defined to be connecting two circuits

(source and load) together via a coupling device or network in such
a way that the maximum transfer of energy occurs between the two
circuits.
This is one of the most important design concepts in amplifier and
oscillator design as shown in Figure 11.1.

The concept of Impedance matching (also referred to as tuning) is

the third step in the overall design process (see Chapter 5, Figure
5.9) and is a very important concept at RF/microwave frequencies
since it allows:
a. Maximum power transfer to occur from source to load, and
b. Signal-to-noise ratio to be improved because matching
causes an increase in the signal level.
(a) and (b) are the primary reasons to employ tuning in practically all
RF/microwave active circuit design. To get a conceptual
understanding of why a matching network is needed in a circuit in
general, we can visualize an active circuit in which a load impedance
is different from the transmission line characteristic impedance
causing power reflections back to the source. To alleviate this
problem and bring about zero power reflection from the load (i.e.
maximum power transfer) a matching network needs to be inserted
between the transmission line and the load.
Ideally, the matching network is lossless to prevent further loss of
power to the load. It acts as an intermediary circuit between the two
non-identical impedances in such a way that the feeding

RF & Microwave Design Essentials

248

transmission line sees a perfect match (eliminating all possible

reflections) while the multiple reflections existing between the load
and the matching network will be unseen by the source.

11.3 SELECTION OF A MATCHING NETWORK

Selection of a lossless matching network is always possible as long
as the load impedance is not purely imaginary and has in fact a nonzero real part.
There are many considerations in selecting a matching network
including:
a. Simplicity-The simplest design is usually highly preferable since
simpler matching networks have fewer elements, require less work
to manufacture, are cheaper, are less lossy and more reliable
compared to a more complicated and involved design.
b. Bandwidth--Any matching network can provide zero reflection at
a single frequency, however, to achieve impedance matching over a
frequency band more complex designs need to be used. Thus, there
is a trade-off between design simplicity and matching bandwidth and
eventually the network price as shown in Figure 11.2

c. Feasibility of Manufacturing
To manufacture a certain design, one needs to consider first the type
of transmission line technology that the matching network will be
implemented with. This means that before the matching circuit is
designed one needs to know before-hand whether it is a microstrip

Matthew M. Radmanesh, Ph.D. 249

or coaxial line type technology, So that the matching circuit will be

designed properly to integrate most efficiently with the rest of the
network.
For example, in microstrip line technology due to its planar
configuration, the use of quarter-wave transformers, stubs and chip
lumped elements for matching are feasible. On the other hand, in
waveguide technology, implementing tuning stubs for matching
purposes is more predominant than lumped elements or /4 multisection transformers.
d. Ease of Tunability
Variable loads require variable tuning. Thus the matching network
design and implementation should account for this. To implement
such an adjustable matching network may require a more complex
design or even switching to a different type of transmission line
technology in order to accommodate such a requirement.
These four considerations form the backbone of all design criteria.
However, for the sake of clarity and ease, we will focus only on the
first consideration, i.e., "simplicity", for the rest of this work and
leave the other three considerations to more advanced texts.
NOTE: There are cases where the matching network has two or more
solutions for the same load impedance. The preference of one
design over the other would greatly depend upon bringing the other
three considerations into view, which will place the ensuing
discussions outside the scope of this work.

11.4 THE GOAL OF IMPEDANCE MATCHING

The most important design tool in amplifier and oscillator design is
the concept of impedance matching. The goal of impedance
matching in all of its different forms can be summed up into one
issue, and that is:
GOAL OF IMPEDANCE MATCHING: Making the input impedance of
the load and the added matching network theoretically equal to the

RF & Microwave Design Essentials

250

characteristic impedance of the feeding transmission line, thus

allowing maximum amount of power to transfer to the load.
Next section will delineate conditions under which maximum power
transfer does take place.

11.4.1 Maximum Average Power Transfer

Consider the circuit (shown in Figure 11.3) which is a problem of
great practical importance. In this circuit, source impedance (ZS) is a
known and fixed value, VS is the phasor representation of the source
sinusoidal voltage at angular frequency ():
VS=Re(|VS|ejt)
(11.1)

The problem is to select the load impedance (ZL) such that the
maximum average power (Pav), at steady state, is obtained from the
source and fed to the load. This problem can be easily solved with
the help of the following theorem:
MAXIMUM POWER TRANSFER THEOREM
Consider the general circuit having a known source impedance and
an unknown load as shown in Figure 11.3. The maximum power
transfer theorem states that the maximum power that can be
delivered to a load is only feasible when the load has an optimum
impedance value (ZL)opt equal to the complex conjugate of the
source impedance value (ZS,), i.e.,
(ZL)opt=ZS*
(11.2)
From Equation (11.2) we can see that:
RL=RS
XL=-XS

Matthew M. Radmanesh, Ph.D. 251

Considering the maximum power transfer as the cornerstone of

matching, we can make the following conclusive observation about
the goal of matching:
The goal of matching is adding a matching network to a load
(ZL) in such a way that the input impedance of the total
combination will be located at the center of the Smith chart
(assuming a real source impedance).
PROOF:
The average power delivered to the load can be written as:
VL=ZLIL
VS
IL =
ZS + ZL
Re(Z L )
1
1
2
2
*
(11.3)
Pav = Re VL I L = Re Z L I L = VS
2
2
2
2 ZS + ZL

Where VS, ZS=RS+jXS and ZL=RL+jXL are the source voltage

phasor, source impedance and load impedance, respectively.
Substitution in Equation (11.3) gives:
RL
2
Pav = VS
2
2
2 (R S + R L ) + (X S + X L )

(11.4a)

In Equation (11.4), VS, RS and XS are given, RL and XL are to be

chosen such that their value will maximize Pav. The reactance XL
is found by differentiating Pav with respect to XL and setting it to
zero, which yields:
Pav/XL=0

XL=-XS

(11.4b)

With this choice, the term (XL+XS)2 in the denominator becomes

zero, which minimizes the denominator and maximizes the
expression with respect to XL. Thus Equation (11.4) can now be
written as:
RL
2
Pav = VS
(11.5)
2
2(R S + R L )

RF & Microwave Design Essentials

252

Now to determine optimum RL, we set the partial derivative of Pav

with respect to RL equal to zero, i.e.,
Pav
=0
(11.6)
R L
Upon differentiation and setting it equal to zero, we obtain:
RL=RS Q.E.D.
(11.7)
Using ZL=ZS* (referred to as a conjugately matched load), from
Equation (11.4) we obtain the maximum power delivered to the load
(Pav)max as:

(Pav )max =

VS

(11.8)

8R S

Furthermore, under these conditions the power produced by the

source is given by:
2
VS 2
1
1 VS
*
=
PS = Re VS I L = Re
(11.9)
2
2 Z S + Z L 4R S

Thus we have:

VS

= 2(Pav )max
(11.10a)
4R S
From Equation (11.10), we can observe that the efficiency of a
conjugately matched load is 50%, i.e.
(Pav )max
= 0.5 = 50%
for ZL=ZS*
(11.10b)
PS
PS =

NOTE 1: For RF and microwave engineers this fact (i.e. 50%

efficiency) is of much significance since the energy in the incoming
electromagnetic waves would have been lost if it were not absorbed
by a conjugately matched load (which is the first or "front-end
stage" of a receiver).

Matthew M. Radmanesh, Ph.D. 253

NOTE 2: For power engineers and electric power companies, this

situation is never allowed to occur and in fact the reverse is desired.
This is because they are extremely interested in efficiency and want
to deliver as much of the average power as possible to the load (i.e.
the customer). Thus, huge power generators are never conjugately
matched.
NOTE 3: The "maximum power transfer theorem" assumes that the
source impedance (ZS) is a fixed and known quantity while the load
impedance is a variable and unknown quantity (ZL), whose value
can be varied to the complex conjugate of ZS, to achieve a maximum
power transfer. Under this condition due to the complex conjugate
condition, the total resistance in the circuit is given by:
Ztot= ZS+ZL =2RS=2RL
NOTE 4: If the reverse is true, namely the load impedance (ZL) is a
known and fixed quantity and the source impedance (ZS) a variable
quantity, the requirement that source and load impedances be
complex conjugate of each other is no longer valid and does not
apply in this case.
Furthermore, to obtain maximum power transfer from the source to
the load for this case, it can easily be observed that we need to have
minimum loss in the source. Thus, we can write the following for the
source:
RS=0
XS=-XL
Under this condition, since the reactances cancel out, the total
resistance in the circuit is given by:
Ztot= ZS+ZL=RL

11.5 DESIGN OF MATCHING CIRCUITS USING

LUMPED ELEMENTS
Considering the size of most modern RF/microwave circuits, actual
discrete lumped element capacitors and inductors are used in the

RF & Microwave Design Essentials

254

design process at low RF/microwave frequencies (up to around 1-2

GHz) or at higher frequencies (up to 60 GHz) if the circuit size is
much smaller than the wavelength (l</10)
Although microwave integrated circuit (MIC) technology has
pushed the frequency limitation of lumped elements into the high
microwave range, there are a large number of circuits whose size has
become comparable with the signal wavelength at higher frequency
ranges where using lumped elements would become completely
impractical.
Thus one of the biggest limitations of the use of lumped elements is
in circuits whose size has become comparable with the signal
wavelength.
Furthermore, if the length (l) of the lumped component is below
(/10) as mentioned above, then they can be used in hybrid or
"Monolithic" MICs at frequencies up to 60 GHz. At these high
frequencies, electrical elements can be realized
via several methods. These methods for each component can be
summarized as follows:
1. CAPACITORS:
a. A single-gap capacitor (C<0.5 pF)
b. An inter-digital gap capacitor in a microstrip line (C<0.5 pF)
c. A short or open transmission line stub (C<0.1 pF)
d. A chip capacitor
e. A metal-insulator-metal(MIM) capacitor (C<25 pF)
These are shown in Figure 11.4.

Matthew M. Radmanesh, Ph.D. 255

2. RESISTORS:
a. Thin film technology using NiChrome or doped semiconductor
material
b. Chip resistor
These are shown in Figures 11.5 (a) and (b).

RF & Microwave Design Essentials

256

3. Inductors
a. A loop of a transmission line
b. A short length of a transmission line
c. A spiral inductor using an air bridge
These are shown in Figures 11.6 (a), (b), and (c).

11.5.1 Matching Network Design Using L-Sections

As already discussed in Chapter 6, the simplest type of matching
network is the L(ell)- section consisting of two reactive elements

Matthew M. Radmanesh, Ph.D. 257

that match a load to a transmission line. The actual configuration is

of the form of an inverted L. The two possibilities are shown in
Figure 11.7.

Considering the fact that either of the two reactive elements can be
an inductor or a capacitor, circuit configurations L1 and L2 provide
a total of eight different possibilities for a given load. The location
of the load on the "smith chart" determines the useful configuration
as discussed in the next section.

11.5.2 Design Based on the Load Location

Depending on the location of the normalized load impedance on the
Smith chart, one or both of the two configurations L1 and L2 may
become practical.
The location of the load becomes crucial in the choice of circuit
configuration for the purpose of matching. The load location can
have three distinct possibilities:
CASE I. THE LOAD IS LOCATED INSIDE THE (1+jx) CIRCLE
(RESISTANCE UNITY CIRCLE)
In this case, we can see from Figure 11.8 that the first element has to
be a shunt element thus configuration L2 is the only practical one.

RF & Microwave Design Essentials

258

Using configuration L2, two possible solutions exist:

Solution (1): shunt L & series C

Solution (2): shunt C & series L

These are shown in Figure 11.9.

Case II. The load is located inside the (1+ jb) circle (conductance
unity circle)
In this case, from Figure 11.10 we can see that the first element has
to be a series element thus only configuration L1 becomes useful.
Similar to case 1, there are two solutions possible as shown in Figure
11.10:

Matthew M. Radmanesh, Ph.D. 259

Solution (1): series L & shunt C

Solution (2): series C & shunt L

The circuits for these two solutions are shown in Figure 11.11.

CASE III.
CIRCLE

THE LOAD IS LOCATED OUTSIDE THE (1+ jx) AND (1+ jb)

In this case, there are four solutions possible as shown in Figure

11.12. These four possibilities are described below:
Solutions (1) and (2): both require a series element first which
makes configuration L1 the only practical one. Thus the solutions
are:

Solution (1): series C & shunt C

Solution (2): series C & shunt L

RF & Microwave Design Essentials

260

The circuits for the two solutions are shown in Figure 11.13.

Solutions (3) and (4) below, Both require a shunt element inserted
first, which makes configuration L2 useful. These two solutions are:

Solution (3): shunt C & series L

Solution (4): shunt C & series L

These are shown in Figure 11.14.

Matthew M. Radmanesh, Ph.D. 261

11.5.3 Design Flexibility

Considering all three cases (i.e. I, II and III), it appears that case III
has the highest flexibility since when the load is located outside the
unity circle, we have the highest amount of design flexibility to suit
the designers needs.
If the load falls inside any of the unity circles, one may be able to
add a reactive element to the load in such a way as to bring the
combined load to the outside of the unity circle and then take
advantage of the matching possibilities that are available at the
outside of these two unity circles.
______________________________________________________
EXAMPLE 11.1
Consider a load (ZL)N located inside the (1+jb) circle. Discuss the
matching possibilities for this load.
Solution:
This load obviously has two matching possibilities (1 and 2) as
discussed earlier in case II.
Let us add a series "L" to take the load ZL to ZL' outside the (1+jB)
circle as shown in Figure 11.15.

Now since (ZL')N is outside, we have two additional possibilities (3

and 4) which may be more suitable for our design needs.
Selecting solution 3, we can see that the final matching circuit
will have the following three elements as shown in Figure 11.16.

RF & Microwave Design Essentials

262

_______________________________________________________
As discussed in Section 11.3, there seems to be four main
considerations that govern the matching circuit design process i.e.:
simplicity, bandwidth, feasibility of manufacturing and ease of
tunability. These four criteria heavily influence ones decision in the
choice of the matching circuits design.
Therefore, even though from the simplicity point of view solution
(3) shown above seems to be more complex than a simple L-design,
there are instances where the other three considerations would
become Paramount and thus make this design a valuable one.

11.5.4 Design Rules for Matching Networks--Lumped

Elements
Based on the discussion presented in the previous two sections, there
are certain rules that if followed would simplify and even speed up
the matching circuit design process. These rules can be summarized
as follows:
Rule #1. Use a ZY Smith chart at all times.
Rule #2. Always start off from the load end and travel "toward
Generator" (in order to prevent uncertainty and confusion
about the starting point).

Matthew M. Radmanesh, Ph.D. 263

Rule #3. Always move on a constant-R or constant-G circle in such

a way as to arrive eventually at the center of the Smith
chart.
Rule #4. Each motion along a constant-R or constant-G circle gives
the value of a reactive element.
Rule #5. Moving on a constant-R circle yields series reactive
elements, whereas moving on a constant-G circle yields
shunt reactive elements.
Rule#6. The direction of travel (or motion) on a constant-R or
constant-G circle determines the type of element to be
used, i.e. a capacitor or an inductor.
Rules 5 and 6 lead to the following additional two rules.
Rule #7. When the motion is upward, in most cases it corresponds to
a series or a shunt inductor.
Rule #8. When the motion is downward, in most cases it
corresponds to a series or a shunt capacitor.
NOTE: These rules are merely a guideline to be followed in the
matching circuit design process. They will never replace the
theoretical and practical understandings that goes into making
them. These understandings as contained in this work, are the
essentials from which all of these rules have been derived.
_______________________________________________________
EXAMPLE 11.2
Given the circuit shown in Figure 11.17, design a lumped element
matching network at 1 GHz that would transform
ZL=10+j10 into a 50 transmission line.

RF & Microwave Design Essentials

264

Solution:
a. ZO=50 , YO=0.02 S
(ZL)N=(10+j10)/50=0.2+j0.2
b. Locate (ZL)N on the smith chart as shown in Figure 11.18.

c. Since (ZL)N is inside the unity conductance circle, this would

correspond to case II and has two possible solutions.
SOLUTION (1):
Start from load on a constant-R circle and move up from point "A"
to "B". This Yields Series L:
jL=j0.2 x 50 L=1.59 nH
-Now starting from point "B", a motion downward on the unity
conductance circle yields a shunt C:
jC=j2.0 x 0.2 C=6.37 pF
The final circuit schematic is shown in Figure 11.19.

Matthew M. Radmanesh, Ph.D. 265

SOLUTION (2):
Starting from load, move downward on a constant-R circle
(series C) to point "B'" and the upward (shunt L) to arrive at "C" s
follows:
Series C 1/jC=-j0.6 X 50 C=5.3 pF
Shunt L 1/jL =-j2 x 0.02 L=3.98 nH
The schematic for solution (2) is shown in Figure 11.20.

11.6 MATCHING NETWORK DESIGN USING

DISTRIBUTED ELEMENTS
At higher frequencies where the component or circuit size is
comparable with wavelength, distributed components may be used
to match the load to the transmission line.
The most common technique in this type of design is the use of a
single open-circuited or short-circuited length of transmission line
(called a stub) connected either in parallel or in series with the
transmission feed line at a certain distance from the load as shown in
Figure 11.21.

RF & Microwave Design Essentials

266

Rather than using a two-wire transmission line schematic, alternate

microstrip line schematic for short- and open-stubs can be drawn
more effectively as shown in Figure 11.22 (a),(b),(c) and (d).

Matthew M. Radmanesh, Ph.D. 267

Such a matching network is easy to build using microstrip or

stripline technology. In single-stub matching networks, the two
variable parameters are the distance "d" (from load to stub) and the
length "l" (stub's length) which provides the value of stub
susceptance or reactance.
Selection of distance "d" is crucial for both shunt and series stub as
explained below:
a. For the shunt stub case, "d" should be chosen such that the input
admittance YP (seen looking into the line before adding the stub) is
of the form YP = YO + jB with the stub susceptance selected as (-jB)
resulting in a matched condition.
b. for the series stub case, "d" should be chosen such that the
impedance ZS (seen looking into the line before addition of the stub)

RF & Microwave Design Essentials

268

is of the form ZS = ZO+ jX with the stub reactance selected as -jX,

resulting in a matched condition.

11.6.1 Choice of Short- or Open-Circuited Stubs

With a /4 difference in length between the two, a short or open

transmission line with proper length, can provide any value of
reactance or susceptance needed for the design.
Structural considerations behind choice of a short- versus an openstub are as follows:
a. OPEN STUBS: For microstrip and stripline technology use of open
stubs are preferred. Use of short stubs require a via-hole through the
substrate to the ground plane, which adds extra work and can be
eliminated through the use of open circuits.
b. SHORT STUBS: For coaxial line or waveguide as a transmission
line media, use of short stubs are preferred because the open stubs
may radiate causing power losses thus making the stub no longer a
purely reactive element.

11.6.2 Stub Realization Using Microstrip Lines

Series transmission lines and shunt stubs (short or open) can easily
be realized using design steps for microstrip line technology (as
outlined in Chapter 7).
Given a dielectric constant (r), its height (h) and a certain
characteristic impedance value (ZO), the width of the microstrip line
(W) can be calculated.
For example, a series transmission line and an open shunt stub
schematic as well as its microstrip realization is shown in Figure
11.23.

Matthew M. Radmanesh, Ph.D. 269

To minimize the microstrip transition interaction and improve the

input VSWR, many designers use a balanced approach for shunt
stubs rather than a single stub. Using balanced shunt stubs technique,
two stubs of the same length (as the single stub) but twice the
characteristic impedance are placed in parallel as shown in Figure
11.24. That is,
l2'=ll2

(11.11)

ZO2'=2ZO2

(11.12)

The reason we use twice the characteristic impedance for each open
shunt stub is due to the fact that each half of the balanced stub (Ystub)
must provide half the total admittance (Ytot), i.e.,
Ystub= jYO2' tan
l2'=
Where
Ytot=j YO2 tanl2

1
Ytot
2

(11.13)
(11.14)

RF & Microwave Design Essentials

Thus substituting Equation (11.14) in (11.13), we have:

1
YO2' tan
l2'= YO2 tanl2
2
'
Choosing l2 =l2 yields:
1
YO2' = YO2
2
Or,
ZO2'=2ZO2
A similar discussion applies to a short shunt stub with
conclusions.

270

(11.15)

(11.16)
(11.17)
the same

11.6.3 Design Steps for Single Stub Matching (Using

The Same Characteristic Impedance)
The Smith chart can be used effectively to design the distance "d" of
the stub to the load and the length "l" of the stub to create the proper

Matthew M. Radmanesh, Ph.D. 271

value of susceptance or reactance (for more details see application

#8, Chapter 10).
There are two circuit configurations where the stub is either in
parallel or in series. Each case needs to be treated separately as
follows:
a. PARALLEL STUB DESIGN
This is shown in Figure 11.23. The design process has the following
steps:
Step 1. plot (YL)N on the ZY chart
(Please note that a single Y-chart could also be used as well but the
load impedance has to be inverted first)
Step 2. draw the appropriate VSWR circle which goes through (ZL)N
Step 3. On the VSWR circle, move (toward generator) to intersect
the (1+jb) conductance unity circle at two solutions located at points
"A1" and "A2" as shown in Figure 11.23:

Solution #1: Y1 = 1+ jb1 (distance d1)

Solution #2: Y2 = 1+ jb2 (distance d2)

Step 4. Now add a shunt susceptance of either -jb1(solution #1) or

-jb2 (solution #2) to arrive at the center of the chart.
Step 5. to determine lengths l1 and l2, we first locate -jb1 and -jb2 on
the Smith chart as shown in Figure 11.24a.
Then starting from Z = (for open stubs) or Z = 0 (for short stubs),
we travel along the outer edge of the chart "toward generator" to
arrive at -jb1 (for open) or -jb2 (for short) stubs. The lengths can be
read off on the circular scale on the outer edge of the chart. These
are shown in Figures 11.24b and 11.24c.
_______________________________________________________
EXAMPLE 11.3 (PARALLEL STUB DESIGN)
Design a matching network using a single shunt open stub as a
tuning element to match a load impedance ZL=15+j10 to a 50
transmission line(see Figure 11.25)

RF & Microwave Design Essentials

272

Solution:
a. Plot (ZL)N=(15+j10)/50=0.3+j0.2 on the ZY-chart
b. Draw the VSWR circle through (ZL)N as shown in Figure 11.26.

c. Move toward generator to meet the (1+jb) circle at two points,

giving two solutions:
(Y1)N=1-j1.33 with d1=0.044 and -jb1=j1.33
(Y2)N=1+j1.33 with d2=0.387 and -jb2=-j1.33
d. From Figure 11.27 we can read off l1 and l2 as:
l1=0.147
l2=0.353

Matthew M. Radmanesh, Ph.D. 273

The two possible design schematics are shown in Figures

11.28a,b.

_______________________________________________________
NOTE: Design #1 is more desirable Since it is usually preferred to
keep the matching stub as close as possible to the load (i.e.
smaller "d") in order to improve:

The sensitivity of the matching network to frequency
and thus providing a larger bandwidth,

The standing wave losses occurring on the line
between the stub and the load possibly due to a high
VSWR.
b. SERIES STUB DESIGN
From Figure 11.29, the design steps are as follows:
Step 1. Plot (ZL)N on the ZY-chart.

RF & Microwave Design Essentials

274

Step 2. Draw the appropriate VSWR circle.

Step 3. Move "toward generator" to intersect the (1+jx), unity
resistance circle, at two points (1) and (2) as shown in Figure 11.29:

Solution #1: Z1 = 1+ jx1 (distance d1), Z1stub=-jx1

Solution #2: Z2 = 1+ jx2 (distance d2) Z2stub=-jx2

Step 4. Now add a series reactance of either -jx1 for solution#1 (or jx2 for solution #2) to cancel the existing reactance.
Step 5. stub lengths l1 (or l2) is now calculated by first locating
reactance -jx1 (or -jx2) on the Smith chart (see Figures 11.30a,b).
For a series open stub, start from Z = and travel on the outer edge
of the chart "Toward generator" to arrive at -jx1 for solution #1 (or jx2 for solution #2). On the other hand, for a series short stub, repeat
the above procedure except start from Z = 0 as shown in Figures
11.30a,b.

Matthew M. Radmanesh, Ph.D. 275

_______________________________________________________
EXAMPLE 11.4 (SERIES STUB DESIGN)
Using a single series open stub, design a matching network that will
transform a load impedance ZL=100+j80 to a 50 Feed
transmission line as shown in Figure, 11.31.

Solution:
a. Plot (ZL)N=(100+j80)/50=2+j1.3 on the smith chart.
b. Draw the VSWR circle (see Figure 11.32)
c. Move "toward generator" to intersect (1+jx) circle at two points
(1) and (2) giving:
Z1=1-j1.33 with d1=0.120 and Z1stub=j1.33
Z2=1+j1.33 with d2=0.463 and Z2stub=-j1.33

RF & Microwave Design Essentials

276

d. Lengths l1 and l2 can be read off from Figure 11.33 as:

l1=0.397
l2=0.103

The two possible design schematics are shown in Figures 11.34a,b.

Matthew M. Radmanesh, Ph.D. 277

11.6.5 Generalized impedance matching network for

non-zero reflection coefficient
When the input and the output reflection coefficients are both nonzero, then the matching network would transform the load
impedance to a point (other than the center of the chart)
corresponding to the desired input reflection coefficient.
This case is a very generalized concept of a matching network and
is in contrast with the case where either the input or the output
reflection coefficient was at the center of the chart.
This type of matching network could occur for example in an
intermediate matching stage of a two-stage amplifier. Figure 11.41
show the generalized concept of a matching network along with the
plot of the input and output reflection coefficients in the Smith chart.

RF & Microwave Design Essentials

278

There are a number of possible solutions which would lead to the

desired matching network.
a. SOLUTION #1:
Assuming the input and output reflection coefficients to be 1 and
2, respectively, an obvious solution is to convert the load (2) to 50
first and then match 1 to 50 , which is located at the center of
the chart as shown in Figure 11.42a.
To realize the matching circuit, we start from the load at point A
and travel toward generator on the constant VSWR circle to arrive
at point B. Adding an opposite shunt stub
at this point will move this point to the center of the Smith chart
(2'=0) which is the new load at point "O" as shown.
Now starting from point "O", add a series stub (to arrive at point
"C") and then a series transmission line to end up at 1 which is the
input reflection coefficient. The resulting distributed circuit is shown
in Figure 11.42b.

Matthew M. Radmanesh, Ph.D. 279

NOTE 1: As described earlier, it is best to keep the matching

philosophy of moving from the load impedance toward the center of
the chart which means to move from the output (load) and progress
backward to the input end.
NOTE 2: The matching network above could have been alternately
realized with lumped elements using the constant conductance or
resistance circle (as described in the "Lumped element design"
section) rather than the distributed element design.
b. SOLUTION #2
First we draw the constant VSWR circle for (1) as shown in Figure
11.43. Then starting from (2) at point "A", we travel on a constant
conductance circle to intersect the (1) VSWR circle at point B.
This corresponds to a shunt stub element.

RF & Microwave Design Essentials

280

Next, we travel on the constant VSWR circle "toward generator "

(clockwise) to arrive at point "C". This would correspond to a series
transmission line of length "l", as shown in the schematic in Figure
11.44.

NOTE 1: There is a second solution which uses the second

intersection of the constant-G circle with the constant VSWR circle
at point B' leading to a similar design procedure as described above.
NOTE 2: Solutions #2 provide only two elements (compared to four
elements given by solution #1 which usually results in the best
bandwidth for the interstage design.
POINT OF INTEREST: In some cases it is more convenient to work
with the equivalent problem of conjugate reflection coefficient which
functionally yields an equivalent circuit as shown in Figures 11.45
and 11.46.

Matthew M. Radmanesh, Ph.D. 281

In Figure 11.47, since one is matching 1* to 2*, we have to start

from 1* (as the load) and progress backwards to 2* at the other
end. Traveling from point "A" to "B" gives a series transmission
line, followed by going from "B" to "C" (producing a shunt stub),
which is identical to what was obtained earlier.

RF & Microwave Design Essentials

282

Chapter 11- Symbol list

A symbol will not be repeated again, once it has been identified and
defined in an earlier chapter, with its definition remaining
unchanged.
Pav - Average power
(Pav)max - Maximum average power
(ZL)opt - Optimum load impedance

CHAPTER-11 PROBLEMS
11.1) Design a single stub matching circuit (see Figure P11.1) that
transforms a load (ZL=30+j50 ) to a transmission line as
follows (use a 100 system) :
a) Assume ZO1=ZO2=100 .
b) Assume ZO1=100 and ZO2=200

Matthew M. Radmanesh, Ph.D. 283

11.2) Design a lumped matching network to match the load YL=(4j6)x10-3 S to a transmission line (ZO=100 ). Find the
element values at 10 GHz.

11.3) Design the matching network shown in Figure P11.3 to match

the load, ZL=100+j100 to a 50 transmission line.

11.4) Design a matching network to transform a load impedance

(ZL=50+j50) to the input impedance (Zin=25-j25 ) at 1
GHz as shown in Figure P11.4 for the following two cases:
a. Lumped element design,
b. Distributed element design.

11.5) Design a matching network that will match a 50 load to an

input reflection coefficient of in=0.5150 in a 50 system
as shown in Figure P11.5. The matching network should use a
quarter-wave transformer.

RF & Microwave Design Essentials

284

11.6) In the circuit shown below, a load ZL=90+j60 is to be

matched to a line as shown in Figure P11.6. Determine Z1, ZO1 and
l.

11.7) A certain microwave device has Zd=50-j50 . Design a

matching
network to match the device impedance to a 25 system for:
a) Lumped element circuit design.
b) Distributed element circuit design.

11.8) A lossless transmission line (ZO=50 ) is to be matched to a

load,
ZL=5.5-j10.5, by means of a short circuited stub(ZO1=100 )
as shown in Figure P11.8. Determine the position and length
of the stub.

REFERENCES
[11.1] Anderson, E. M. Electric Transmission Line
Fundamentals. Upper Saddle River: Prentice Hall, 1985.

Matthew M. Radmanesh, Ph.D. 285

[11.2] Cheng, D. K. Fundamentals of Engineering

Electromagnetics. Reading: Addison Wesley, 1993.
[11.3] Cheung, W. S. and F. H. Levien. Microwave Made Simple,
Principles and Applications. Norwood: Artech House, 1985.
[11.4] Gonzalez, G. Microwave Transistor Amplifiers, Analysis
and Design, 2 nd ed. Upper Saddle River: Prentice Hall,
1997.
[11.5] Liao, S. Y. Microwave Circuit Analysis and Amplifier
Design. Upper Saddle River: Prentice Hall, 1987.
[11.6] Pozar, D. M. Microwave Engineering, 2nd ed. New York:
John Wiley & Sons, 1998.
[11.7] Schwarz, S. E. Electromagnetics for Engineers. Orlando:
Sanders College Publishing, 1990.

RF & Microwave Design Essentials

286

CHAPTER 12

Stability in Active
Networks
12.1 INTRODUCTION
In any amplifier design, one of the very important considerations is
the stability of the circuit under different source or load conditions.
This term needs to be defined at this point:
DEFINITION- STABILITY: Is defined to be in general, the ability of an
amplifier to maintain effectiveness in its nominal operating
characteristics in spite of large changes in the environment such as
physical temperature, signal frequency, source or load conditions,
etc.
In this chapter, the stability requirements for an amplifier circuit or a
more general two-Port network with known S-parameters will be
discussed (see Figure 12.1). Furthermore, we will limit our stability
considerations primarily to source and load conditions and will
develop exact criteria for unconditional stability as well as
conditional stability (also called potentially unstable condition).

Matthew M. Radmanesh, Ph.D. 287

When an amplifier becomes unstable (i.e. not able to maintain its

nominal characteristics), it no longer acts as an amplifier but an
oscillator.
From Figure 12.1, we can observe that since in and out depend on
the source and load matching networks, therefore the stability of the
amplifier circuit depends on S and L.
It should be noted that since the S-parameters of a two-Port network
is frequency dependent, thus the stability condition of a circuit
depends on the frequency of operation. Thus it is possible to have a
circuit functioning well as an amplifier at an intended frequency
while possibly oscillating for out of band frequencies.
12.1.1 ANALYSIS
In a two-Port network (see Figure 12.2), when the input impedance
or output impedance presents a negative resistance, i.e.,
Re(Zin)<0
(12.1)
Or,
Re(Zout)<0
(12.2)

This negative resistance condition at the input (or the output) Port
means that the reflected signal from the input (or output) port has

RF & Microwave Design Essentials

288

higher power than the incident signal thus making |in| or |out| larger
than unity, i.e.,
|
in|>1
(12.3)
Or,
|
out|>1
(12.4)
Thus considering a passive source or load impedance where:
|S|<1, and |L|<1, two types of "stability conditions" can be defined
as follows:
1. Unconditional stability: a network is said to be "Unconditionally
stable" in a frequency range if, and only if:
|
in|<1,
(12.5a)
and
for all |
S|<1 and |
L|<1
(12.5b)
|
out|<1
2. Conditional stability: a network is set to be "conditionally
stable" or "Potentially unstable", in a frequency range, if
|
in|<1
(12.6a)
and,
|
out|<1
(12.6b)
only for a limited range of values of passive source and load
impedances (or |S| and |L|), but not for all values.
Using signal flow graphs from earlier chapters, in and out are
derived and are given by:
S S
in = S 11 + 12 21 L
(12.7)
1 S 22 L
S S
out = S 22 + 12 21 S
(12.8)
1 S 11 S

A SPECIAL CASE--UNILATERAL TRANSISTOR

If the transistor is unilateral, i.e. S12=0, then the equations for
unconditional stability simplifies to the following:
|S11|<1
(12.9a)
|S22|<1
(12.9b)

Matthew M. Radmanesh, Ph.D. 289

If the above condition can not be met by the transistor, then the
amplifier circuit is considered to be conditionally stable for a certain
range of S and L which will be discussed in more detail shortly.

12. 2 STABILITY CIRCLES

As discussed earlier, the unconditional stability for a general twoPort network having passive source/load impedances (as shown in
Figure 12.3) requires the conditions as given by equations (12.5a,b).

However, the boundaries between stable and potentially unstable

regions of S and L are determined by replacing the inequality
signs in (12.5) with equality signs, giving:
S S
in = S 11 + 12 21 L =1
(12.10a)
1 S 22 L
out = S 22 +

S 12 S 21 S
=1
1 S 11 S

|
S|<1
|
L|<1

(12.10b)
(12.10c)
(12.10d)

12.3 GRAPHICAL SOLUTION OF STABILITY

CRITERIA

Considering the L and S planes, the loci of points for which |in|=1
and |out|=1 are found to be two circles:
a. Input Stability Circle
b. Output Stability Circle

RF & Microwave Design Essentials

290

These two stability circles define the boundaries between "stable"

and "unstable" regions for different values of S and L. For passive
matching networks these values lie inside the standard Smith chart
(|S|1, |L|1). Thus the intersection of the two stability circles
with the standard smith chart provides the stable and unstable
regions.
The equation for the output stability circle (|in|=1) drawn in the L
plane can be written as (see Equation12.11):
|
L-CL|=RL
(12.12)
where:

(S
=

S 11
CL
DL
RL=|S12S21/DL|
DL=|S22|2-|
|2
=S11S22-S12S21
22

* *

(12.13a)
(12.13b)
(12.13c)
(12.13d)

Similarly, the input stability circle (|out|=1) drawn in the S-plane is

obtained by interchanging S11 for S22 in Equations (12.13) and
(12.13) and is given by:
|
S-CS|=RS
(12.14)
where:

(S
=

S 22
CS
DS
RS=|S12S21/DS|
DS=|S11|2-|
|2
=S11S22-S12S21
11

* *

(12.15a)
(12.15b)
(12.15c)
(12.15d)

NOTE: The general equation of a circle in the plane can be written

as:
|
-C|=R
(12.15e)
Where "C" is a complex number representing the center of the circle
and "R" is a real positive number representing the circle's radius.

Matthew M. Radmanesh, Ph.D. 291

Figures 12.4 and 12.5 illustrate the graphical plot of the input and
output stability circles where the circles divide their respective
planes (S-plane or L-plane) into two regions:

RF & Microwave Design Essentials

292

a.

The first region is the stable region and is characterized by

|out|<1 for input stability circle in the S-plane (or |in|<1 for
output stability circle in the L-plane), and

b.

The second region is the unstable region and is characterized by

|out|>1 for input stability circle (or |in|>1 for output stability
circle)

We now need a method to determine which region corresponds to

the stable region that can be subsequently used for amplifier design.
OBSERVATION: Because the S-parameters are frequency dependent,
therefore the size and position of the input and output stability
circles would change as frequency is varied. This is because the
center and radius of each circle are expressed in terms of the Sparameters.

12.3.1 OUTPUT STABILITY CIRCLE

The output stability circles are plotted in the L plane as shown in

Figure 12.4. If we set L=0 (i.e. ZL=ZO, the center of the chart), then
from (12.7) we have:
|
in|=|S11|
(12.16)
Now if the |S11| of the device is less than unity (i.e. |S11|<1), then that
region of the output stability circle which includes the center of the
smith chart is the "stable region" and the second region would be the
unstable area.
For example, if the stability circle does not include the center of the
smith chart as shown in Figure 12.4, then L values located outside
of the stability circle are in the stable region and L values inside the
stability circle are in the unstable region.
Vice versa, if |S11| of the device is more than unity (i.e. |S11|>1), then
the center of the smith chart is in unstable region. For example, in
Figure 12.4, the Stable region would be inside of the stability circle
and the outside would be the unstable region.

Matthew M. Radmanesh, Ph.D. 293

12.3.2 Input Stability Circle

Similar to the output stability circles, the input stability circles are
plotted in the S plane. If we set S=0 (i.e. the center of the smith
chart), then from (12.8) we have:
(12.17)
|
out|=|S22|
In this case, if |S22| of the device is less than unity (i.e. |S22|<1) then
one region of the circle containing the center of the smith chart is the
stable region and the other region is unstable. For example, the
input stability circle as shown in Figure 12.5 does not include the
center of the Smith chart. Thus outside of the circle is stable and the
inside unstable. Vice versa, for |S22|>1 the outside of the input
stability circle is unstable and the inside stable

12.3.3 Special Case: Unconditional Stability

For |S11|<1 and |S22|<1, the amplifier circuit will be unconditionally
stable when either of the following two conditions hold true:
a. Both stability circles fall completely outside the Smith chart, or
b. Both stability circles completely enclose the smith chart.
Therefore for all passive source and load impedances, the
unconditional stability can be concisely stated in mathematical form
as:
||CL|-RL|>1 for |S11|<1
(12.18a)
||CS|-RS|>1 for |S22|<1
(12.18b)
Equations (12.18) in essence state that the distance between the
center of the Smith Chart and the center of the stability circle must
be larger than the stability circle's radius by an amount equal to the
Smith Charts radius. An example of the input stability circle in the
S plane is shown in Figure 12.6.
NOTE: If |S11|>1 or |S22|>1, then the amplifier can not be
unconditionally stable since at the center of the Smith chart (where
s=0, L=0), we have |in|=|S22|>1 and |out|=|S11|>1 which
obviously is an unstable region.

RF & Microwave Design Essentials

294

12.4 ANALYTICAL SOLUTION OF STABILITY

CRITERIA
The necessary and sufficient conditions for an amplifier to be
unconditionally stable can be mathematically derived from
Equations (12.10). The result of this derivation is a set of
mathematical conditions which can be concisely referred to as "twoor three-parameter test criteria", as follows:

Matthew M. Radmanesh, Ph.D. 295

Let us define the determinant of the S-matrix (), factors K and B1

as:
=S11S22-S12S21,
1 | S 11 | 2 | S 22 | 2 + | | 2
K=
2 | S 12 S 21 |
and,
B1=1+|S11|2-|S22|2-|
|2

(12.19)
(12.20)
(12.21)

Then based on these definitions, a two-Port network will be

unconditionally stable if, and only if, either one of the following
mathematically equivalent criteria are satisfied:
CRITERION #1: THREE-PARAMETER TEST CRITERION
K>1,
(12.22)
and
1 | S 11 | 2
> 1,
(12.23)
| S 12 S 21 |
and
1 | S 22 | 2
>1
(12.24)
| S 12 S 21 |
We can shrink the three parameter test into a two-parameter test as
the "K- Test" below.
CRITERION #2: TWO-PARAMETER TEST CRITERION: (K-
TEST)
K>1
(12.25)
|
|<1
(12.26)
This is often referred to as "K- Test".
CRITERION #3: TWO-PARAMETER TEST CRITERION: (K-B1 TEST)
K>1
(12.27)
B1>0
(12.28)
This may also be called "K-B1 Test".

RF & Microwave Design Essentials

296

These three criteria are mathematically equivalent and if a device

satisfies any one of the three criteria, the other two are automatically
satisfied.
Thus a two-Port network will be unconditionally stable if and only if
any one of the above three criteria are satisfied.
NOTE 1: The two-parameter test criteria ( #2 and #3) are more
popular and more often used than criterion #1, primarily due to its
simplicity and ease of calculations.
NOTE 2: For a unilateral transistor we have:
S12=0 K=
>1
and,
|
|=|S11S22|

(12.29)
(12.30)

Since K>1 has already been satisfied, therefore in order to satisfy

the condition for unconditional stability we desire ||<1, which
requires:
(12.31)
a. |S11|<1
b. |S22|<1
(12.32)
for all passive values of ZS and ZL.
This conclusion is in agreement with the earlier discussion.
_______________________________________________________
EXAMPLE 12.1
Determine the stability of a GaAs FET that has the following Sparameters at 2 GHz in a 50 system both graphically and
mathematically:
S11=0.89-60
S21=3.1123
S12=0.0262
S22=0.78-27
Solution:
a. Graphical method- we calculate the following values:
CL=1.3647, RL=0.5
CS=1.1368, RS=0.2
Input and output stability circles are plotted in Figure 12.7.

Matthew M. Radmanesh, Ph.D. 297

From this Figure we can see that the GaAs FET is "potentially
unstable" and the center of the Smith Chart (s=0, L=0), being
outside of the circles, represents stable regions since at this point we
have:
|in|=|S11|=0.89<1
|out|=|S22|=0.78<1
b. Mathematical method- we calculate the following values:
K=0.6
=0.7-83
Since K<1, the transistor is potentially unstable which is in
agreement with the graphical method.
_______________________________________________________

Criterion 4: Single-Parameter Test (

-Parameter Test)
Considering the more "popular two-parameter test", we can see that
the "two-parameter test" (in particular the "K- test") described
above, provides a set of mathematical conditions on two parameters
for unconditional stability, and only indicates whether a device is
stable or not. However, due to the fact that certain constraints are
imposed on the two parameters, the "two-parameter" test can not be
used to show the degree of stability of one device relative to other
similar devices.

RF & Microwave Design Essentials

298

To determine the unconditional stability of a device as well as its

degree of stability relative to other devices, a new criterion has been
derived that combines the "K- parameters" into a single-parameter
test and is often referred to as the "
-parameter test". The
parameter "" is defined as:
1 | S 11 | 2
=
| S 22 S *11 | + | S 21S 12 |
For unconditional stability, the following must be satisfied:
>1
(12.33)
Furthermore, if "device A" has a parameter "A" which is greater
than "B" corresponding to "device B" i.e.,
(12.34)
A > B
Then "device A" is said to be more stable than "device B".
Equation (12.34) indicates that a device with a larger value of "" is
more desirable for an amplifier design since it implies a greater
degree of stability.

12.5 POTENTIALLY UNSTABLE CASE

Sometimes when s and L are chosen such that:

|
in|>1 or |
out|>|
(12.35)
then the amplifier circuit becomes potentially unstable. In these
situations, the device could be made unconditionally stable if the
total input and output loop resistance is made to be positive, i.e.,
Re(ZS+Zin)>0
(12.36)
Re(ZL+Zout)>0
(12.37)
To achieve a positive loop resistance and thus making a potentially
unstable transistor into conditionally stable one, two methods
normally are employed:
a. Resistively loading the transistor, or
b. Adding negative feedback (not commonly done)

Matthew M. Radmanesh, Ph.D. 299

Use of these techniques brings about a reduction in the gain, an

increase in the noise Figure and a degradation of the amplifier power
output.
These two techniques are useful in broadband potentially unstable
amplifiers where the wide frequency range increases the probability
of unstability. First the resistive loading is used to stabilize the
transistor and then negative feedback is used to provide a relatively
constant gain in low input and output VSWR.
In narrowband amplifiers use of these techniques are not
recommended, instead careful selection of s and L in the early
stages of the design is necessary to ensure a stable amplifier.
_______________________________________________________
EXAMPLE 12.2
A BJT has the following S-parameters:
S11=0.65-95
S21=5.0115
S12=0.03540
S22=0.8-35
Is this transistor unconditionally stable? If not, use resistive loading
to make the transistor conditionally stable. What are the resistor
values?
Solution:
Simple calculations give us:
K=0.547
=0.504-110.4
Since K<1, therefore the transistor is potentially unstable!
To draw the output stability circles, we find the output stability
circle CL and RL from (12.13) as follows:

(S
=

* *

S 11
,
CL
DL
RL=|S12S21/DL|
DL=|S22|2-|
|2
=S11S22-S12S21
CL=1.3
48, RL=0.45
22

RF & Microwave Design Essentials

300

Similarly, for the input stability circle, we calculate CS and RS from

(12.15) as follows:

(S
=

* *

S 22
CS
DS
RS=|S12S21/DS|
DS=|S11|2-|
|2
CS=1.79
122, RS=1.04
These two circles are drawn in Figure 12.8.
11

There are four types of resistive loading possible to improve stability

as shown in Figure 12.9.

Matthew M. Radmanesh, Ph.D. 301

Using the input stability circle, it can be seen that:

a. A series resistor of R1=9 (Figure 12.9a) or,
b. a shunt resistor of R2=71 (Figure 12.9b) at the input of the
transistor will restore the stability.
On the other hand, using the output stability circle, we can see that:
c. A series resistor of R3=43 (Figure 12.9c) or,
d. A shunt resistor of R4=500 (Figure 12.9d) at the output of the
transistor will assure stability.
It should be noted that in most cases, stabilizing either the input or
the output port will restore stability to the transistor. Thus any one
of the four types of resistive loading should be sufficient to create a
stable amplifier.
POINT OF CAUTION: Use of resistive loading at the input of the
transistor (see Figures 12.9a, b) is not recommended due to an
increase in the input loss which translates into a higher noise Figure
at the output of the amplifier. Any resistive loading is preferred to
take place at the output of the amplifier to minimize the increase in
amplifier's noise Figure. This effect will be studied in more depth in
Chapter 14.

RF & Microwave Design Essentials

302

Chapter 12- Symbol List

A symbol will not be repeated again, once it has been identified and
defined in an earlier chapter, with its definition remaining
unchanged.
K - Stability factor used to evaluate the stability of a two-port
network (K- Test)
- Determinant of the S-matrix used to evaluate stability (K- Test)
- A parameter used to evaluate the stability of a network, The test
is called the - parameter test.

CHAPTER-12 PROBLEMS
12.1) In each of the stability circle drawings shown in Figure
P12.1, clearly indicate the possible locations for a stable
source reflection coefficient.
12.2) Output stability circles are shown in Figure P12.2. Determine
the stable region for the load reflection coefficient.
12.3) The scattering parameters for three different transistors are
given below. Determine the stability in each case and in a
potentially unstable case, draw the input and output stability
circles:
a.

0.67 67 0.0756.2
S=

1.74 36.4 0.6 92.6

b.

0.04590
0.385 55
S=
0.89 26.5
2.778

c.

0.7 50 0.2775
S=
0.680
5120

Matthew M. Radmanesh, Ph.D. 303

RF & Microwave Design Essentials

304

12.4) The S parameters of a GaAs FET at a certain Q-point are given

in the following table:

Matthew M. Radmanesh, Ph.D. 305

Draw the input stability circles (at each frequency) in a smith chart
and the output stability circles. Indicate the unstable regions.
12.5) The S parameters of several two-Port networks are given by:
0.70 0.7180
a.
S=

0.70 0.70
b.

0.7 1.7
S=

1.7 0.7

1 0.7
S=

0.7 1
Determine K and ||. Draw the input and output stability circles for
each case as well.
12.6) a. show that in the limit as S12 approaches zero, the center and
radius of the stability circles are given by:
CS1/S22, rS0, CL1/S11 and rL0.
b. The S parameters of a two-Port network are:
0
290
S=
0.145
2
Draw the stability circles and show the unstable regions.
12.7) Show how resistive loading can stabilize a resistor whose Sparameters at f=750 MHz are:
0.6978 0.03341.4
S=
0.8425
5.67123
Consider all four types of resistive loading for this problem.
12.8) A microwave GaAs FET has the following S parameters
measured at VDS=3 V and ID=10 mA at 4 GHz:
0.89 50 0.0666
S=

3.26141 0.58 24
a. Calculate the delta factor ().
b. The stability factor (K).
c. Find the center and radius of the input stability circle and plot the
circle.
d. Determine the center and radius of the output stability circle and
plot the circle.
c.

RF & Microwave Design Essentials

306

12.9) Considering the diagram shown in Figure P12.9, indicate the

possible locations for a stable source impedance and stable load
impedance. The solid circle is the smith chart.

12.10) A microwave transistor has the following S parameters:

5.180
0.8 170
S=
0.62 40
0.370

Matthew M. Radmanesh, Ph.D. 307

Determine the stability, and plot the stability circles if the device is
potentially unstable.

REFERENCES
[12.1] Cheung, W. S. and F. H. Levien. Microwave Made Simple,
Principles and Applications. Norwood: Artech House, 1985.
[12.2] Gonzalez, G. Microwave Transistor Amplifiers, Analysis
and Design, 2 nd ed. Upper Saddle River: Prentice Hall,
1997.
[12.3] Liao, S. Y. Microwave Circuit Analysis and Amplifier
Design. Upper Saddle River: Prentice Hall, 1987.
[12.4] Pozar, D. M. Microwave Engineering, 2nd ed. New York: John
Wiley & Sons, 1998.
[12.5] Schwarz, S. E. Electromagnetics for Engineers. Orlando:
Saunders College Publishing, 1990.
[12.6] Vendelin, George D. Design of Amplifiers and Oscillators by
the S-Parameter Method. New York: John Wiley & Sons,
1982.
[12.7] Vendelin, George D., Anthony M. Pavio, and Ulrich L. Rhode.
Microwave Circuit Design, Using Linear and Non-Linear
Techniques. New York: John Wiley & Sons, 1990.
[12.8] Woods, D. Reappraisal of the Unconditional Stability
Criteria. IEEE Transactions on Circuits and Systems, Feb.
1976.

RF & Microwave Design Essentials

308

CHAPTER 13

Gain Concepts in
Amplifiers
13.1 INTRODUCTION
Gain consideration in an amplifier plays an important role in the
design process. As discussed earlier, the primary consideration in an
amplifier is its stability with its power gain following very closely as
second in importance.

13.2 POWER GAIN CONCEPTS

Consider the single stage microwave transistor amplifier with the
transistor straddled by two matching networks on either side as
shown in Figure 13.1.
There are several power gain concepts that are commonly used in
amplifier design process. Each one of these power gains has a
specific name with a specific definition.

Matthew M. Radmanesh, Ph.D. 309

Therefore at the outset let us first define the various power levels
existing in the circuit as shown in Figure the 13.2.

PIN Power input to the transistor or to the input matching network.

PAVS Power available from the source under matched condition.
This is a special case of PIN when IN= S*.
PL Power delivered to the load or the output matching network.
PAVN Power available from the transistor under matched
condition; a special case of PL when L= OUT*.
Based on these definitions of power, we can now define the
following power gain equations:
P
1. G T L (Transducer power gain)
PAVS
P
2. G P L (Operating power gain, also called power gain)
PIN

RF & Microwave Design Essentials

3. G A

310

PAVN
(Available power gain)
PAVS

Using signal flow graphs and Mason's rule, the three power ratios as
defined above, are derived and the results are stated as follows
1. Transducer Gain (GT)
1- | S | 2
1 | L | 2
2
GT =
|
S
|
(13.1)
21
| 1 IN S | 2
| 1 S 22 L | 2
Equation (13.1) can be written as :
GT=GS.GO.GL
Where
1- | S | 2
GS =
| 1 IN S | 2
GO=|S21|2,
1- | L | 2
GL =
| 1 S 22 L | 2

(13.2)
(13.3)
(13.4)
(13.5)

From Equation (13.2) we may attribute GO to the gain of the

transistor while GS and GL are attributable to the effective gains of
the input and output matching networks.
Alternately, Equation (13.1) can also be written in terms of OUT
(rather than IN) as:
1- | S | 2
1 | L | 2
2
GT =
| S 21 |
(13.6)
| 1 S 11 S | 2
| 1 OUT L | 2
2. Operating power gain (GP)
1 | L | 2
1
2
GP =
|
S
|
21
1 | IN | 2
| 1 S 22 L | 2
3. Available Power gain (GA)

(13.7)

Matthew M. Radmanesh, Ph.D. 311

GA =

1- | S | 2
1
| S 21 | 2
2
| 1 S 11 S |
1 | OUT | 2

(13.8)

Where

IN = S 11 +
OUT = S 22

S 12 S 21 L
1 S 22 L
S S
+ 12 21 S
1 S 11 S

(13.9)
(13.10)

NOTE: From Equation (13.1), the terms represent input and output
matching network's degree of matching to the transistor at its input
or its output. The matching networks are made up of passive
components and have no inherent gain, thus are incapable of
generating power. Nevertheless, since input and output matching
networks are capable of increasing the degree of match in the circuit
as the signal flows through, they can be considered to have a
positive gain in a relative manner. Thus we can write Equation
(13.2) in dB as:
GT(dB)= GS(dB) + GO(dB) + GL(dB)
(13.11)

13.3 A SPECIAL CASE-UNILATERAL

TRANSISTOR
If the transistor is unilateral, i.e. S12=0, then GS and GL gain blocks
as well as IN and OUT simplify into:
GTU=GSU.GO.GLU
(13.12)
Where:

IN=S11 ,

(13.13)

OUT=S22 ,

(13.14)

G SU

1- | S | 2
=
| 1 S 11 S | 2

(13.15)

RF & Microwave Design Essentials

G LU =

1- | L | 2
| 1 S 22 L | 2

312

(13.16)

13.6 MAXIMUM GAIN DESIGN

From (13.4) we can observe that since GO is fixed for any given
transistor, the overall gain of the amplifier is controlled by the gain
blocks GS and GL corresponding to the input and output matching
networks, respectively.
Therefore in order to obtain the maximum possible gain from the
amplifier circuit, we must maximize GS and GL values, which
effectively implies that the input and output matching sections must
provide a conjugate match at the transistor's input and output port.
Furthermore, under this conjugate matched condition at the input
and the output of the transistor, maximum power will be transferred
into the input port and out of the output port of the transistor as
shown in Figure 13.5.
Based on conjugate impedance matching concepts, maximum power
transfer a) from the input matching network to the transistor and, b)
from the transistor to the output matching network, will occur when:
IN =
S*
(13.30)
*
OUT =
L
(13.31)
NOTE: Due to the inherent mismatch between the transistor and the
matching networks (M1,M2) and the fact that the conjugate match
and maximum power transfer will occur theoretically only at one
particular frequency, this type of a circuit design is considered to be
a "narrow-band design".

Matthew M. Radmanesh, Ph.D. 313

13.7 UNILATERAL CASE (MAXIMUM GAIN)

When the transistor is unilateral (i.e. S12=0) then Equations (13.3)
through (13.5) simplify into the following:
(13.32)
IN =S11
OUT =S22
(13.33)
GTU=GS.GO.GL
(13.34)
Where
1- | S | 2
GS =
(13.35)
| 1 IN S | 2
GO=|S21|2,
(13.36)
2
1- | L |
GL =
(13.37)
| 1 S 22 L | 2
Under conjugately matched (maximum gain) conditions:
S =S11*
(13.38)
*
L =S22
(13.39)
GTU,max = GS,max.GO.GL,max
(13.40)
Where
1
(13.41)
G S ,max =
1 | S 11 | 2
GO=|S21|2,
(13.42)
1
G L ,max =
(13.43)
1 | S 22 | 2
Thus we can Equation (13.40) write as:

314

RF & Microwave Design Essentials

G TU ,max = G S ,max G O G L ,max =

1
1 S 11

S 21

1
1 S 22

NOTE: Each gain block (GS or GL) is bound at the lower end by a
gain of zero and at the upper end by the maximum gain (GS,max or
GL,max), as follows:
0
GS GS,max
(13.44)
0
GL GL,max
(13.45)
We can normalize these two equations to obtain:
0
gS 1
0
gL 1
where the normalized gain factors (gS, gL) are defined as:
GS
1 | S | 2
gS =
=
(1 | S 11 | 2 )
2
G S ,max | 1 S 11 S |
gL =

GL
1 | L | 2
(1 | S 22 | 2 )
=
2
G L ,max | 1 S 22 L |

(13.46)
(13.47)
(13.48)
(13.49)

13.8 CONSTANT GAIN CIRCLES (UNILATERAL

CASE)
Considering Equations (13.35) and (13.37), it can be shown (see
appendix) that the values of S and L that produce a constant gain
(or normalized gain) lie in a circle in the Smith chart. These circles
are called constant GS or GL circles, respectively.
To obtain the equations for these circles, we start with Equations
(13.48) and (13.49). It is shown that the values of S (or L) that
produce a constant value of gS (or gL) lie in a circle described by the
following equations:
|
S - CgS| = RgS,
(13.50a)
|
L - CgL| = RgL
(13.50b)
Where the center and radius (CS, RS) and (CL, RL) for each of the
two circles are given by:

Matthew M. Radmanesh, Ph.D. 315

C gS =

RgS =

g S S 11 *
1 | S 11 | 2 (1 g S )

(1 g S ) (1 | S 11 | 2 )
1 | S 11 | 2 (1 g S )

(13.51a)
(13.51b)

and

C gL =
R gL =

g L S 22 *
1 | S 22 | 2 (1 g L )
(1 g L ) (1 | S 22 | 2 )
1 | S 22 | 2 (1 g L )

(13.52a)
(13.52b)

The two equations in (13.50) represent equations of two families of

circles where the centers of each family of circles lie along the
straight line given by the angle of S11* and S22* as shown in Figures
13.6a,b.

OBSERVATIONS:
a. From Equations (13.51) and (3.52) we note that when gS or gL=1
(i.e. maximum gain condition), we have:
RgS=0,
(13.53a)
RgL=0,
(13.53b)
CgS=S11*,
(13.53c)
And
CgL=S22*.
(13.53d)
This indicates that the maximum gain occurs only at one point
located at S11* and S22* in the S or L plane, respectively. This

RF & Microwave Design Essentials

316

observation is in agreement with our earlier results as expressed

by Equations (13.38) and (13.39).
b. The 0-dB circles (i.e., GS=1, GL=1 circles) will always pass
through the origin (i.e. S=0, L=0 points). This can be shown by
noting that gS from (13.48) can be written as:
GS=1 and S=0 gS=1-|S11|2
S 11 *
| C gS |= R gS =
(13.54)
1+ | S 11 | 2
Similarly for gL:
GL=1 and L=0 gL=1-|S22|2
These results show that the radius and distance from the origin to
the center of the 0-dB constant GS or GL circle are identical and
thus prove our observation.

c.

At the outer edge of the smith chart:

|
S|=1 GS=0=-
dB
|
L|=1 GL=0=-
dB
Since this gain value is impossible to achieve, therefore the gain
circles never intersect the outer edge of the Smith chart.

d.

For a particular gain value, there are infinite number of points

on the constant gain circle that provide the same gain. Thus the
choice of S and L along the constant gain circles are not
unique, but in order to minimize mismatch loss and maximize
bandwidth it is best to choose points close to the center of the
smith chart. This is true only for cases where noise is not of
importance. In fact for low-noise amplifier design, (as we will
see in the next chapter) one needs to use mismatch at the input
matching network in order to obtain minimum noise from the
amplifier.

13.9 UNILATERAL FIGURE OF MERIT

We already noticed that under the unilateral assumption, power gain
analysis greatly simplifies. However, in most cases S120. Thus if
we still wish to use the unilateral assumption and the simplified

Matthew M. Radmanesh, Ph.D. 317

unilateral gain equations for the amplifier design (when S120), we

need to determine the error involved in our analysis.
The error involved lies in the magnitude ratio of GT/GTU which is
obtained by dividing Equation (13.1) by (13.12):
GT
1
=
(13.55)
G TU | 1 X | 2
Where
S 12 S 21 S L
X=
(13.56)
(1 S 11 S )(1 S 22 L )
It can be shown that ratio of GT/GTU is bounded by:
G
1
1
< T <
2
G TU (1 | X |) 2
(1+ | X |)

(13.57)

When S=S11* and L=S22*, GTU achieves its maximum value,

GTU,max. The maximum error introduced using the unilateral
assumption (i.e. using GTU instead of GT) is bounded by:
GT
1
1
<
<
(13.58)
2
G TU ,max (1 U ) 2
(1 + U )
where
| S 12 || S 21 || S 11 || S 22 |
U=
(1 | S 11 | 2 )(1 | S 22 | 2 )

(13.59)

Where U is defined to be the "unilateral Figure of merit" which

varies with frequency due to its S-parameter dependence. Thus "U"
needs to be calculated at each frequency in order to obtain the limits
of the error involved due to unilateral assumption.
From table 13.1, which lists various values of "U" vs. GT/GTU, one
can determine if the calculated value of "U" gives a tolerable error
value for GT/GTU.

RF & Microwave Design Essentials

318

Usually an error of a few tenths of a dB in the GT/GTU,max ratio is

justifiable when using the unilateral assumption. The following
example illustrates this point further:
_______________________________________________________
EXAMPLE 13.2
Assume U=0.05 at 1 GHz for a microwave amplifier where S120.
a. Find the maximum error if we use unilateral gain equations for
this transistor;
b. If the transistor is used in an amplifier design with a gain of 15
dB, can unilateral assumption be used in this case?
Solution:
a. From table 13.1 we can see that for U=0.05, the GT/GTU,max is
bound by:
0.907 < GT/GTU,max <1.108
Or in dB, we have:
-0.42 dB< GT/GTU,max <0.45 dB

Matthew M. Radmanesh, Ph.D. 319

b. Since the error due to unilateral assumption is bound between 0.42 dB and +0.45 dB, it is small enough (compared to 15 dB) to
justify the unilateral assumption.
_______________________________________________________
EXAMPLE 13.3
Find the maximum error range for the transducer gain value if use
unilateral gain equations for transistor that has the following Sparameters:
S11=0.6-160, S12=0.04516,
S21=2.530, S22=0.5-90
Solution:
From Equation (13.59) we have:
| S 12 || S 21 || S 11 || S 22 | 0.045x 2.5x0.6x0.5
U=
=
= 0.070
(1 | S 11 | 2 )(1 | S 22 | 2 ) (1 0.6 2 )(1 0.5 2 )
The lower limit is:
1
1
= 0.87 = 0.59 dB
=
2
(1 + u )
(1 + 0.070) 2

The upper limit is:

1
1
=
= 1.156 = 0.63 dB
2
(1 u)
(1 0.070) 2
Thus the error range for GT is given by:
-0.59<GT<0.63 dB
______________________________________________________

13.10 BILATERAL CASE

When S120 and unilateral Figure of merit causes an unjustifiably

high error in the gain equations, we are faced with the bilateral case
where S12 can no longer be ignored.
We know that from Equations (13.30) and (13.31) the maximum
power transfer occurs when:

RF & Microwave Design Essentials

IN = S * = S 11 +

OUT = L * = S 22

S 12 S 21 L
1 S 22 L
S S
+ 12 21 S
1 S 11 S

320

(13.60)
(13.61)

Under these conditions, the overall maximum gain using lossless

matching networks is given by:
1 | L | 2
1
2
GT =
| S 21 |
1 | S | 2
| 1 S 22 L | 2

(13.62)

From Equation (13.60), we note that for a bilateral transistor S

depends on L and vice versa, from (13.61) L depends on S. This
means that these two equations are cross-coupled and must be solved
simultaneously to obtain the simultaneous conjugate match values of
S and L.
Solving Equations (13.60) and (13.61) simultaneously, we obtain the
simultaneous conjugate match values of S and L (referred to as
MS and ML):

MS

B 1 B 1 4 | C1 | 2
=
2C1

(13.63a)

ML

B 2 B 2 4 | C 2 |2
=
2C 2

(13.63b)

where
B1=1+|S11|2-|S22|2-|
|2
B2=1+|S22|2-|S11|2-|
|2
*
C1=S11-
S22
C2=S22-
S11*

(13.64a)
(13.64b)
(13.65a)
(13.65b)

Matthew M. Radmanesh, Ph.D. 321

NOTE: It can be shown that for an unconditionally stable two-Port

network (K>1, ||<1), the solutions from (13.63a,b) with a minus
sign (-) should be considered in order to obtain meaningful values
for MS and ML (i.e., |MS|<1 and |ML|<1).
Under simultaneous conjugate matched conditions, GT,max from
Equation (13.62) is obtained to be:
IN*=
MS
(13.66a)
S=
L=
OUT*=
ML
(13.66b)
G T ,max

1 | ML | 2
1
2
=
| S 21 |
1 | MS | 2
| 1 S 22 ML | 2

(13.67)

Substituting for MS and ML from Equations (13.63a,b) into (13.67),

we obtain:

G T ,max =

| S 21 |
(K K 2 1 )
| S 12 |

(13.68)

When K = 1, we obtain the maximum stable gain GMSG from

(13.68):
GMSG=GT,max|K=1 =|S21|/|S12|
GMSG is a figure of merit showing the maximum value that GT,max
can achieve. Thus simply by looking at a transistor forward (S21)
and reverse (S12) transmission coefficients, one can decide if the
transistor is useful in providing the needed gain for a particular
amplifier design or not.

13.11 SUMMARY
Before embarking upon the task of designing a functional amplifier
we need to consider one more aspect, namely "Noise Figure" which
for a sensitive receiver plays a very important role in the design
process.

RF & Microwave Design Essentials

322

This topic will be considered in detail in the next chapter and will
consummate all of the major considerations that accompany the
design process of an amplifier. This is done so that when we start
discussing the actual design process for several types of amplifiers,
where each type accentuates one or more of these major
considerations (e.g. stability, gain, noise figure, etc.), a full
knowledge of the subject has already been fully experienced and
explored by the reader.

Chapter 13 - List of Symbols

A symbol will not be repeated again, once it has been identified and
defined in an earlier chapter, with its definition remaining
unchanged.
GA - Available power gain
G0 - Transistor power gain
GL - Output matching network power gain
GLU - Output matching network power gain for unilateral case
GMSG - Maximum stable gain
GP - Operating power gain or input matching network power gain
GS - Available power gain
GSU - Available power gain for special case-unilateral transistor
GT - Transducer power gain
GT,max - Maximum transducer power gain
GTU - Transducer power gain for special case-unilateral transistor
ML - Load mismatch factor
MS - Source mismatch factor
PAVS - Power available from the source under matched condition
PAVN - Power available from the transistor under matched condition
PIN - Input power to the transistor or the input matching network
PL - Output power to the load or the output matching network
POUT - Output power from network under consideration
ML - Reflection coefficient for conjugate match value of L
MS - Reflection coefficient for conjugate match value of S

Matthew M. Radmanesh, Ph.D. 323

CHAPTER-13 PROBLEMS
13.1) The S parameters of a transistor are:
0
0.7 30
S=

490 0.50
The transistor is used in the amplifier shown in Figure P13.1,
where the output matching network produces L=0.590.
Determine the values of GT, GP and GA.

13.2) a. Determine GT, GP and GA in a microwave amplifier as

shown in Figure P13.2, if S=0.49-150, L=0.5690 and
the S-parameters of the transistor are:
0.54165 0.0920
S=
0.5 80
230
b. Calculate PAVS, PIN, PAVN and PL if E1=1030, Z1=50
and Z2=50 .

RF & Microwave Design Essentials

324

13.3) Prove that the maximum unilateral transducer power gain is

obtained when S=S11* and L=S22*.
13.4) a. Design a microwave transistor amplifier for GTU,maxusing a
BJT whose S parameters in a 50 system at VCE=10 V, IC=10
mA and f=1 GHz are given by:
0
0.7 160

S=
0.5 20
585
13.5) A microwave amplifier is to be designed for GTU,max using a
transistor with:
0
0.5140

S=
0.6 95
545
The S parameters were measured in a 50 system at f=900
MHz, VCE=15 V and IC=20 mA.
a. Determine GTU,max.
b. Draw the constant gain circle for GL=1 dB.
c. If the S parameters at 1 GHz are:
0
0.48137

S=

4.648 0.57 99
Calculate the gain GT at 1 GHz if S=0.49-150 and
L=0.5690.
13.6) An GaAs FET amplifier has the following S parameters for the
active device:
0
0.5180

S=
0.5 45
490

Matthew M. Radmanesh, Ph.D. 325

a. Is the amplifier stable?

b. What is the maximum gain in dB?
c. What is the input impedance Zin, in a 50 system?
d. What is the load impedance ZL for the maximum-gain
case?
13.7) An amplifier is operating at 10 GHz using a GaAs FET with
the following S parameters:
0
0.45 45

S=
0.8 160
5 30
Design an amplifier for maximum gain using 50
transmission lines in the output mating network, as shown in
Figure P13.7.
a. Is the amplifiers stable?
b. What is the maximum gain in dB?
c. Design the input matching network for maximum gain
using series C1 and shunt L1 elements as shown.
d. What are the values of in and its corresponding Zin?

13.8) An FET device has the following S parameters at 3GHz:

0
0.3 60

S=
0.8 30
245
Design an amplifier for maximum gain using this transistor
and 50 input and output transmission lines:
a. Check the stability
b. What is the maximum gain in dB?
c. Design an input-matching network for maximum gain
using series L and shunt C elements to match a 50 line to
ZS.
13.9) The FET amplifier shown in Figure P13.9 has the following S
parameters in a 50 system:

RF & Microwave Design Essentials

326

0
0.5180

S=

3.090 0.5 90
The circuit is terminated by RS= 50 and RL=100 . Find:
a. Zin and Zout,
b. The unilateral power gain, in dB.
c. If matching networks are used at the input and output ports
such that maximum power transfer occurs, find the
maximum unilateral power gain in dB.

13.10) Consider the microwave network shown in Figure 13.10,

consisting of a 50 source, a 50 -3 dB matched attenuator and a
50 load:
a. Compute the available power gain, the transducer power
gain, and the actual power gain.
b. How do these gains change if the load is changed to 25
? How do these gains change if the source impedance is
changed to 25?

13.11) Use the new -parameter test to determine which of the

following devices are unconditionally stable, and of these,
which has the greatest stability:

Matthew M. Radmanesh, Ph.D. 327

13.12) Show that for a unilateral device where S12=0, the parameter test implies that:
|S11|<1 and |S22|<1 (for unconditional stability).

REFERENCES
[13.1] Bahl, I. and P. Bhartia. Microwave Solid State Circuit
Design. New York: Wiley Interscience, 1988.
[13.2] Gonzalez, G. Microwave Transistor Amplifiers,
Analysis and Design, 2 nd ed. Upper Saddle River:
Prentice Hall, 1997.
[13.3] Ha, T. T. Solid State Microwave Amplifier Design. New
York: John Wiley & Sons, 1987.
[13.4] Liao, S. Y. Microwave Circuit Analysis and
Amplifier Design. Upper Saddle River: Prentice
Hall, 1987.
[13.5] Pozar, D. M. Microwave Engineering, 2nd ed. New
York: John Wiley & Sons, 1998.
[13.6] Vendelin, George D. Design of Amplifiers and
Oscillators by the S-Parameter Method. New York:
John Wiley & Sons, 1982.
[13.7] Vendelin, George D., Anthony M. Pavio, and Ulrich
L. Rhode. Microwave Circuit Design, Using Linear
and Non-Linear Techniques. New York: John
Wiley & Sons, 1990.

RF & Microwave Design Essentials

328

Matthew M. Radmanesh, Ph.D. 329

CHAPTER 14

Noise in Active Networks

14.1 INTRODUCTION
Having done a stability check and having met the gain requirements
of an amplifier, the next important point to consider is noise. In an
RF/microwave amplifier, the existence of noise signal plays an
important role in the overall design procedure and needs to be
grasped before a meaningful design process can be developed.
Noise power results from random processes that exist in nature.
These random processes can be classified in several important
classes each generating a certain type of noise which will be
characterized shortly.
Some of the most important types of random processes are:
1. Thermal vibrations of atoms, electrons and molecules in a
component at any temperature above 0 K.
2. Flow of charges (electrons and/or holes) in a wire or a device.
3. Emission of charges (electrons or ions) from a surface such as
the cathode of a diode or an electron tube, etc.
4. Wave propagation through atmosphere or any other gas

RF & Microwave Design Essentials

330

14.2 IMPORTANCE OF NOISE

Noise is passed into a microwave component or system either from
an external source or is generated within the unit itself. Regardless
of the manner of entrance of the noise signal, the noise level of a
system greatly affects the performance of the system by setting the
minimum detectable signal in the presence of noise. Therefore it is
often desirable to reduce the influence of external noise signals and
minimize the generation of noise signals within the unit, in order to
achieve the best performance.

14.3 NOISE DEFINITION

Since noise considerations are of important consequences, we need
to define it first:
DEFINITION- ELECTRICAL NOISE (OR NOISE): Is defined to be any
unwanted electrical disturbance or spurious signal. These unwanted
signals are random in nature and are generated either internally in
the electronic components or externally through impinging
electromagnetic radiation.
Since signals are totally random and uncorrelated in time, they are
best analyzed though statistical methods. Their statistical properties
can be briefly summarized as:
a. The "Mean value" of the noise signal is zero, i.e.,

V n = Lim T 1 / T

t1 + T

V (t )dt = 0
n

(14.1)

t1

_
Where Vn is the noise mean value, Vn(t) is the instantaneous noise
voltage, t1 is any arbitrary point in time and T is any arbitrary period
of time ideally a large one approaching .

b.

The "mean-square-value" of the noise signal is a constant

value, i.e.,
t1 + T

Vn = Lim T 1 / T [Vn (t )]2 dt = Constant

2

t1

(14.2)

Matthew M. Radmanesh, Ph.D. 331

c. The "root-mean-square" (rms) of a noise signal is given by:

Vn ,rms = Vn

(14.3)

(14.4)

or,
2
( Vn ) rms
= Vn

The concept of "root-mean-square value" of noise as given by

Equation (14.3), is based on the fact that the "mean-square value",
2

Vn , is proportional to the "noise power". Thus if we take the

square root of Equation (14.2), we obtain the "rms value" of the
noise voltage which is the "effective value" of the noise voltage.

14.4 SOURCES OF NOISE

There are several types of noise which needs to be defined:
a. DEFINITION-THERMAL NOISE (ALSO CALLED JOHNSON NOISE OR
NYQUIST NOISE): is the most basic type of noise which is caused by
thermal vibration of bound charges and thermal agitation of
electrons in a conductive material. This is common to all passive or
active devices.
b. DEFINITION-SHOT NOISE (OR SCHOTTKY NOISE): is caused by
random passage of discrete charge carriers (causing a current "I",
due to motion of electrons or holes) in a solid state device while
crossing a junction or other discontinuities. It is commonly found in
a semiconductor device (e.g. in a pn junction of a diode or a
transistor) and is proportional to (I)1/2 .
c. DEFINITION-FLICKER NOISE (ALSO CALLED 1/f
vibrations of a current due to the following factors:

NOISE):

is small

1. Random injection or recombination of charge carriers at an

interface, such as at a metal semiconductor interface (in
semiconductor devices).

RF & Microwave Design Essentials

332

2. Random charges in cathode emissions of electric charges such

as at a cathode-air interface (in a thermionic tube).
Flicker noise is a frequency-dependent noise, which distorts the
signal by adding more noise to the lower part of the signal band
than the upper part. It exists at lower frequencies, almost from
DC extending down to approximately 500 kHz to 1 MHz at a rate of
10 dB per decade.

14.5 THERMAL NOISE ANALYSIS

To analyze noise, let us consider the circuit shown in Figure 14.1a
where a noisy resistor is connected to the input port of a two-port
network. Focusing primarily on thermal noise, we note that the
available noise power (i.e. maximum power available under matched
conditions) from any arbitrary resistor has been shown by Nyquist to
be:
PN = kTB
(14.5)
Where,
k= Boltzmann's constant (=1.374x10-23 J/K).
T= The resistor's physical temperature.
B= The 2-port network's bandwidth (i.e., B= fH-fL).
Since the noise power does not depend on the center frequency of
operation but only on the bandwidth, it is called "white noise" as
shown in Figure 14.1b.
There are a few observations about noise power (PN) which is worth
considering:
a. As bandwidth (B) is reduced, so does the noise power which
means narrower bandwidth amplifiers are less noisy,

Matthew M. Radmanesh, Ph.D. 333

b. As temperature (T) is reduced, the noise power is also lessened

which means cooler devices or amplifiers generate less noise power,
c. Increasing bandwidth to infinity causes an infinite noise power
(called ultraviolet catastrophe) which is incorrect since (14.5) for
noise power is only valid up to approximately 1000 GHz.

14.6 NOISE MODEL OF A NOISY RESISTOR

A noisy resistor (RN) at a temperature (T) can be modeled by an
ideal noiseless resistor (RNO) at 0 K in conjunction with a noise
voltage source (Vn,rms) as shown Figure 14.2. If we assume that the
resistor value is independent of temperature then RNO=RN.

From this model, the available noise power to the load (under
matched condition) is given by (see Figure 14.3):
2
Vn ,rms
PN =
(14.6)
4R N

RF & Microwave Design Essentials

334

Equation (14.17) provides the noise power available from a noisy

resistor which equals Equation (14.5) for any arbitrary resistor.
Thus:
(14.7a)
PN=kTB
Vn,rms = 2 PN R N = 2 kTBR N
(14.7b)
From Equation (14.7) we can observe that the noise voltage is
proportional to ( R1/2 ). Thus higher-valued resistors have higher
noise voltage even though they provide the same noise power level
as the lower-valued resistors.
______________________________________________________
EXAMPLE 14.1
Calculate the noise power (in dBm) and rms noise voltage at
T=290K for
a. RN=1 , B=1 Hz
b. RN=2 M, B=5 kHz.
Solution:
a. The noise power is given by:
k=1.374x10-23 J/K
PN=kTB=1.374x10-23 x290x1=3.985x10-21 W
Or in dBm, we have:
PN(dBm)=10log(3.985x10-21/10-3)=-174 dBm
This is the power per unit Hz. The corresponding noise voltage for a
1 resistor is given by:

Vn,rms = 2 PN R N = 2 3.985x10 21 x1 = 12.6x10 11 V

=12.6x10-5 V
b. For a 5 kHz bandwidth, we have
PN=kTB=3.985x10-21x5000=19.925x10-18 W

Matthew M. Radmanesh, Ph.D. 335

The corresponding noise voltage for a 2 M resistor is given by

Vn,rms = 2 PN R N = 2 19.925x10 18 x 2x10 6 = 12.6x10 6 V

=12.6 V
_______________________________________________________

14.7 EQUIVALENT NOISE TEMPERATURE

Any type of noise, in general, has a power spectrum which can be
plotted in the frequency domain. If the noise power spectrum is not a
strong function of frequency (i.e., it is "White" noise) then it can be
modeled as an equivalent thermal noise source characterized by an
"equivalent noise temperature" (Te).
To define "the equivalent noise temperature" (Te), we consider an
arbitrary white noise source with an available power (PS) having a
noiseless source resistance (RS) as shown in Figure 14.4a. This white
noise source can be replaced by a noisy resistor with an equivalent
noise temperature (Te) defined by:
PS
(14.8)
kB
Where B is the bandwidth of the system or the component under
consideration.
Te =

_______________________________________________________
EXAMPLE 14.2:
Consider a noisy amplifier with available power gain (GA) and
bandwidth (B) connected to a source and load resistance ( R ) both

RF & Microwave Design Essentials

336

at T=TS as shown in Figure 14.5. Determine the overall noise

temperature of the combination and the total output noise power if
the amplifier all by itself creates an output noise power of Pn.

Solution:
To simplify the analysis, let us first assume that the source resistor is
at T=0K. This means that no noise enters the amplifier, i.e., PNi=0.
The noisy amplifier can be modeled by a noiseless amplifier with an
input resistor at an equivalent noise temperature of:
Pn
(14.9)
Te =
G A kB
Te is called the equivalent noise temperature of the amplifier
"referred to the input" as shown in Figure 14.6.

Since source resistor ( R ) is at a physical temperature other than

zero (T=TS) , then as a result the combined equivalent noise
temperature (Te') is the addition of the two noise temperatures:
Te'=Te+TS
(14.10)
Assuming the noise power at the input terminals of the amplifier is
PNI (=kTSB), the total output noise power due to the amplified input
thermal noise power will be (GAPNi) which adds to the amplifier's

Matthew M. Radmanesh, Ph.D. 337

generated noise power (Pn) linearly by using the superposition

theorem (see Figure 14.7), i.e.,
PNo,tot=GAPNi+Pn=GAkB(TS+Te)
(14.11)
PNo,tot= GAkBTe'

NOTE: It is important to note that from (14.11), the "equivalent

noise temperature" (Te') is defined by "referring" the total output
noise power to the input port. Thus the same noise power is
delivered to the load by driving a "noiseless amplifier" with a
resistor at an equivalent temperature (Te'=Te+TS).
_______________________________________________________

14.7.1 A Measurement Application: Y-Factor Method

The concept of equivalent noise temperature is commonly used in
the measurement of noise temperature of an unknown amplifier
using the "Y-factor method". In this method, the physical
temperature of a matched resistor is changed to two distinct and
known values:
a. One temperature (T1) is at boiling water (T1=100C) or at room
temperature (T1=290 K),
b. The second temperature (T2) is obtained by using either a noise
source
(hotter source than room temperature) or a load immersed in
liquid nitrogen at T=77 K (a colder source than room
temperature) as shown in Figure 14.8.

RF & Microwave Design Essentials

338

The amplifier's unknown noise temperature (Te) can be obtained as

follows:
PNo,1=GAkB(T1+Te)
(14.12)
PNo,2=GAkB(T2+Te)
(14.13)
Now define:
PNo ,1
Y
PNo , 2
Thus we can write:
T + Te
(14.14)
Y= 1
T2 + Te
Or,
T YT2
Te = 1
(14.15)
Y -1
From a measurement of T1, T2 and Y the unknown amplifier's noise
temperature (Te) can be found.
POINT OF CAUTION: To obtain an accurate value for Y, the two
temperatures ideally must be far apart; otherwise Y1 and the
denominator of Equation (14.15) will create relatively inaccurate
results.
NOTE: A noise source "hotter" than room temperature, as used in
the Y-factor measurement, would be a solid state noise source (such

Matthew M. Radmanesh, Ph.D. 339

as an IMPATT diode) or a noise tube. Such active sources,

providing a calibrated and specific noise power output in a
particular frequency range, are most commonly characterized by
their "excess noise ratio" values vs. frequency. The term "excess
noise ratio" or ENR is defined as:
P PO
T TO
= 10 log 10 N

ENR (dB ) = 10 log 10 N

(14.16)
PO
TO
where PN and TN are the noise power and equivalent noise
temperature of the active noise generator, and PO and TO are the
noise power and temperature of a passive source(e.g. a matched
load), respectively.

14.8 DEFINITIONS OF NOISE FIGURE

As discussed earlier, a noisy amplifier can be characterized by an
equivalent noise temperature (Te). An alternate method to
characterize a noisy amplifier, is through the use of the concept of
noise Figure which we need to define first.
DEFINITION- NOISE FIGURE: Is defined to be the ratio of the total
available noise power at the output, (PO)tot, to the output available
noise power (PO)i due to thermal noise coming only from the input
resistor at the standard room temperature (TO=290 K).
To formulate an equation for noise figure (F), let us transfer the
noise generated inside the amplifier (Pn) to its input terminals and
model it as a "noiseless" amplifier which is connected to a noisy
resistor ( R ) at noise temperature (Te) in series to another resistor (
R ) at T=TO, both connected at the input terminals of the "noiseless"
amplifier as shown in Figure 14.9. From this configuration we can
write:
Pn=GAkTeB
(14.17a)
(PO)i=GAPNi=GAkBTO
(14.17b)
(PO)tot=PNO=Pn+(PO)i
(14.18)
(P )
( P ) + Pn
Pn
F = O tot = O i
= 1+
(14.19a)
( PO ) i
( PO ) i
G A PNi
Or,

RF & Microwave Design Essentials

Te
TO
Or, in dB we can write:

T
F = 10 log 10 1 + e
TO

F = 1+

340

(14.19b)

(14.20)

FIGURE 14.9 A noisy amplifier

From (14.19) we can see that "F" is bounded by:

1 F

(14.21)

The lower boundary (F=1) is the best case scenario and is the noise
Figure of an ideal noiseless amplifier where Te=0.
From Equation (14.19b), we can write:
Te=(F-1)TO

(14.22)

NOTE1: Temperature (Te) is the equivalent noise temperature of the

amplifier referred to the input.
NOTE 2: Either F or Te can interchangeably be used to describe the
noise properties of a two-port network. However, For small noise
Figure values (i.e., when F1), use of Te becomes preferable.
POINT OF CAUTION: It is interesting to note that the noise Figure is
defined with reference to a matched input termination at room
temperature (TO=290 K). Therefore if the physical temperature of
the amplifier changes to some value other than TO, we still use the
room temperature (TO=290 K) to find the noise figure value.

14.8.1 Alternate Definition of Noise Figure

From Equations (14.17) and (14.18), we can write:
PNO=GAPNi+Pn

(14.23)

Matthew M. Radmanesh, Ph.D. 341

(Po)i=GAPNi
(14.24)
Where Pn=GAkTeB is the generated noise power inside the
amplifier. The noise figure can now be written as:
PN O
PN O
Pn
F=
=
= 1+
(14.25)
( PO ) i G A PN i
G A PN i
The available power gain (GA) by definition is given by:
PS
GA = O
PS i
where PSO and PS i are the available signal power at the output and
the input, respectively. Thus Equation (14.25) can now be written as:
PS /PN i
(SNR) i
(14.26)
F= i
=
PSo /PN o (SNR) o
where (SNR)i and (SNR)o are the available signal-to-noise ratio at
the input and output ports, respectively.
Equation (14.26) indicates that the noise figure can also be defined
in terms of the ratio of available signal-to-noise power ratio to the
available signal-to-noise power ratio at the output

14.8.2 Noise Figure of a Lossy Two-Port Network

This is an important case, where the two-port network considered
earlier, is a lossy passive component such as an attenuator or a lossy
transmission line, as shown in Figure 14.10.

RF & Microwave Design Essentials

342

PO
) less than unity which can be
Pi
expressed in terms of the loss factor or attenuation (L) as:
1
GA =
(GA<1)
(14.27)
L

A lossy network has a gain ( G A =

Since the gain of a lossy network is less than unity it follows that the
loss or attenuation factor (L) is more than unity (i.e., L=Pi/Po>1) for
any lossy network or component.
Expressing the attenuation factor (L) in "dB" gives the following:
P
(14.28)
L(dB ) = 10 log 10 i
Po
For example, if the lossy component attenuates the input power by
ten times, then we can write:
P
G A = O = 0.1 L=1/GA=10=10 dB
Pi
If the lossy network is held at a temperature (T), the total available
output noise power according to Equation (14.5) is given by:
PNO=kTB
(14.29)
On the other hand, from (14.23) the available output noise power is
also given by the addition of the input noise power and the generated
noise inside the circuit (Pn):
PNO=GAkTB + Pn=KTB/L+Pn
(14.30)
where Pn is the noise generated inside the two-port. Equating
Equations (14.29) and (14.30), we obtain Pn as:
L 1
(14.31a)
Pn =
kTB
L
NOTE: If we refer the noise generated inside the amplifier (Pn) to the
input side (Pn)i, from (14.31a) we have:
(Pn)i=Pn/GA=LPn=(L-1)kTB
(14.31b)

Matthew M. Radmanesh, Ph.D. 343

Using Equations (14.31) we can now define the equivalent noise

temperature (Te) of a lossy two-port, referred to the input terminals,
as:
(p )
Te=(L-1)T
(14.32)
Te = n i
kB
Thus the noise figure of a lossy network is given by:
T
F = 1+ e
TO
T
= 1 + (L 1)
TO

(14.33)

A SPECIAL CASE:
For a lossy network at room temperature, i.e., T=To, Equation
(14.33) gives:
F=L
(14.34)

Equation (14.34) indicates that the noise figure of a lossy network at

room temperature equals the attenuation factor (L). For example: if
GA=1/5 then L=1/GA=5, giving F=5 or 7 dB for T=To=290 K.
______________________________________________________
Example 14.3
A wideband amplifier (2-4 GHz) has a gain of 10 dB, an output
power of 10 dBm and a noise figure of 4dB at room temperature.
Find the output noise power in dBm.
Solution:
B=2 GHz
GA=10 dB
F=4 dB
F=PNO/GAPNi=PNO/GAkToB
Thus:
PNO=FGAkToB
10 log10PNO=PNO(dB)=F(dB)+GA(dB)+10 log10(kTo)+10 log10(B)
=4+10-174+10 log10(2x109)= -67 dBm
______________________________________________________

RF & Microwave Design Essentials

344

14.9 NOISE FIGURE OF CASCADED NETWORKS

A microwave system usually consists of several stages or networks
connected in cascade where each adds noise to the system thus
degrading the overall signal-to-noise ratio.
If the noise figure (or noise temperature) of each stage is known, the
overall noise figure (or noise temperature) can be determined.

14.9.1 Cascade of Two Stages

To analyze a two stage amplifier, let us consider a cascade of two
amplifiers each with its own gain, noise temperature or noise figure
as shown in Figure 14.11. The noise power of each stage is given as
follows:
PNO1=GA1kB(To+Te1)
(14.35)
PNO2=GA2PNO1+GA2kTe2B
(14.36)

Combining Equations (14.35) and (14.36) we get:

PNO2=GA1GA2kB(To+Te1+Te2/GA1)
(14.37)
The two-stage amplifier as a whole has an total gain of GA=GA1GA2,
an overall equivalent noise temperature (Te) and a total output noise
power (PNO) given by:
PNO=GAkB(TO+Te)
(14.38)
Comparing Equation (14.38) to (14.37) we have:
Te=Te1+Te2/GA1
(14.39)
The overall noise figure (F) for the two-stage amplifier is found by
using (14.39):
F=1+Te/To=1+(Te1+Te2/GA1)/To
(14.40)
By noting that:
F1=1+Te1/To,
(14.41)
F2=1+Te2/To
(14.42)
Equation (14.40) can be written as:

Matthew M. Radmanesh, Ph.D. 345

F2 1
(14.43)
G A1
Equations (14.39) and (14.43) show that the first stage noise figure
F1 (or noise temperature, Te1), and gain (GA1) have a large influence
on the overall noise figure (or noise temperature). This is because
the second stage noise figure, F2 (or noise temperature, Te2) is
reduced by the gain of the first stage (GA1).
F = F1 +

Thus the key to low overall noise figure is a primary focus on the
first stage by reducing its noise and increasing its gain. Later stages
have a greatly reduced effect on the overall noise figure.
NOISE MEASURE
In order to determine systematically the order or sequence in which
two similar amplifiers need be connected to produce the lowest
possible noise figure, we first must define a quantity called "noise
measure" as:
F -1
M=
(14.44)
1 - 1/G A
If amplifier #1 (AMP1) has a noise measure (M1) and amplifiers #2
(AMP2) a noise measure (M2) then there are two possible cases
which needs to be addressed (in order to obtain the lowest possible
noise form the cascade), as follows:

Case I: M1<M2 -- Then AMP1 should precede AMP2 since F12<F21

Case II: M2<M1 -- Then AMP2 should precede AMP1 since F21<F12
Where F12 is the overall noise figure of the two stage amplifier when
AMP1 precedes AMP2; and vice versa F21 is for the case when
AMP2 precedes AMP1.

NOTE: It can easily be shown mathematically that for example:

If M1<M2 then F12<F21
(14.45)
where
F 1
F12 = F1 + 2
(14.46)
G A1

RF & Microwave Design Essentials

F21 = F2 +

F1 1
G A2

346

(14.47)

And vice versa, if M2<M1 then F21<F12.

_______________________________________________________
Example 14.4
An antenna is connected to an amplifier via a transmission line
which has an attenuation of 3 dB (see Figure 14.12). The amplifier
has the following specifications:
GA=20 dB
B=200 MHz
Te=145 K
Calculate the overall noise figure and gain of the cascade at 300 K.

Solution:
a. For the transmission line we have:
Since L=1/GTL L(dB)=-GTL(dB)
L=3 dB=2 GTL=-3 dB=1/2
FTL=1+(L-1)T/To=1+(2-1)300/290=2.03=3.1 dB
b. For the amplifier we have:
FAMP=1+Te/To=1.5=1.8 dB
c. The overall noise figure and gain are calculated to be:
FTOT=FTL+(FAMP-1)/GTL=2.03+(1.5-1)/0.5=3.03=4.8 dB
GTOT=GTL+GAMP=-3+20=17 dB
Therefore we can see that due to the addition of a lossy transmission
line in front of the amplifier, we have three deleterious effects: 1) the
overall noise figure has increased (from 1.8 dB to 4.2 dB) 2) the
second stage noise contribution has been intensified since the

Matthew M. Radmanesh, Ph.D. 347

transmission line has a gain less than unity (GTL<1), and 3) the
overall gain dropped by 3 dB which represents the third side effect.
______________________________________________________

14.10 CONSTANT NOISE FIGURE CIRCLES

As derived in reference [14.1], the noise Figure of a two-Port
amplifier is given by
2
r
F = Fmin + n YS Yopt
(14.61)
gS
Where
R
rn = n (the equivalent normalized noise resistance of the two-port).
ZO

YS=gS+jbS (the normalized source admittance corresponding to S

as defined in Chapter 13)
Yopt=gopt+jbopt (the normalized source admittance for minimum
noise figure, i.e. at S=opt F=Fmin)
Since YS and Yopt are related to S and opt by the relations:
1 S
YS =
(14.62)
1 + S
1 opt
Yopt =
(14.63)
1 + opt
Then using S and opt instead of YS and Yopt in Equation (14.61), we
can write:
F = Fmin +

4rn S opt

(1 )1 +
2

(14.64)

opt

We now define a parameter called the noise Figure parameter (N):

N=

S opt
1 S

(14.65)

348

RF & Microwave Design Essentials

Thus Equation (14.64) can be written as:

4rn N
F = Fmin +
2
1 + opt

(14.66)

Parameters rn, opt, and Fmin are called the "noise parameters" of the
transistor and are usually provided in the data sheets by the
manufacturer.
NOTE 1: Using Equation (14.66), we can write Equation (14.65) as:
N=

S opt
1 S

F Fmin
1 + opt
4rn

(14.67)

From Equation (14.67) we can see that for a fixed (F), the
parameter (N) is a positive real number.
NOTE 2: Noise parameters may also be determined experimentally
by the following procedure:
a. Vary S until a minimum noise figure occurs which is
subsequently recorded (i.e., F=Fmin).
b. Now using a vector network analyzer, measure S which provides
the value for opt.
c. To find rn, we set S to zero and measure the noise figure ( FS =0 )

at this point. Then using Equation (14.64) and the value of opt
from step (b), we can obtain rn as:
1+ 2
opt

rn = F
(14.68)
2
4 opt

Where
F = FS =0 Fmin

14.10.1 Analysis
By using Equation (14.67) and through rearranging terms and further
mathematical manipulation of Equation (14.64), we obtain an
equation of a circle in the S-plane as:

Matthew M. Radmanesh, Ph.D. 349

|
S - CF|= RF
(14.69)
where CF and RF are the center and radius of noise figure circles
given by:
opt
CF =
(14.70)
N +1
2

RF =

N 2 + N(1 opt )

(14.71)
1+ N
Equation (14.69) represents a family of noise figure circles with the
noise figure (F) value as a parameter. For derivation of constant noise
figure circles, please refer to Appendix M.

OBSERVATIONS:
When F=Fmin then:
S=
opt N=0
(14.72)
CF=
opt
(14.73)
RF=0
(14.74)
Equation (14.73) and (14.74) indicate that Fmin is a point uniquely
located at opt. From Equation (14.67) we can see that since "N" is a
real positive number, then all noise figure circles have centers
located along opt vector in the S plane as shown in Figure 14.14.

_______________________________________________________

350

RF & Microwave Design Essentials

EXERCISE: Derive Equation (14.69) from Equations (14.64) through

(14.67).
Hint: Use the following identity:
S opt

= (S opt )(S opt )

to write (14.52) as:

+ S opt
*

)+

N
1+ N
1+ N 1+ N
2
2
Now add |opt| /(N+1) to both sides of the above equation to obtain
the desired relation for a circle as given by Equation (14.69).
_______________________________________________________
S

opt

opt

Chapter 14- Symbol List

A symbol will not be repeated again, once it has been identified and
defined in an earlier chapter, with its definition remaining
unchanged.
B - Bandwidth
F - Noise Figure
k - Boltzmanns constant
M - Noise measure
N - Overall noise Figure
PN - Noise power
PNI - Input Noise power
PN0,tot - Total output noise power
RN - Noisy resistor
RN0 - Noiseless resistor
T - Temperature
Te - Equivalent noise temperature
T0 - Standard room temperature (290 K)
TS - Source and load temperature
Vn,rms- Root-mean-square (rms) of noise
2

Vn - The mean-square value of noise

CHAPTER-14 PROBLEMS
14.1) The Y-factor method is to be used to measure the equivalent
noise temperature of a component. A hot load of T1=300 K

Matthew M. Radmanesh, Ph.D. 351

14.2)

14.3)

14.4)

14.5)

14.6)

and a cold load of T2=77 K will be used. If the noise

temperature of the amplifier is Te=250 K, what will be the
ratio of power meter readings at the output of the component
for the two loads?
A transmission line has a noise figures F=1 dB at a
temperature TO=290 K. Calculate and plot the noise figure of
this line as its physical temperature ranges from T=0 K to
1000 K.
Assume that measurement error introduces an uncertainty of
Y into the measurement of Y in a Y-factor measurement.
Derive an expression for the normalized error of the
equivalent noise temperature (Te/Te) in terms of (Y/Y)
and the temperatures T1, T2 and Te. Plot (Te/Te) as a
function Te for two values of (Y/Y): a) 0.1, and b) 0.20,
and from these plots establish the minimum normalized error
(Te/Te) and the corresponding Te for each case.
An amplifier with a bandwidth of 1 GHz has a gain of 15 dB
and a noise temperature of 250 K. If it is used as a
preamplifier in a cascade, preceding a microwave amplifier
of 20 dB gain 5 dB noise figure, determine the overall noise
temperature.
An amplifier with a gain of 12 dB, a bandwidth of 150 MHz
and a noise figure of 4 dB feeds a receiver with a noise
temperature of 900K. Find the noise figure of the overall
system.
Consider the microwave system shown in Figure P14.6,
where the bandwidth is 1 GHz centered at 20 GHz and the
physical temperature of the system is TO=300 K. What is the
equivalent noise temperature of the source? What is the noise
input of the amplifier in dB? When the noisy source is
connected to the system what is the total noise power output
of the amplifier in dBm?

RF & Microwave Design Essentials

352

14.7) Consider the wireless local area network (WLAN) receiver

front-end shown in Figure P14.7, where the bandwidth of the
bandpass filter is 100 MHz centered at 2.4 GHz. If the
system is at room temperature, find the noise figure of the
overall system. What is the resulting signal to noise ratio at
the output if the input signal power level is -90 dBm? Can
the components be rearranged to give a better noise figure?

14.8) A two-way power divider has one output port terminated in a

matched
load as shown in Figure P14.8. Find the equivalent noise
temperature of the resulting two-Port network if the divider
is an equal-split two-way resistive divider.

Matthew M. Radmanesh, Ph.D. 353

14.9) For a two-stage cascaded network with gain values of G1 and

G2 and
noise figures of F1 and F2 as shown in Figure P14.9, the input
noise power is Ni=kTB. The output noise power is N1 and N2
at the output of the first and second stages. Are the following
expressions correct:
a. F1=N1/G1Ni
b. F2=N2/G2N1
c. F2=N2/G1G2F2Ni

14.10) A receiver has the block diagram shown in Figure P14.10.

Calculate:
a. The total gain (or loss) in dB,
b. The overall noise Figure in dB.

RF & Microwave Design Essentials

354

14.11) Two satellite receiver systems have the following

specifications for
their components:
RF Amplifier: F=5 dB, G=10 dB
Mixer:
Lc=5 dB
IF amplifier:
F=2 dB, G=15 dB
Bandpass filter: IL=2 dB
Compare the two systems in terms of the overall gain and
noise figure values (see Figure P14.11).

14.12) Calculate the overall noise Figure and gain for the receiver
system shown in Figure P14.12.

14.13) The S parameters and the noise parameters of a GaAs FET at

10 GHz are:
0.6 170 0.0516
S=
0.5 95
2 30
Fmin=2.5 dB

Matthew M. Radmanesh, Ph.D. 355

O=0.5
145
Rn =5
a. Is the transistor unconditionally stable?
b. Determine GA,max
c. Determine the noise figure if the transistor is used in an
amplifier designed for maximum available gain (GA,max)
14. 14) Consider the low noise block (LNB) shown in Figure P14.14.
Calculate the total noise figure and the available gain of this
block.

REFERENCES
[14.1] Ambrozy, A. Electronic Noise. New York: McGraw-Hill,
1982.
[14.2] Cappy, A. Noise Modeling and Measurement Techniques.
IEEE Transactions on Microwave Theory and Technique,
Vol. MTT-36, January 1988, pp. 110.
[14.3] Friis, H. T. Noise Figure of Radio Receivers. Proceedings of
IRE, Vol. 32, July 1944, pp. 419 22.
[14.4] Fukui, H. Available Power Gain, Noise Figure, and Noise
Measure of Two-Ports and Their Graphical
Representation. IEEE Transactions on Circuit Theory,
June 1966, pp. 13742.
[14.5] Gonzalez, G. Microwave Transistor Amplifiers,
Analysis and Design, 2 nd ed. Upper Saddle River:
Prentice Hall, 1997.
[14.6] Haus, H. A. and R. B. Adler. Circuit Theory of Linear
Noisy Networks. Cambridge: MIT Press, 1959 and New
York: John Wiley & Sons, 1959.
[ 14.7] Pozar, D. M. Microwave Engineering, 2nd ed. New York:
John Wiley & Sons, 1998.
[14.8] Radmanesh, M. M., and J. M. Cadwallader. Millimeter-

RF & Microwave Design Essentials

356

Wave Noise Sources at V-Band (50 to 75 GHz),

Microwave Journal, Vol. 36, No. 2, pp. 128134, Sept.
1993.
[14.9] Radmanesh, M. M. and J. M. Cadwallader. Solid State
Noise Sources at mm-Waves: Theory and Experiment.
Microwave Journal, Vol. 34, No. 10, pp. 125133, Oct. 1991.
[14.10] Radmanesh, M. M. The Gateway to Understanding: Electrons
to Waves and Beyond, AuthorHouse, 2005.
[14.11] Radmanesh, M. M. Cracking the Code of Our Physical
Universe, AuthorHouse, 2006.
[14.12] Vendelin, George D., Anthony M. Pavio, and Ulrich L.
Rhode. Microwave Circuit Design, Using Linear and
Non-Linear Techniques. New York: John Wiley &
Sons, 1990.

Matthew M. Radmanesh, Ph.D. 357

CHAPTER 15

Amplifier Design
15.1 INTRODUCTION
One of the most basic concepts in microwave circuit design is
amplification. In the past, microwave tubes and microwave diodes
(biased in the negative resistance region) was commonly used,
however, nowadays use of microwave transistors (BJT or FET) has
become very popular.
Transistor amplifiers are built rugged and are reliable for low-power
to medium-power applications. In this chapter, we will consider the
design of small-signal, narrow-band, low-noise, broad-bound and
finally multi-stage amplifiers. The design methods presented are
based on the S-parameters of the transistor in the amplifier circuit.
We will relegate the design of high-power amplifiers to the next
chapter when we consider large signal amplifiers.

15.2 TYPES OF AMPLIFIERS

Having considered basic concepts in stability, gain and noise of a
two-Port amplifier in the previous three chapters, we are now ready
to embark upon the task of designing a functional amplifier.

RF & Microwave Design Essentials

358

In general, any type of amplifier requires an optimization of either

the gain, noise figure, power or bandwidth. Therefore each type of
design will be an interplay of constant-gain and noise-figure circles
or power contours depending on the design requirements.

15.2.1 Classes of Amplifiers Based on Operating-Point

An amplifier usually operates in one of the following classes:
a. Class A amplifier
b. Class B amplifier
c. Class AB amplifier
d. Class C amplifier
These are described as follows:
a.
Class A amplifier: in this mode, each transistor in the
amplifier operates
in its active region for the entire signal cycle.
b.
Class B amplifier: In this mode of operation, each transistor
is in its active region for approximately half of the signal
cycle.
c.
Class AB amplifier: In this mode, an amplifier operates in
class A for small signals and in class B for large signals.
d.
Class C amplifier: In this mode of operation, each transistor
is in its active region for significantly less than half of the
signal cycle.

15.2.2 Classes of Amplifiers Based on Signal-Level

Even though the above four classes clearly describe the different
types of transistor amplifiers under different conditions, however,
most microwave amplifiers are usually classified according to the
signal level: a) small signal mode or b) large signal mode. The
method of analysis of an amplifier under each signal condition is
drastically different. Each mode of operation offers its own unique
style of analysis as follows:
1. Small Signal Analysis: Is a method of analysis of an active
circuit in which it is assumed that the signals deviate from (or
fluctuate to either side of) the steady bias levels by such a small

Matthew M. Radmanesh, Ph.D. 359

amount that only a small part of the operating characteristic of the

device is covered and thus the operation is always linear.
2. Large Signal Analysis: Is a method of analysis of an active
circuit under high amplitude signals that traverse such a large part
of the operating characteristics of a device that nonlinear portions
of the characteristic are usually encountered causing nonlinear
operation of the device.
In this chapter the primary focus will be on small signal amplifiers
and the operating point most suitable for this purpose would be
Class A amplifiers which will be studied here. Large signal
amplifiers will be studied in the next chapter.

15.3 SMALL SIGNAL AMPLIFIERS

To properly design small signal amplifiers, we need to determine the
biasing values for the transistor first and then use great skill to
design and connect the DC circuitry correctly to the RF portion of
the amplifier.
Thus generally speaking, the design of an amplifier (or for that
matter any microwave circuit) consists of the design of two separate
circuits:
a. DC circuit design
b. RF/MW circuit design
These two circuits are then integrated and packaged seamlessly into
one complete unit as discussed next.

15.3.1 Amplifier's DC-Bias Circuit Design

To operate an amplifier under small signal condition, linear
operation is required. This demands that the amplifier operate in
class A mode. Under this condition, the D.C. Q-point of the
amplifier must be chosen approximately midrange in IC-VCE (for
BJT) or ID-VDS (for FET) characteristic curves insuring active mode
operation of the transistor as shown in Figure 15.1.

RF & Microwave Design Essentials

360

Furthermore, it is essential that the amplifier's bias circuitry be

connected to the RF circuit in such a way that it will create
minimum interaction and leakage for the RF/microwave signals
(which are traveling from the input to the output). In other words,
the DC circuitry should be completely isolated from the path of
RF/microwave signals so that none of the signals leak or couple
improperly to the DC source or worse yet, to ground. To
successfully achieve such a task, we use several schemes which can
be briefly stated as follows:
a. Connect an inductor, commonly referred to as an RF choke
(RFC), between the DC source and the RF/MW circuitry. One
simple method to implement an RF choke in practice, is through
the use of a "ferrite bead".
b. Connecting a quarter-wave transformer between the DC source
and the RF/MW circuitry. The characteristic impedance (ZO) of
the transformer should preferably be very high (i.e., ZO>>1),
since it will create a high impedance path for any traveling
RF/MW signal.
c. Connecting a high-value capacitor (as a load) to the quarterwave transformer to effectively short out any residual RF/MW

Matthew M. Radmanesh, Ph.D. 361

signal that would leak into the DC circuitry. The use of a highvalue capacitor as a load for the quarter-wave transformer,
effectively creates an open circuit at the input end where it is
connected to the RF/MW circuitry.
Combining all of the three schemes into any RF/MW design
simultaneously would guarantee a high degree of isolation between
the DC and RF/MW circuitry and proper operation of the amplifier
(see Figure 15.2).

Symmetrical stubs may also be used for better input VSWR as

shown in Figure 15.3. This arrangement will minimize transition
interactions between shunt and series TLs.

RF & Microwave Design Essentials

362

15.3.2 Amplifier's RF/MW Circuit Design:

To design an RF/Microwave circuit, we need to follow these steps:
Step 1. Based on the microwave amplifiers specifications, select an
appropriate device e.g. if gain (G) is given choose a transistor with
typical |S21/S12|>G that is in the desired frequency range. Or, if the
noise figure (Fo) is given make sure that it is greater than the Fmin of
the selected transistor, i.e., Fo>Fmin.
Step 2. Bias the transistor in midrange of IC-VCE curves (BJT) or
ID-VDS curves (FET).
Step 3. Measure the S-parameters of the transistors at the selected
Q-point.
Step 4. Check stability conditions( K>1, ||<1) at a particular
frequency. If the condition is not met, draw the input and output
stability circles and determine the stable regions.
Step 5. There are two cases:
a. If S12=0, then use unilateral design formulas; Otherwise,
b. For S120, compute the unilateral figure of merit (U) and find
the error range, and if small enough, then use unilateral
assumption. Otherwise, use bilateral analysis and design
formulas.
Step 6. The input and output matching networks should now be
designed based on any one of the following requirements:
a. Narrow-band Amplifier design (NBA),
b. High-gain amplifier design (HGA),
c. Maximum-gain amplifier design (MGA)--This is a special case
of HGA,
d. Low-noise amplifier design (LNA)
e. Minimum-noise amplifier design (MNA)--This is a special case
of LNA,
We will discuss the design techniques and criteria for each one of
the categories (a)-(e), in the next sections. A summary of all of the
steps taken in the design of RF/microwave amplifiers is shown in
Figure 15.4.

Matthew M. Radmanesh, Ph.D. 363

Figure 15.4

Overall view of design steps.

15.4 DESIGN OF DIFFERENT TYPES OF

AMPLIFIERS
Now we will discuss the exact steps required for the design of each
of the seven types of small-signal amplifiers (a)-(g), introduced in
the previous section. Each type of amplifier optimizes a particular
amplifier characteristic (such as gain, noise figure, etc.) which needs
to be dealt with properly in the design process.

RF & Microwave Design Essentials

364

15.4.1 Narrow-Band Amplifiers (NBA) Design

By definition, a narrow-band amplifier is an amplifier in which the
amplification takes place over a bandwidth which is 10% (or less) of
the center frequency of operation.
Since the small signal amplifiers (b)-(e) normally have a narrow
bandwidth of less than 10% of the center frequency, therefore they
may be referred to as "narrow-band amplifiers" as well.
This means that the design considerations and techniques which will
be discussed for small signal amplifiers (b)-(e), also apply to the
design of narrow-band amplifiers where the S-parameters, the noise
figure and other parameters are measured at the center of the
frequency bandwidth

15.4.2 High-Gain Amplifier (HGA) design

High gain amplifier design requires a specific gain, which is not
equal to the maximum gain available from the amplifier. Therefore,
for high gain operation, we need to draw input and output constant
gain circles and select S and L, which are appropriate points on the
respective constant-gain circles located in stable regions.
Once S and L are selected, the rest of the design reduces to the
design of appropriate input and output matching networks.
_______________________________________________________
Example 15.1
Design a high-gain amplifier for a power gain of 15 dB at the
frequency of 3 GHz, if the selected bipolar transistor has the
following S-parameters (at VCE=4 V and IC=5 mA):

[S] =

0.7 155

4180

0.51 20
0

Solution:
1. Using the selected transistor, we need to check the stability
condition first:

Matthew M. Radmanesh, Ph.D. 365

S12=0 K=
||=0.357 <1
Therefore the transistor is unconditionally stable.
2. Since S12=0, therefore this is a unilateral design.
GT=GSGOGL
GO=|s21|2=16=12 dB
Since the transistor can provide a gain of 12 dB, thus we need to
obtain the remaining 3dB from the input and output matching
network as follows:
GS,max=1/(1-|S11|2)=1.96=2.9 dB
GL,max=1/(1-|S22|2)=1.35=1.3dB
From GS,max and GL,max values, we assign the following gain values
to the input and output matching networks:
GS=2 dB, and GL=1 dB.
3. Input 2-dB constant gain circle and output 1-dB constant gain
circle is now plotted:
a. GS=2 dB=1.59
gS=1.59/1.96=0.81
CgS=0.63155
RgS=0.25
b. GL=1 dB=1.26
gL=1.26/1.35=0.93
CgL=0.4820
RgL=0.20
The 2-dB and 1-dB constant gain circles are plotted in the Smith
chart as shown in Figure 15.5.
4. Since the transistor is unconditionally stable, therefore
S and L can be chosen anywhere on the constant-gain circles. One
possible design is given by:
a. Input matching network:
Shunt capacitance: CS=2.31 pF
Series inductor: LS=1.14 nH
b. Output matching network:
Series capacitance: CL=0.73 pF
Shunt inductor: LL=6.17 nH

RF & Microwave Design Essentials

366

5. The final schematic for the high-gain amplifier is shown in

Figure 15.6.

______________________________________________________

15.4.3 Maximum-Gain Amplifier (MGA) Design

Maximum-Gain amplifier design is a special case of High-Gain
Amplifier (HGA) design where the input and output gain circles are
reduced to single points and thus the design process is reduced to a
single design choice.
Assuming that the first five steps of the design are successfully
completed (see section 15.2), we now turn to step 6 and require the
following conditions to be met for this type of amplifier design:
S=
IN*
L=
OUT*
We need to consider two cases as follows:

a. UNILATERAL DESIGN
If the unilateral assumption holds valid, then GT/GTU,max (in dB)
should be within a tolerable error range, i.e.,

Matthew M. Radmanesh, Ph.D. 367

GT
G TU ,max

(dB ) 2

(15.1)

where -1 and 2 are the maximum tolerable errors (in dB). Thus for
maximum gain design, we can write:
(15.2a)
S=S11*
L=S22*
(15.2b)
This condition provides the maximum transducer gain (GTU,max):
(15.3)
GTU,max = GS,max GO GL,max
Where
1
G S ,max =
(15.4a)
2
1 S 11
G L ,max =

1
2

1 S 22

GO=|S21|2

(15.4b)
(15.4c)

b. BILATERAL DESIGN

In this case S and L should be chosen equal to MS and ML as

given in Chapter 13 by:
B 1 B 1 - 4 | C1 | 2
=
2C1
2

MS

(15.5a)

B 2 B 2 - 4 | C 2 |2
ML =
(15.5b)
2C 2
where
B1=1+|S11|2-|S22|2-|
|
(15.6a)
B2=1+|S22|2-|S11|2-|
|
(15.6b)
C1=S11-
S22*
(15.7a)
C2=S22-
S11*
(15.7b)
_______________________________________________________
EXAMPLE 15.2
A GaAs MESFET is measured to have the following S-parameters
for a mid range Q-point where VDS=5 V , ID=10 mA at 10 GHz with:
S11=0.55-150
2

RF & Microwave Design Essentials

368

S12=0.0420
S21=2.82180
S22=0.45-30
Using this transistor , design a microwave amplifier for maximum
power gain at 10 GHz.
Solution:
1. Check stability condition:
||=|S11S22- S12S21|=0.16 <1
K=2.28 >1
Therefore the device is unconditionally stable at 10 GHz.
2. Since S120, we need to check the unilateral figure of merit:
U=0.05
1/(1+0.05)2 < GT/GTU,max < 1/(1-0.05)2
-0.41 dB< GT/GTU,max < 0.45 dB
If we consider approximately 0.5 dB error as acceptable, then we
will use unilateral assumption and at the end will verify this
assumption when we find the maximum gain value.
3. GS,max=1.43 =1.55 dB
GO=|S21|2= 7.95 = 9 dB
GL,max=1.25= 0.97 dB
GTU,max= 1.55 + 9 + 0.97 =11.55 dB
4. Verification: Compared to a gain of 11.55 dB, an error of 0.5 dB
is justifiable!
5. Now we design the input and output matching network using a 50
system. From Figure 15.7 we can see that:

Matthew M. Radmanesh, Ph.D. 369

Input matching network:

First element- a shunt capacitor:
jC=j1.66/50 CS=0.59 pF
Second element- a series inductor:
jL=j0.71x50 LS=0.63 nH
Output matching network:
First element- a series capacitor:
1/jC=-j1.2x50 CL=0.59 pF
Second element- a series inductor:
1/jL=-j0.73/50 LL=1.21 nH
The final design is shown in Figure 15.8.

15.4.4 Low-Noise Amplifier (LNA) Design

In this type of amplifier design, the objective is not to exceed a
specified noise figure value while achieving the highest possible
gain. Since maximum power gain and minimum noise figure can not
be achieved simultaneously, in most cases we need to trade off one
for the other in order to obtain our design objective.
Given a design requirement with a specific noise figure and an exact
gain value, we need to carry out the following design procedure
(assuming that we have successfully performed the first five design
steps as outlined in section 15.2):
Step 1. Compute the allocated gain values for the input and the
output matching networks.

RF & Microwave Design Essentials

370

Step 2. Plot the source constant-gain circle and the constant-noise

circle on the same smith chart
Step 3. Choose a source constant-gain circle to intercept the desired
constant noise-figure circle.
Step 4. Using a value of S (selected anywhere on the constant gain
circle, such that it is between the intercept points and inside the
constant noise figure circle), the input matching network can be
designed.
Step 5. Plot the load constant-gain circle with the allocated gain of
step (a) above, and choose a L on this circle; finalize the amplifier
circuit with the design of output matching network.
_____________________________________________________
Example 15.3
Design a low noise amplifier (LNA) for a noise Figure of 3.5 dB and
a power gain of 16 dB. A bipolar transistor has been selected and is
biased at midrange for class A amplifier design: VCE=4 V, IC=30
mA. The S-parameters and the noise parameters at 1 GHz are as
follows:
0.7 37
0

[S] =

5.0180 0.51 20
Fmin=3 dB
opt =0.45180
Rn=4
Solution:
1. S12=0 K=
||=0.36 <1
Therefore the amplifier is unconditionally stable.
2. This is a unilateral amplifier design:
GT,max= GS,maxGOGL,max
GS,max=1/(1-|S11|)2=1.96=3 dB
GO=|S21|2=25=14 dB
GL,max=1/(1-|S22|)2=1.35=1.30 dB
GT,max=3+14+1.3=18.3 dB

Matthew M. Radmanesh, Ph.D. 371

Since the maximum gain is more than the required gain, thus the
transistor will serve our purpose splendidly!
2. We know the transistor provides 14 dB of gain, Thus to obtain the
required gain of 16 dB, we allocate the gain for each stage as
follows:
GO=14 dB (Transistor),
GS=1.22 dB (input matching network),
GL=0.78 dB (output matching network)
3. Plot the input the 1.22-dB gain circle:
GS=1.22 dB=1.32
gS=GS/GS,max =1.32/1.96=0.67
CS=0.56-155
RS=0.35
The 1.22-dB gain circle is plotted in Figure 15.9.
4. The 3.5 dB noise figure circle is also plotted in Figure 15.9 with:
F=3.5 dB=2.24
N=0.23
CF=0.37180
RF=0.39
5. The input matching network is obtained by choosing any point
located both on the 1.22 dB gain circle and inside (or on) the 3.5 dB
noise-figure circle. One such point is shown in Figure 15.9, giving
the following element values:
Shunt capacitor: jB=j0.75/50=j15X10-3 S Cs=2.39 pF
Series inductor: jX=j0.44x50=j22.0 Ls=3.5 nH

RF & Microwave Design Essentials

372

6. Plot the output 0.78 dB gain circle

GL=0.78 dB=1.20
gL=GL/GL,max =1.20/1.35=0.89
CL=S22*=0.4720
RS=0.25
The 0.78-dB gain circle is plotted in Figure 15.9.
7. The output matching network is obtained by choosing an arbitrary
point on the 0.78 dB gain circle giving the following element values:
Series capacitor: jX=j1.4X50=j70 CL=2.27 pF
Shunt inductor: jB=-j0.4/50=-j8.0X10-3 S LL=19.9 nH
8. The final amplifier Schematic is drawn in Figure 15.10

15.4.5 Minimum-Noise Amplifier Design

Minimum-Noise Amplifier (MNA) design is a special case of LowNoise Amplifier (LNA) design where the noise figure circles are
reduced to a single point (opt) and thus the design process is
reduced to a single design choice.
In this type of amplifier, after performing the first five steps of the
design (see section 15.2) we are ready now to design the input and
output matching networks.
From Chapter 14, we know that to achieve minimum noise, we need
to select
S = opt
(15.8)
and for best VSWR at the output, choose:

Matthew M. Radmanesh, Ph.D. 373

S 12 S 21 opt
L = OUT * = S 22 +

1 S 11 opt

(15.9)

Once S and L are known, the input and output matching networks
can be easily designed.
_______________________________________________________
Example 15.4
Design a minimum noise amplifier (see Figure 15.11) using a
MESFET with its S-parameters measured at a Q-point in midrange
of ID-VDS characteristic curves (i.e. VDS=8 V, ID=5 mA) at 12 GHz:
0.144 89
0.1546
2.34 84

[S] =

0.6137

Fmin=1.2 dB
O=0.47-65
Rn=40

FIGURE 15.11 Circuit for Example 15.4

Solution:
1. Since the transistor has already been selected and measured, we
check the stability condition as the next step:
K=1.10 > 1
||=0.37 <1
Thus the transistor is unconditionally unstable.
2. For minimum noise design we choose:
S=O=0.47-65
3. Now we choose L=OUT* for maximum power transfer:

374

RF & Microwave Design Essentials

*

S S
L = OUT * = S 22 + 12 21 O =0.303-85
1 S 11 O

4. Using the selected S and L we obtain:

GT=9.12 dB
5. Using established techniques, the input and output matching
network can be realized. The final circuit is shown in Figure 15.12.

Figure 15.12 Final design for Example 15.4.

______________________________________________________

Chapter 15- Symbol List

A symbol will not be repeated again, once it has been identified and
defined in an earlier chapter, with its definition remaining
unchanged.
BBA - Broadband amplifier
Fmin - Minimum Noise Figure
Fcas - Noise Figure for cascaded components
HGA - High gain amplifier
GaAs - Gallium Arsenic
GL,MAX - Maximum output matching network power gain
GS,MAX - Maximum available power gain
LNA - Low Noise Amplifier
MESFET - Metal Semiconductor Field Effect Transistor
Mmin - Minimum Noise measure
MGA - Maximum gain amplifier
MNA - Minimum Noise Amplifier
MSA - Multi-stage ampliifer
Te,cas - Equivalent noise temperature for cascaded components
OPT - Reflection coefficient for minimum noise match of a device.

Matthew M. Radmanesh, Ph.D. 375

IN,M1 - Reflection coefficient for input of stage one of a multi stage

amplifier.
OUT,M2 - Reflection coefficient for output of stage two of a multi
stage amplifier.

CHAPTER-15 PROBLEMS
15.1) Design a microwave transistor Amplifier for maximum gain
using a BJT whose S parameters in a 50 system are:
0
0.7 160

S=
0.5 20
585
15.2) The S parameters of a GaAs FET in a 50 system are given
by:
0
0.7 135

S=
0.8 60
4 60
a. Determine the unstable region in the smith chart.
b. Design the input matching network for GS=4 dB with the
greatest degree of stability.
c. Draw the complete RF amplifier schematic.
15.3) An amplifier is operating at 10 GHz using an FET device with
the following S parameters:
0
0.5 45

S=
0.8 160
5 30
Design the amplifier for maximum gain using a 50- input
and output and transmission lines as shown in Figure P15.6
a. Is the amplifiers stable?
b. What is the maximum gain in dB?
c. Design the input matching network to work for maximum
gain using series capacitors and shunt inductors to match the
50- line to ZS.
d. What are the values of in and Zin ?

RF & Microwave Design Essentials

376

15.4) A GaAs FET device has the following S parameters at 3 GHz:

0
0.3 60

S=
0.8 30
245
Design an amplifier for maximum gain using this transistor
and 50- input and output Transmission lines. The matching
network should use lumped elements as shown in Figure
P15.7.

15.5) Design a FET Amplifier for a minimum noise Figure and

maximum possible gain. Use open stub and quarterwavelength transformers for the matching networks. The Sparameters of the device at 3 GHz over a 100-MHz
bandwidth at a low-noise bias point (VDS=5 V, ID=10 mA) is
given by:
0
0.9 60

S=
0.5 45
290
Fmin=3 dB
opt=0.5-135
Rn=4

Matthew M. Radmanesh, Ph.D. 377

Show the final DC and RF schematic with the input and

output matching networks.
15.6) Design an amplifier with maximum unilateral gain at 6.0 GHz
using a GaAs FET transistor with the following S- parameters:
0
0.61 170

S=
0.72 83
2.24 32
Design L-section matching networks using lumped elements.
15.7) A GaAs FET has the following S parameters in a 50-
system:
0.7 170 0.0260
S=
0.8 70
3.560
Fmin=2 dB
opt=0.7120
Rn=15
Design an amplifier with minimum noise figure and
maximum possible gain. Use transmission lines in the
matching networks.
15.8) If the RF output voltage (vo), of a FET Amplifier is
represented by:
vo=C1vi+C2vi2+C3vi3
Where the input voltage is: vi=|vi|cost
Show that the gain of the Amplifier, G in dB, is given by:
G=20log10(C1+3C3|vi|2/4)
REFERENCES
[15.1] Anderson, R. W. S-Parameter Techniques for Faster, More

Accurate Network Design. Hewlett-Packard Application Note

951, 1967.
[15.2] Bahl, I. and P. Bhartia. Microwave Solid State Circuit Design.
New York: Wiley Interscience, 1988.
[15.3] Gonzalez, G. Microwave Transistor Amplifiers, Analysis
and Design, 2 nd ed. Upper Saddle River: Prentice Hall,
1997.
[15.4] Carson, R. S. High-Frequency Amplifiers. New York: Wiley
Interscience, 1975.

RF & Microwave Design Essentials

[15.5]

378

Chang, K. Microwave Solid-State Circuits and

Applications. New York: John Wiley & Sons, 1994.
[15.6] Froehner, W. H. Quick Amplifier Design with Scattering
Parameters, Electronics, October 1967.
[15.7] Ha, T. T. Solid-State Microwave Amplifier Design. New
York: John Wiley & Sons, 1987.
[15.8] Liao, S. Y. Microwave Circuit Analysis and Amplifier
Design. Upper Saddle River: Prentice Hall, 1987.
[15.9] Microwave Transistor Bias Configurations. Hewlett-Packard
Application Note 9441. April 1975.
[15.10] Pozar, D. M. Microwave Engineering, 2nd ed. New York:
John Wiley & Sons, 1998.
[15.11] Vendelin, George D., Anthony M. Pavio, and Ulrich L.
Rhode. Microwave Circuit Design, Using Linear and NonLinear Techniques. New York: John Wiley & Sons, 1990.

Matthew M. Radmanesh, Ph.D. 379

PART IV
____________________________________

APPENDICES
APPENDIX A PHYSICAL CONSTANTS
APPENDIX B INTERNATIONAL SYSTEM OF UNITS (SI)
APPENDIX C UNIT PREFIXES & CONVERSIONS
APPENDIX D GREEK ALPHABETS
APPENDIX E FRAGMENTED ENERGY FORMS
APPENDIX F CLASSICAL LAWS OF ELECTRICITY AND
MAGNETISM
APPENDIX G MATERIALS CONSTANTS
APPENDIX H CONVERSION AMONG TWO-PORT NETWORK
P A R AM E T E R S
APPENDIX I CONVERSION AMONG THE Y-PARAM ETERS
APPENDIX J MATHEMATICAL IDENTITIES
APPENDIX K DC BIAS NETWORKS FOR AN FET
APPENDIX L THE SCIENTISTS BEHIND THE DISCOVERIES
APPENDIX M DERIVATION OF THE CONSTANT GAIN AND
NOISE FIGURE CIRCLES
APPENDIX N CD ROM DOWNLOAD

RF & Microwave Design Essentials

380

Matthew M. Radmanesh, Ph.D. 381

APPENDIX A

Physical Constants

RF & Microwave Design Essentials

APPENDIX B

International System of
Units (SI)

382

Matthew M. Radmanesh, Ph.D. 383

APPENDIX C

Unit Prefixes &

Conversions

RF & Microwave Design Essentials

384

Matthew M. Radmanesh, Ph.D. 385

APPENDIX D

Greek Alphabets

RF & Microwave Design Essentials

APPENDIX G

Materials Constants

386

Matthew M. Radmanesh, Ph.D. 387

RF & Microwave Design Essentials

388

Matthew M. Radmanesh, Ph.D. 389

APPENDIX H

Conversion Among
Two-Port Network
Parameters

RF & Microwave Design Essentials

390

Matthew M. Radmanesh, Ph.D. 391

RF & Microwave Design Essentials

392

APPENDIX J

MATHEMATICAL
IDENTITIES
A) BINOMIAL FORMULAS

(x
y)2=x22xy+y2
(x
y)3=x3 3x2y + 3xy2 y3
(x
y)4=x4 4x3y +6x2y2 4xy3 + y4
Or, in general:

(x + y )n

= x n + nx n 1 +

n(n 1) n 2 2 n(n 1)(n 2) n 3 3

x y +
x y + ..... + y n
2!
3!

Where factorial n (n!) is defined by:

n!=1.2.3.......n
Note: Zero factorial is defined by: 0!=1

B) SPECIAL PRODUCTS

x2-y2=(x-y)(x+y)
x3-y3=(x-y)(x2+xy+y2)
x3+y3=(x+y)(x2-xy+y2)
x4-y4=(x2-y2)(x2+y2) =(x-y)(x+y)(x2+y2)

Matthew M. Radmanesh, Ph.D. 393

C) TRIGONOMETRIC FUNCTION RELATIONS

sin( x ) = sin x
cos( x ) = cos x
sin x
cos x
cos x
cot x =
sin x
1
sec x =
cos x
1
csc x =
sin x
2
sin x + cos 2 x = 1
tan x =

sin 2x=2sinx cosx

cos 2x=cos2x-sin2x=1-2sin2x=2cos2x-1
sin3x = 3sinx 4sin 3 x
cos3x = 3cosx + 4cos 3 x
1 cos2x
2
1
cos2x
+
cos 2 x =
2
sin 2 x =

3 sin x sin 3x
4
3
cos
x
+ cos 3x
cos 3 x =
4
xy
xmy
sin x sin y = 2 sin(
) cos(
)
2
2
x+y
xy
cos x + cos y = 2 cos(
) cos(
)
2
2
x+y
xy
cos x cos y = 2 sin(
) cos(
)
2
2
sin 3 x =

RF & Microwave Design Essentials

1
[cos(x y ) cos(x + y )]
2
1
cos x cos y = [cos( x y ) + cos( x + y )]
2
1
sin x cos y = [sin( x y ) + sin( x + y )]
2
sin x sin y =

D) HYPERBOLIC FUNCTION RELATIONS

e x e x
2
x
e + e x
cosh x =
2
sinh x e x e x
tanh x =
=
cosh x e x + e x

sinh x =

1
tanh x
1
sec hx =
cosh x
1
cosh x =
sinh x
2
cosh x sinh 2 x = 1
sinh( x ) = sinh x
cosh( x ) = cosh x
tanh( x) = tanh x
sinh( x y ) = sinh x cosh y cosh x sinh y
cosh( x y ) = cosh x cosh y sinh x sinh y
tanh x tanh y
tanh( x y ) =
1 tanh x tanh y
coth x =

394

Matthew M. Radmanesh, Ph.D. 395

sinh 2x = 2 sinh x cosh x

cosh 2x = cosh 2 x + sinh 2 x = 2 cosh 2 x 1 = 1 + 2 sinh 2 x
2 tanh x
1 + tanh 2 x
1
sinh 2 x = [cosh 2x 1]
2
1
cosh 2 x = [cosh 2x + 1]
2
(x y )
(x m y )
sinh x sinh y = 2 sinh
cosh
2
2
(x + y )
(x y )
cosh x + cosh y = 2 cosh
cosh
2
2
(x + y )
(x y )
cosh x cosh y = 2 sinh
sinh
2
2
tanh 2x =

1
[cosh( x + y ) cosh(x y )]
2
1
cosh x cosh y = [cosh( x + y ) + cosh( x y )]
2
1
sinh x cosh y = [sinh( x + y ) + sinh( x y )]
2
sinh x sinh y =

E) LOGARITHMIC RELATIONS

log a xy = log a x + log a y

log a

x
= log a x log a y
y

log a x y = y log a x
log b x
log b a
log a a = 1
F) COMPLEX NUMBERS
log a x =

RF & Microwave Design Essentials

396

x + jy = re j (conversion from rectangular to polar form)

and
re j = r(cos + sin ) (Eulers Identity)
Where,
j=

1,

r = x2 + y2
y
= tan 1 ( )
x

(re j ) n = r n e jn
(r1e j1 )(r2 e j 2 ) = r1r2 e j( 1 + 2 )

G) RELATIONSHIP BETWEEN EXPONENTIAL, TRIGONOMETRIC AND

HYPERBOLIC FUNCTIONS

e j = 1
e j / 2 = j
e j( x + 2k ) = e jx
e j[ x +( 2k +1) ] = e jx

e jx = cosx jsinx (Euler' s identity)

sinx =

e jx e jx
2j

e jx + e jx
2
jx
e e jx
tanx = j( jx
)
e + e jx

cosx =

sin( jx ) = j sinh x
cos( jx ) = cosh x
tan( jx) = j tanh x
sinh( jx) = j sin x
cosh( jx) = cos x
tanh( jx ) = j tan x

Matthew M. Radmanesh, Ph.D. 397

Where j = 1 and k is an integer.

H) DERIVATIVES

d n
du
(u ) = nu n 1
dx
dx
d
dv
du
(uv ) = u
+v
dx
dx
dx
d u
v(du / dx ) u(dv / dx )
( )=
dx v
v2
d
du
sin u = cos u
dx
dx
d
du
cos u = sin u
dx
dx
d
du
tan u = sec 2 u
dx
dx
d
du
cot u = csc 2 u
dx
dx
log a e du
d
log a u =
dx
u dx
d
1 du
log e u =
dx
u dx
d u
du
a = a u log e a
dx
dx
d u
du
e = eu
dx
dx

RF & Microwave Design Essentials

d
du
sinh u = cosh u
dx
dx
d
du
cosh u = sinh u
dx
dx
d
du
tanh u = sec h 2 u
dx
dx
d
du
coth u = csc h 2 u
dx
dx
I) INTEGRALS

u n +1
+C
n+1
udv = uv vdu

n
u du =

du
= log e | u | + C
u
au
u
a
du
=
+C

log e u

e du = e + C
log x = x log x x + C
u

sin udu = cos u + C

cos udu = sin u + C
tan udu = log cos u + C
cot udu = log sin u + C
sinh udu = cosh u + C
cosh udu = sinh u + C
e

398

Matthew M. Radmanesh, Ph.D. 399

tanh udu = log cosh u + C

coth udu = log sinh u + C
e

u sin u cos u
+C
2
u + sin u cos u
2
+C
cos udu =
2

sin

udu =

J) TAYLOR SERIES EXPANSION

f (x) |x=a = f (a) + f ' (a)(x a) +

e x |x=0 = 1 + x +

f '' (a)(x a)2

f (n1) (a)(x a)n1
+ +
+
2!
(n 1)!

x2 x3
+
+
2! 3!

sinx|x=0 =x- x3/3! + x5/5!- x7/7! +.

cosx| x=0 =1- x2/2! + x4/4!- x6/6! +.
ln(1+x)|x=0 =x- x2/2! + x3/3!- x4/4! +.
K) EQUATION OF A CIRCLE

(x-a)2 + (y-b)2=R2
where (a,b) is the center of the circle having a radius R.

RF & Microwave Design Essentials

400

APPENDIX K

DC Bias Networks for an

FET
There are five basic DC bias networks for an FET that uses one or
two power supplies as described next.

___________________________________________________________________

Vendelin, G. D. Five Basic Bias Design for GaAs FET Amplifiers,

Microwaves & RF, February 1978. (Reproduced with the permission of
Microwaves & RF.)

Matthew M. Radmanesh, Ph.D. 401

RF & Microwave Design Essentials

402

Matthew M. Radmanesh, Ph.D. 403

APPENDIX L

APPENDIX M

Derivation of the Constant

Gain & Noise Figure
Circles
From Chapter 13 the unilateral gain equations are given by:
GS
1 | S | 2
gS =
=
(1 | S 11 | 2 )
(M.1)
2
G S ,max | 1 S 11 S |
gL =

GL
1 | L | 2
(1 | S 22 | 2 )
=
G L ,max | 1 S 22 L | 2

(M.2)

We will derive the constant gain circle equation in the S plane

using Equation (M.1). A similar procedure could be utilized for
Equation (M.2) in the L plane.
We will present the general equation of a circle first, and then
proceed to obtain the desired derivation from Equation (M.1).

a. GENERAL EQUATION OF A CIRCLE

RF & Microwave Design Essentials

404

The general equation for a circle in the S plane, with radius (Rgs)
and a center at (Cgs), is given by:
|
S -Cgs |=Rgs
(M.3)
Squaring both sides of Equation (M.3), we have:
|
S -Cgs |2=Rgs2

(M.4)

Equation (M.4) can be rewritten as:

(
S -Cgs )(
S -Cgs )*=Rgs2
S -Cgs )*=Rgs2
(
S -Cgs )(

(M.5)

Applying the conjugate operation and Multiplying the terms in

Equation (M.5) yields the desired equation form for a circle as:
|
S|2 -Cgs*S -CgsS*=Rgs2 - |Cgs|2
(M.6)

b. CONSTANT GAIN CIRCLES

Multiplying the terms, Equation (M.1) can be written as:
gs(1+|S11S|2 -S11S -S11*S*)=1-|
S|2 -|S11|2+|
S|2|S11|2

(M.7)

Separating the various terms, we can write Equation (M.7) as:

|
S|2(1 -|S11|2 +gs|S11|2) - gs S11S - gs S11*S*=1- gs-|S11|2 (M.8)
Dividing both sides by the |S|2 coefficient, Equation (M.8) can be
recasted as:
*
2

*
g S S 11
g S S 11
2

= 1 g S | S 11 |
S
2
1 S 2 (1 g ) S 1 S 2 (1 g ) S
1 S 11 (1 g S )
11
S
11
S

(M.9)
Comparing Equation (M.6) with (M.9), we obtain:
g S S 11 *
C gS =
(M.10)
1 | S 11 | 2 (1 g S )

R gs C gs
2

1 g S S 11

1 | S 11 | 2 (1 g S )

(M.11)

Matthew M. Radmanesh, Ph.D. 405

Substituting for Cgs from Equation (M.11) in (M.11), we obtain Rgs

as:
(1 g S ) (1 | S 11 | 2 )
RgS =
(M.12)
1 | S 11 | 2 (1 g S )
Equations (M.10) and (M.12) provide the equations for Cgs and Rgs
as exactly given in Chapter 13.

c. CONSTANT NOISE FIGURE CIRCLES

The preceding procedure can also be used successfully to derive the
constant noise figure circle's center and radius (CF, RF). From
Chapter 14 we can write:
N=

S opt
1 S

CF =

(M.13)

opt

+ S opt

1+ N

)+

opt

1+ N

opt

N
1+ N

(M.14)

(M.15)

N +1
2

RF =

N 2 + N(1 opt )
1+ N

(M.16)

RF & Microwave Design Essentials

406

APPENDIX N

CD ROM Download
A. INSTRUCTIONS
1. To Download the CD ROM, please type in the following link
exactly:

http://www.csun.edu/~matt/RFMWDesignEssentials.zip
2. Once a pop up window shows up, click on open to unzip the
file. Make sure you have the WinZip software to unzip the files
properly.
3. After unzipping the files, create a folder called E-book in the C:
drive.
4. Save all of the files in this folder.
5. To start the software, double click on the Microsoft file called
StartMenu file.
NOTE: There is a stand-alone pdf file called Smith ZY_chart.pdf
included in the downloaded bundle of files, which can be printed out
and used for solving problems that require a ZY Smith Chart.

B. MAIN FEATURES
A CD containing software in the form of an electronic book (E-book),
which contains all numerical examples from the text, is bound into
the back of each textbook. The solutions are programmed using
Visual Basic software, which is built into the Microsoft Excel

Matthew M. Radmanesh, Ph.D. 407

application software. The main features of this CD are as follows:

1. It is a powerful interactive tool for learning the textbook content

and also for solving the numerical problems.

2. The software includes 90 solved problems based on the numerical

examples in the book.

3. The big advantage of the interactive software tool is its use of live
math. Every number and formula is interactive. The reader can
change the starting parameters of a problem and watch as the final
results change before his or her eyes. This feature allows the reader
to experiment with every number, formula, etc.
4. Each solved problem becomes a worksheet that the reader can
modify to solve dozens of related problems.
5. The electronic book takes advantage of the powerful Microsoft
Excel environment to perform many tedious and complicated RF
and microwave design calculations (usually using complex
numbers), allowing the student to focus on the essential concepts.
6. This is an excellent tool for students, engineers, and educators, to:
a. Understand the fundamentals and practical concepts of RF
and Microwaves and,
b. Encourage applications and new RF/Microwave circuit
designs using the concepts presented in the book.

C. HOW TO START THE PROGRAM

The following steps need to be carried out before the software is
ready to use:
1. Either one of the following two methods may be used to
utilize the contents of the CD-ROM:
a. Read all the files directly from the CD-ROM or,
b. Create a folder called E-book in the C: Drive. Copy the
entire content of the CD-ROM into the folder entitled E-book.
2. Open Microsoft Excel 2000 software (or Excel 97 with SR-1
or SR-2 revision) and open the E-book folder.
3. You may begin the program by double clicking on the StartMenu
file.
NOTE: You may double click on the Start Menu file directly as a
shortcut without opening the Excel software.

RF & Microwave Design Essentials

408

4. Once the StartMenu file opens up, click on Analysis Hub

and then open up the about the CD file at the top left corner by
clicking on it.
5. Carefully read all the information in the about the CD file
and close it by either
a. holding the ALT + TAB keys and choosing EXCEL or,
b. clicking return to Start Menu ARROW to return to the
E-book Start Menu.
6. Turn on both Analysis Toolpack and Analysis Toolpack VBA as
discussed below.
7. Proceed to the desired example by clicking on it and selecting Enable
Macros.

D. HOW TO USE THE E-BOOK SOFTWARE

Before proceeding to the worked-out examples, we need to select from
the toolbar menu Tools, Add-ins, and, from the dialog box, select
both Analysis Tool-pack and Analysis ToolpackVBA in order
to set up the software properly.
When a particular example is selected and clicked for interactive
use, the user will encounter a dialogue box where Enable Macros
must be selected. When the example is opened, the user will observe
that each numerical example consists of several sections, which can
be briefly summarized as:
1. Problem statement: Word for word text taken from the book that
describes the nature of the problem.

2. Input data: This section provides all the manipulatable data,

which the reader may have at his or her own disposal to vary
interactively and experiment with, in order to examine different
scenarios and obtain answers to what if questions. Inside the input
data box, the user may change the values of the parameters only, and
not any of the units. The user should type the new value in the
appropriate box and press enter/return to observe the desired change.
This is the only place where the user is allowed to make any changes
to the software.

Matthew M. Radmanesh, Ph.D. 409

3. Output data: This section contains a step-by-step solution of the

problem as well as easy explanations provided for quick assimilation
of the results. Most of the complicated calculations are done in
complex numbers using the Visual Basic programming technique,
which is part of the Microsoft Excel software.
4. Problem format and color codes: All problems are formatted and
color-coded in the same manner throughout the software. This is done
for the users easy recognition and reference, and is delineated as
follows:
Example xx.xx
Cyan
Problem text
Tan

Solution:

Navy Blue

Input data (interactive part)

Heading
Content

Red
Turquoise/Brown

Output data
Heading
Content
Interactive Answers
Caution
Note/Conclusion
Reference

Sky Blue
Yellow
Green
Red
Pink
Violet

E. SOFTWARE KNOWLEDGE REQUIRED

A rudimentary knowledge of Microsoft Excel is required to operate
the software successfully. The user does not need to know Visual
Basic programming techniques to work with the examples solutions
interactively.

F. MINIMUM SOFTWARE/HARDWARE REQUIREMENTS

The user needs to have the following:
1.
Hardware requirements: A personal computer (PC) with a
Pentium chip, preferably.

RF & Microwave Design Essentials

2.

410

Software requirements: Windows 95/98/XP/NT operating

system and Microsoft Excel 2000/2002 (or Excel 97 with:
SR-1 or SR-2 revisions).

NOTE 1: To obtain the Service Release 1 or 2 (SR-1 or SR-2), the

user should download the required software from the following
Website: http://www.microsoft. com/
Go to the search link and find the SR-1 or SR-2 revision, which is
suitable to the version of Microsoft Office that you own. Download
and install the SR-1 or SR-2 upgrade to repair all known bugs in
Microsoft Excel 97. Without this correction, Excel 97 gives incomplete
values for the worked-out examples in the textbook CD.
NOTE 2: If you own Microsoft Excel 2000/2002, please ignore
Note 1. Install the textbook CD directly without any changes to the
Microsoft Excel software using the procedure outlined in the previous
section. If you experience any problems, you need to download and
install the SR-1A revision for Microsoft Excel 2000/2002 from the
site mentioned previously.

G. TROUBLESHOOTING PROBLEMS
If the following problems occur, you may correct them as follows:
1. If ###### appears in place of a numerical answer, it means that
the cell is too small and you have to resize that cell in order to display
the final numerical result correctly. To resize the cell, go to the Excel
toolbar menu and select Format, Column, and Autofit Selection.
2. If #VALUE! appears, it means that any of the following
conditions may have occurred:
a. Divide by zero.
b. Negative number under a square root.
c. The number is out of range.
d. Excel 97 software is not used with SR-1 or SR-2 revision.
e.
The Analysis Toolpack and Analysis Toolpack VBA are not
turned on.*
* You need to select from the toolbar menu Tools, Add-ins, and from
the dialog box select Analysis Toolpack and Analysis ToolpackVBA in

Matthew M. Radmanesh, Ph.D. 411

order to set up the software properly.

NOTE: There is a stand-alone pdf file called Smith ZY_chart.pdf

included in the downloaded bundle of files, which can be printed out
and used for solving problems that require a ZY Smith Chart.

RF & Microwave Design Essentials

412

Matthew M. Radmanesh, Ph.D. 413

Glossary of Technical
Terms
The following glossary supplements the presented materials in
the text, but does not replace the use of an unabridged technical
dictionary, which is a must for mastery of sciences.

A
Absolute
a) That which is without reference to anything else and thus not
comparative or dependent upon external conditions for its existence
(opposed to relative), b) That which is free from any limitations or
restrictions and is thus unconditionally true at all times.
Absolute Temperature Scale
A scale with which temperatures are measured relative to absolute
zero (the temperature of 273.15 C or 459.67 F or 0 K). The
absolute temperature scale leads to the absolute temperatures, which
are: a) The temperature in Celsius degrees, relative to 273.15 C
(giving rise to the Kelvin scale), and b) The temperature in
Fahrenheit degrees, relative to 459.67 F (giving rise to the
Rankine scale). At absolute zero temperature, molecular motion
theoretically vanishes and a body would have no heat energy. The
absolute zero temperature is approachable but never attainable. See
also Temperature.

RF & Microwave Design Essentials

414

Active Device
An electronic component such as a transistor that can be used to
produce amplification (or gain) in a circuit.
Admittance
The measure of ease of AC current flow in a circuit, the reciprocal of
impedance expressed in Siemens (symbol Y or y).
Ampere (A)
The unit of electric current defined as the flow of one Coulomb of
charge per second. Alternately, it can also be defined as the constant
current that would produce a force of 2x10-7 Newton per meter of
length in two straight parallel conductors of infinite length, and of
negligible cross section, placed one meter apart in a vacuum.
Ampere's Law
Current (either conduction or displacement) flowing in a wire or in
space generates a magnetic flux that encircles the wire in a
clockwise direction when the current is moving away from the
observer. The direction of the magnetic field follows the right hand
rule (This law may also be referred to as the law of magnetic field
generation).
Differential form: CurlH = J +
Integral form :

D
,
t

D
dS

t
S

H dl = I +
C

Amplitude
The extent to which an alternating current or pulsating current or
voltage swings from zero or a mean value.
Analog
Pertaining to the general class of devices or circuits in which the
output varies as a continuous function of the input.
Anode
The positive electrode of a device (such as a diode, etc.) toward
which the electrons move during current flow.

Matthew M. Radmanesh, Ph.D. 415

Application Mass
All of the related masses that are connected and/or obtained as a
result of the application of a science. This includes all physical
devices, machines, experimental setups, and other physical materials
that are directly or indirectly derived from and are a result of the
application. In this book when we say application mass, we really
mean "technical application mass." See also Generalized
application mass, Technical application mass and personalized
application mass.
Attenuation
The decrease in amplitude of a signal during its transmission from
one point to another.
Attenuation Constant
The real component of the propagation constant.
Attenuator
A resistive network that provides reduction of the amplitude of an
electrical signal without introducing appreciable phase or frequency
distortion.
Average Power
The power averaged over one cycle.
Axiom
A self-evident truth accepted without proof.

B
Bias
The steady and constant current or voltage applied to an electrical
device to establish an operating point for proper operation of the
device.
Bidirectional
Responsive in both directions.
Bilateral
Having a voltage-current characteristic curve that is symmetrical
with respect to the origin. If a positive voltage produces a positive

RF & Microwave Design Essentials

416

current magnitude, then an equal negative voltage produces a

negative current of the same magnitude.
Brewster Angle
The angle of incidence of light reflected from a dielectric surface at
which the reflection coefficient becomes zero when the lights
electrical field vector lies in the plane of incidence (parallel
polarization). In other words, if a parallel polarized wave is incident
at a dielectric surface at the Brewster angle, all of the wave will be
transmitted through and there will be no reflection. Generally
speaking, the concept of Brewster angle applies to any electronic
wave of any frequency, not just light waves [B=tan-1(2/1)1/2].

C
Capacitance
The property that permits the storage of electrically separated
charges when a potential difference exists between two conductors.
The capacitance of a capacitor is defined as the ratio between the
electric charge of one electrode, and the difference in potential
between the electrodes.
Capacitor
A device consisting essentially of two conducting surfaces separated
by an insulating material (or a dielectric) such as air, paper, mica,
etc., that can store electric charge.
Cathode
The portion or element of a two-terminal device that is the primary
source of electrons during operation.
Cavity (Also Called a Cavity Resonator)
A metallic enclosure inside which resonant fields at microwave
frequencies are excited in such a way that it becomes a source of
electromagnetic oscillations frequencies.
Cell
A single and basic unit for producing electricity by electrochemical
or biochemical action. For example, a battery consists of a series of
connected cells.

Matthew M. Radmanesh, Ph.D. 417

Celsius (C)
1/100th of the temperature difference between the freezing point of
water (0C) and the boiling point of water (100C) on the Celsius
temperature scale given by:
T(C)=T(K)-273.15 =

5
{T(F)32}.
9

Characteristic Impedance
The driving-point impedance of a transmission line if it were of
infinite length. This can also be defined as the ratio of the voltage to
current at every point along a transmission line on which there are
no standing waves. It is given in general by:

Z0= ( R + jL) /(G + jC)

Charge
A basic property of elementary particles of matter (electrons,
protons, etc.) that is capable of creating a force field in its vicinity.
The built-in force field is a result of stored electric energy.
Chip
A single substrate upon which all the active and passive circuit
elements are fabricated using one or all of the semiconductor
techniques of diffusion, passivation, masking, photoresist, epitaxial
growth, etc.
Circuit
The interconnection of a number of devices in one or more closed
paths to perform a desired electrical or electronic function.
Classical Mechanics (Also Called Classical Physics, NonQuantized Physics or Continuum Physics)
Is the branch of physics based on concepts established before
quantum physics, and includes materials in conformity with
Newton's mechanics and Maxwell's electromagnetic theory.
Coaxial Transmission Line (Also Called Coaxial Cable)
A concentric transmission line in which one conductor completely
surrounds the other, the two being separated by a continuous solid
dielectric or by dielectric spacers. Such a line has no external field
and is not susceptible to external fields.

RF & Microwave Design Essentials

418

Coulomb (C)
The unit of electric charge defined as the charge transported across a
surface in one second by an electric current of one ampere. An
electron has a charge of 1.602x10-19 Coulomb.
Coulomb's Laws
The laws that state that the force (F) of attraction or repulsion
between two electric charges (or magnetic poles) is directly
proportional to the product of the magnitude of charges, Q (or
magnetic pole strengths, M), and is inversely proportional to the
square of distance (d) between them; that is,
Electric: F =

Q1Q 2
,
4d 2

Magnetic: F =

M1M 2
4d 2

The force between unlike charges, Q1 and Q2 (or poles, M1 and M2)
is an attraction, and between like charges (or poles) is a repulsion.

Communication Principle (Also Called Universal

Communication Principle)
A fundamental concept in life and livingness that is intertwined
throughout the entire field of sciences that states for communication
to take place between two or more entities, three elements must be
present: a source point, a receipt point, and an imposed space or
distance between the two.
Complex Power
Power calculated based on the reactance of a component.
Component
A packaged functional unit consisting of one or more circuits made
up of devices, which in turn may be part of an operating system or
subsystem.
Conductivity
The ratio of the current density (J) to the electric field (E) in a
material. It represents the ability to conduct or transmit electricity.

Matthew M. Radmanesh, Ph.D. 419

Conductor
a) A material that conducts electricity with ease, such as metals,
electrolytes, and ionized gases; b) An individual metal wire in a
cable, insulated or un-insulated.
Curl Operation
Curl is an operation on a vector field, which creates another vector
whose magnitude measures the maximum net circulation per unit
area of the vector field at any given point and has a direction
perpendicular to the area, as the area size tends toward zero. The
cause of the curl of a vector field is a vortex source. For example
electric current (conduction or displacement) is the vortex source for
magnetic field.
Current
Net transfer of electrical charges across a surface per unit time,
usually represented by (I) and measured in Ampere (A). Current
density (J) is current per unit area.

D
DC (Also Called Direct Current)
A current which always flows in one direction (e.g., a current
delivered by a battery).
Decibel (dB)
The logarithmic ratio of two powers or intensities or the logarithmic
ratio of a power to a reference power, multiplied by 10. It is onetenth of an international unit known as Bel: N(dB)=10log10(P2/P1).
Device
A single discrete conventional electronic part such as a resistor, a
transistor, etc.
Diamagnetics
are materials (such as glass, wood, lead, sulfur and others), which
avoid magnetic lines of force.
Dichotomy
Two things or concepts that are sharply or distinguishably opposite
to each other.

RF & Microwave Design Essentials

420

Die (Also Called Chip)

A single substrate on which all the active and passive elements of an
electronic circuit have been fabricated. This is one portion taken
from a wafer bearing many chips, but it is not ready for use until it is
packaged and provided with terminals for connection to the outside
world.
Dielectric
A material that is a non-conductor of electricity. It is characterized
by a parameter called dielectric constant or relative permittivity (r).
Dielectric Constant
The property of a dielectric defined as the ratio of the capacitance of
a capacitor (filled with the given dielectric) to the capacitance filled
with air as the dielectric, but otherwise identical in geometry.
Diffraction
Is the redistribution of intensity of waves in space, which results
from the presence of an object (such as a grating, consisting of
narrow slits or grooves) in the path of the beam of light waves. This
shall split up the beam into many rays, causing interference and thus
producing patterns of dark and light bands downstream (i.e., regions
with variations of wave amplitude and phase).
Digital
Circuitry in which data-carrying signals are restricted to either of
two voltage levels.
Discovery
The gaining of knowledge about something previously unknown.
Discrete Device
An individual electrical component such as a resistor, capacitor, or
transistor as opposed to an integrated circuit that consists of several
discrete components.
Distributed Element
An element whose property is spread out over an electrically
significant length or area of a circuit instead of being concentrated at
one location or within a specific component.

Matthew M. Radmanesh, Ph.D. 421

Divergence
a) The emanation of many flows from a single point, or reversely,
the convergence of many flows to one point; b) (of a vector field, F)
The net outflux per unit volume at any given point in a vector field,
as the volume size shrinks to zero (symbolized by divF). The cause
of the divergence of a vector field is called a flow source. For
example, positive electric charge is the flow source for the electric
field and creates a net outflux of electric field per unit volume at any
given point.
Dual
Two concepts, energy forms or physical things that are of
comparable magnitudes but of opposite nature, thus becoming
counterpart of each other.
Duality Theorem
States that when a theorem is true, it will remain true if each quantity
and operation is replaced by its dual quantity and operation. In
circuit theory, the dual quantities are voltage and current and
impedance and admittance. The dual operations are series and
parallel and meshes and nodes.

E
Electric Charge (or Charge)
(Microscopic) A basic property of elementary particles of matter
(e.g., electron, protons, etc.) that is capable of creating a force field
in its vicinity. This built-in force field is a result of stored electric
energy. (Macroscopic) The charge of an object is the algebraic sum
of the charges of its constituents (such as electrons, protons, etc.),
and may be zero, a positive or a negative number.
Electric Current (or Current)
The net transfer of electric charges (Q) across a surface per unit
time.
Electric Field
The region about a charged body capable of exerting force. The
intensity of the electric field at any point is defined to be the force
that would be exerted on a unit positive charge at that point.

RF & Microwave Design Essentials

422

Electric Field Intensity

The electric force on a stationary positive unit charge at a point in an
electric field (also called electric field strength, electric field vector,
and electric vector).
Electrical Noise (or Noise)
Any unwanted electrical disturbance or spurious signal. These
unwanted signals are random in nature, and are generated either
internally in the electronic components or externally through
impinging electromagnetic radiation.
Electricity
Is a form of energy, which can be subdivided into two major
categories: a) Electrostatics, and b) Electrokinetics.
Electrodynamics
Is a scientific field of study dealing with the various phenomena of
electricity in motion, including the interactions between currentcarrying wires as well as the forces on current wires in an
independent magnetic field.
Electrokinetics
Is that broad and general field of study dealing with electric charges
in motion. It studies moving electric charges (such as electrons) in
electric circuits and electrified particles (such as ions, etc.) in electric
fields.
Electrolysis
The action whereby a current passing through a conductive solution
(called an electrolyte) produces a chemical change in the solution
and the electrodes.
Electrolyte
A substance that ionizes when dissolved in a solution. Electrolytes
conduct electricity, and in batteries they are instrumental in
producing electricity by chemical action.
Electrolytic Cell
In general, a cell containing an electrolyte and at least two
electrodes. Examples include voltaic cells, electrolytic capacitors,
and electrolytic resistors.

Matthew M. Radmanesh, Ph.D. 423

Electromagnetic (EM) Wave

A radiant energy flow produced by oscillation of an electric charge
as the source of radiation. In free space and away from the source,
EM rays of waves consist of vibrating electric and magnetic fields
that move at the speed of light (in vacuum), and are at right angles to
each other and to the direction of motion. EM waves propagate with
no actual transport of matter, and grow weaker in amplitude as they
travel farther in space. EM waves include radio, microwaves,
infrared, visible/ultraviolet light waves, X-ray, gamma rays, and
cosmic rays.
Electromagnetics
The branch of physics that deals with the theory and application of
electromagnetism.
Electromagnetism
a) Magnetism resulting from kinetic electricity; b) Electromagnetics.
Electron
A stable elementary particle of matter, which carries a negative
electric charge of one electronic unit equal to q= -1.602x10-19 C and
has a mass of about 9.11x10-31 kg and a spin of .
Electronics
The study, control, and application of the conduction of electricity
through different media (e.g., semiconductors, conductors, gases,
vacuum, etc.).
Electroplating
Depositing one metal on the surface of another by electrolytic
action.
Electrostatics
The branch of physics concerned with static charges and charged
objects at rest.
Elementary Particle
A particle, which can not be described as a compound of other
particles and is thus one of the fundamental constituents of all matter
(e.g. electron, proton, etc.).

424

RF & Microwave Design Essentials

Energy
The capacity or ability of a body to perform work. Energy of a body
is either potential motion (called potential energy) or due to its
actual motion (called kinetic energy).

F
Fahrenheit (F)
1/180th of the temperature difference between the freezing point of
water (32F) and the boiling point of water (212F) on the
Fahrenheit temperature scale.
T(F) =T(R) - 459.67 =

9
T(C)+32
5

Where R and C are symbols for degrees Rankin and Celsius,

respectively.
Farad (F)
The unit of capacitance in the MKSA system of units equal to the
capacitance of a capacitor that has a charge of one Coulomb when a
potential difference of one volt is applied.
Faraday's Law (also called the law of electromagnetic induction)
When a magnetic field cuts a conductor, or when a conductor cuts a
magnetic field, an electrical current will flow through the conductor
if a closed path is provided over which the current can circulate; i.e.,

Differential form: CurlE =

Integral form:

,
t

E dl = t dS = dt
C

Ferrimagnetics
Ferrimagnetics are materials with the relative permeability (r) much
greater than that of vacuum having r=1. Ferrimagnetic materials are
materials made of iron oxides (chemical formula: XFe2O3, where X
is a metal ion), where their internal magnetic moments are not all
aligned in one direction, that is to say some are aligned antiparallel,

Matthew M. Radmanesh, Ph.D. 425

but with smaller magnitudes, so that the net magnetic field output is
still much higher than a paramagnetic material. Examples of
ferrimagnetics include materials such as manganese-zinc ferrite,
barium ferrite, and a whole class of materials, having a high
electrical resistance, called ferrites. Ferrimagnetic materials exhibit
hysteresis, which is a type of material behavior characterized by an
inability to retrace exactly the input-output curve when the
magnetizing force is reversed. This nonlinear behavior is caused by
the fact that the material will retain some of the magnetic effects
internally (called the remnant magnetism) even when the external
magnetizing force is completely removed.
Ferromagnetics
ferromagnetics are materials with the relative permeability (r) much
greater than that of vacuum (r=1), the amount depending on the
magnetizing force. Ferromagnetic materials are a group of materials
whose internal magnetic moments align in a common direction such
as iron, nickel, cobalt, and their alloys. Ferromagnetic materials
exhibit hysteresis, which is a type of material behavior characterized
by an inability to retrace exactly the input-output curve when the
magnetizing force is reversed. This nonlinear behavior is caused by
the fact that the material will retain some of the magnetic effects
internally (called the remnant magnetism) even when the external
magnetizing force is completely removed.
Field
An entity that acts as an intermediary agent in interactions between
particles, is distributed over a region of space, and whose properties
are a function of space and time, in general.
Field Theory
The concept that, within a space in the vicinity of a particle, there
exists a field containing energy and momentum, and that this field
interacts with neighboring particles and their fields.
Flow
The passage of particles (e.g., electrons, etc.) between two points.
Example: electrons moving from one terminal of a battery to the
other terminal through a conductor. The direction of flows are from
higher to lower potential energy levels.

RF & Microwave Design Essentials

426

Force
That form of energy that puts an unmoving object into motion, or
alters the motion of a moving object (i.e., its speed, direction or
both). Furthermore, it is the agency that accomplishes work.
Frequency
The number of complete cycles in one second of a repeating
quantity, such as an alternating current, voltage, electromagnetic
waves, etc.

G
Gain
The ratio that identifies the increase in signal or amplification that
occurs when the signal passes through a circuit.
Gauss
The unit of magnetic induction (also called magnetic flux density) in
the cgs system of units equal to one line per square centimeter,
which is the magnetic flux density of one Maxwell per square
centimeter, or 10-4 Tesla.
Gauss's Law (electric)
The summation of the normal component of the electrical
displacement vector over any closed surface is equal to the electric
charges within the surface, which means that the source of the
electric flux lines is the electric charge; i.e.,
Differential form: Div D =
Integral form:

D dS = dv = Q
S

Gauss's Law (magnetic)

The summation of the normal component of the magnetic flux
density vector over any closed surface is equal to zero, which in
essence means that the magnetic flux lines have no source or
magnetic charge; i.e.,

Differential form: Div = 0 ,

Matthew M. Radmanesh, Ph.D. 427

Integral form:

B dS = 0
S

Generalized Application Mass (G.A.M.)

In general, is any created space, which contains created energies and
created matter of any form, shape or size existing as a function of
time. In simple terms, generalized application mass is any matter and
energy, condensed and packaged into an object form, which exists in
a time-stream (from its inception to now). The generalized concept
of application mass includes the entire mechanical space containing
all energies and matter such as electrons, atoms, molecules and all
the existing gigantic masses of planets, stars, galaxies, which are not
the direct byproduct of Mans sciences.
Generalized Ohm's Law
When dealing with linear circuits under the influence of time
harmonic signals, Ohm's law can be restated under the steady-state
condition in the phasor domain as V=ZI, where Z is a complex
number called impedance and V and I are voltage and current
phasors, respectively.
Gilbert (Gi)
The unit of magnetomotive force in the cgs system of units, equal to
the magnetomotive force of a closed loop of one turn in which there
is a current of 10/4 amperes. One Gilbert equals 10/4 Ampereturn.
Gradient (of a scalar function)
Gradient (of a scalar function) is a vector, which lies in the direction
of maximum rate of increase of the function at any given point and
therefore is normal to the constant-value surfaces. Mathematically, it
is a vector obtained from a real function f(x,y,z), whose components
are the partial derivatives of f(x,y,z), e.g., in Cartesian coordinate
system we can write: gradf=(f/x, f/y, f/z).
Ground
(a) A metallic connection with the earth to establish zero potential
(used for protection against short circuit); (b) The voltage reference
point in a circuit. There may or may not be an actual connection to
earth but it is understood that a point in the circuit said to be at

RF & Microwave Design Essentials

428

ground potential could be connected to earth without disturbing the

operation of the circuit in any way.

H
Henry (H)
The unit of self and mutual inductance in the MKSA system of units
equal to the inductance of a closed loop that gives rise to a magnetic
flux of one Weber for each ampere of current that flows through.
Hertz (Hz)
The unit of frequency equal to the number of cycles of a periodic
function that occur in one second.
Hole
A vacant electron energy state near the top of the valence band in a
semiconductor material. It behaves as a positively charged particle
having a certain mass and mobility. It is the dual of electron, unlike
a proton which is the dichotomy of an electron.
Hypothesis
An unproven theory or proposition tentatively accepted to explain
certain facts or to provide a basis for further investigation.

I
Impedance
The total opposition that a circuit presents to an AC signal, and is a
complex number equal to the ratio of the voltage phasor (V) to the
current phasor (I).
Incident Wave
A wave that encounters a discontinuity in a medium, or encounters a
medium having a different propagation characteristics.
Inductance (L)
The inertial property of an element (caused by an induced reverse
voltage), which opposes the flow of current when a voltage is
applied; it opposes a change in current that has been established.

Matthew M. Radmanesh, Ph.D. 429

Inductor
A conductor used to introduce inductance into an electric circuit,
normally configured as a coil to maximize the inductance value.
Input
The current, voltage, power, or other driving force applied to a
circuit or device.
Insulator
A material in which the outer electrons are tightly bound to the atom
and are not free to move. Thus, there is negligible current through
the material when a voltage is applied.
Integrated Circuit (IC)
An electrical network composed of two or more circuit elements on
a single semiconductor substrate.
Isolation
Electrical separation between two points.

J
Joule (J)
The unit of energy or work in the MKSA system of units, which is
equal to the work performed as the point of application of a force of
one Newton moves the object through a distance of one meter in the
direction of the force.
Junction
A joining of two different semiconductors or of semiconductor and
metal.
Junction Capacitance
The capacitance associated with a junction such as the capacitance
of a region of transition between p- and n-type semiconductor
materials.

RF & Microwave Design Essentials

430

K
Kelvin (K)
The unit of measurement of temperature in the absolute scale (based
on Celsius temperature scale), in which the absolute zero is at
273.15 C. It is precisely equal to a value of 1/273.15 of the absolute
temperature of the triple point of water, being a particular pressure
and temperature point, 273.15 K, at which three different phases of
water (i.e., vapor, liquid, and ice) can coexist at equilibrium. See
also temperature.
Kinetic
(Adjective) Pertaining to motion or change. (Noun) Something
which is moving or changing constantly such as a piece of matter.
Kinetic Energy (K.E.)
The energy of a particle in motion. The motion of the particle is
caused by a force on the particle.
Kirchhoff's Current Law (KCL)
The law of conservation of charge that states that the total current
flowing to a given point in a circuit is equal to the total current
flowing away from that point.
Kirchhoff's Voltage Law (KVL)
An electrical version of the law of conservation of energy that states
that the algebraic sum of the voltage drops in any closed path in a
circuit is equal to the algebraic sum of the electromotive forces in
that path.
Knowledge
Is a body of facts, principles, data, and conclusions (aligned or
unaligned) on a subject, accumulated through years of research and
investigation, that provides answers and solutions in that subject.

Matthew M. Radmanesh, Ph.D. 431

L
Law
An exact formulation of the operating principle in nature observed to
occur with unvarying uniformity under the same conditions.
Law of Conservation of Energy (Excluding All Metaphysical
Sources of Energy)
This fundamental law simply states that any form of energy in the
physical universe can neither be created nor destroyed, but only
converted into another form of energy (also known as the principle
of conservation of energy).
Leyden Jar
The first electric capacitor (or condenser) capable of storing charge;
it consists of a glass jar with a coat of tin foil outside and inside and
a metallic rod passing through the lid and connecting with the inner
tin lining. It is named after the city of Leyden (also written as
Leiden) in Holland, where it was invented.
Light Waves
Electromagnetic waves in the visible frequency range, which ranges
from 400 nm to 770 nm in wavelength.
Linear Network
A network in which the parameters of resistance, inductance, and
capacitance of the lumped elements are constant with respect to
current or voltage, and in which the voltage or current sources are
independent of or directly proportional to other voltages and currents
or their derivatives, in the network.
Load
The impedance to which energy is being supplied.
Lossless
A theoretically perfect component that has no loss and hence,
transmits all of the energy fed to it.
Lumped Element
A self-contained and localized element that offers one particular
electrical property throughout the frequency range of interest.

RF & Microwave Design Essentials

432

M
Magnet
A piece of ferromagnetic or ferromagnetic material whose internal
domains are sufficiently aligned so that it produces a considerable
net magnetic field outside of itself and can experience a net torque
when placed in an external magnetic field.
Magnetic Field
The space surrounding a magnetic pole, a current-carrying
conductor, or a magnetized body that is permeated by magnetic
energy and is capable of exerting a magnetic force. This space can
be characterized by magnetic lines of force.
Magnetic Field Intensity (H)
The force that a magnetic field would exert on a unit magnetic pole
placed at a point of interest, which expresses the free space strength
of the magnetic field at that point (also called magnetic field
strength, magnetic intensity, magnetic field, magnetic force, and
magnetizing force).
Magnetostatics
The study of magnetic fields that are neither moving nor changing
direction.
Man
Homo sapiens (literally, the knowing or intelligent man); mankind.
Mathematics
Mathematics are short-hand methods of stating, analyzing, or
resolving real or abstract problems and expressing their solutions by
symbolizing data, decisions, conclusions, and assumptions.
Matter
Matter particles are a condensation of energy particles into a very
small volume.
Maxwell (Mx)
The unit for magnetic flux in the cgs system of units, equal to 10-8
Weber.

Matthew M. Radmanesh, Ph.D. 433

Maxwell's Equations
A series of four advanced classical equations developed by James
Clerk Maxwell between 1864 and 1873, which describe the behavior
of electromagnetic fields and waves in all practical situations. They
relate the vector quantities for electric and magnetic fields as well as
electric charges existing (at any point or in a volume), and set forth
stringent requirements that the fields must satisfy. These celebrated
equations are given as follows:
Differential form

Integral form

D
,
t

H dl = I + t

1) Ampere's Law: CurlH = J +

2) Faraday's Law: CurlE =

,
t

3) Gauss's Law (electric): DivD = ,

E dl = t dS = dt
C

D dS = dv = Q
S

4) Gauss's Law (magnetic): Div = 0 ,

dS

B dS = 0
S

From these equations, Maxwell predicted the existence of

electromagnetic waves whose later discovery made radio possible.
He showed that where a varying electric field exists, it is
accompanied by a varying magnetic field induced at right angles,
and vice versa, and the two form an electromagnetic field pair that
could propagate as a transverse wave. He calculated that in a
vacuum, the speed of the wave was given by 1/(00), where 0 and
0 are the permittivity and permeability of vacuum. The calculated
value for this speed was in remarkable agreement with the measured
speed of light, and Maxwell concluded that light is propagated as
electromagnetic waves.
Mechanics
The totality of the three categories of application mass: a)
Generalized application Mass; b) Technical application mass, and c)
Personalized application mass. See also classical mechanics and
quantum mechanics.

RF & Microwave Design Essentials

434

Microelectronics
The body of electronics that is associated with or applied to the
realization of electronic systems from extremely small electronic
parts.
Microstrip Line
A microwave transmission line that is composed of a single
conductor supported above a ground plane by a dielectric.
Microwave Integrated Circuit (MIC)
A circuit that consists of an assembly of different circuit functions
that are connected by Microstrip transmission lines. These different
circuits all incorporate planar semiconductor devices, passive
lumped elements, and distributed elements.
Microwaves
Waves in the frequency range of 1 GHz to 300 GHz.
Millimeter Wave
Electromagnetic radiation in the frequency range of 30 to 300 GHz,
corresponding to wavelength ranging from 10 mm to 1 mm.
Model
A physical (e.g., a small working replica), abstract (e.g., a
procedure) or a mathematical representation (e.g., a formula) of a
process, a device, a circuit, or a system and is employed to facilitate
their analysis.
Monolithic Integrated Circuit
An integrated circuit that is formed in a single block or wafer of
semiconductor materials. The term is derived from Greek,
"monolithos", which means "made of one stone."
Monolithic Circuits
Are integrated circuits entirely on a single chip of semiconductor.
Monolithic Microwave Integrated Circuit (MMIC)
A microwave circuit obtained through a multilevel process approach
comprising of all active and passive circuit elements as well as
interconnecting transmission lines, which are formed into the bulk or
onto the surface of a semi-insulating semiconductor substrate by
some deposition scheme such as epitaxy, ion implantation,
sputtering, evaporation, diffusion, etc.

Matthew M. Radmanesh, Ph.D. 435

Monumental Discovery
Any of the six major un-ravelings or breakthroughs of knowledge
about a significant phenomenon in the field of electricity, which
shifted the subject in a substantial way and expanded all of the
hitherto knowledge amply.

N
Natural Laws
A body of workable principles considered as derived solely from
reason and study of nature.
Neper (Np)
A unit of attenuation used for expressing the ratio of two currents ,
voltages, or fields by taking the natural logarithm (logarithm to base
e) of this ratio. If voltage V1 is attenuated to V2 so that V2/V1= e-N,
then N is attenuation in Nepers (always a positive number) and is
defined by: N (Np)=loge(V1/V2)=ln(V1/V2), where V1>V2.
Neutron
One of uncharged stable elementary particles of an atom having the
same mass as a proton. A free neutron decomposes into a proton, an
electron, and a neutrino. A neutrino is a neutral uncharged particle
but is an unstable particle since it has a mass that approaches zero
very rapidly (a half-life of about 13 minutes).
Network
A collection of electric devices and elements (such as resistors,
capacitors, etc.) connected together to form several interrelated
circuits.
Newton (N)
The unit of force in MKSA system of units equal to the force that
imparts an acceleration of one m/s2 to a mass of one kilogram.
Noise
Random unwanted electrical signals that cause unwanted and false
output signals in a circuit.

RF & Microwave Design Essentials

436

Nomenclature
The set of names used in a specific activity or branch of learning;
terminology.
Nonlinear
Having an output that does not rise and fall in direct proportion to
the input.
Nucleus
The core of an atom composed of protons and neutrons, having a
positive charge equal to the charge of the number of protons that it
contains. The nucleus contains most of the mass of the atom, pretty
much like the sun containing most of the mass of the solar system.

O
Occams (or Ockham's) Razor Doctrine
A principle that assumptions introduced to explain a thing must not
be multiplied beyond necessity. In simple terms, it is a principle
stating that the simplest explanation of a phenomenon, which relates
all of the facts, is the most valid one. Thus by using the Occam's
razor doctrine a complicated problem can be solved through the use
of simple explanations, much like a razor cutting away all undue
complexities (after William of Occam, an English philosopher,
1300-1349, who made a great effort to simplify scholasticism).
Oersted (Oe)
The unit of magnetic field in the cgs system of units equal to the
field strength at the center of a plane circular coil of one turn and 1cm radius when there is a current of 10/2 ampere in the coil.
Ohm (
)
The unit of resistance in the MKSA system of units equal to the
resistance between two points on a conductor through which a
current of one ampere flows as a result of a potential difference of
one volt applied between the two points.

Matthew M. Radmanesh, Ph.D. 437

Ohm's Law
The potential difference V across the resistor terminals is directly
proportional to the electrical current flowing through the resistor.
The proportionality constant is called resistance (R); i.e., V=RI.
Ohm's Law can also be interpreted as the conversion of potential
energy (V) into kinetic energy (I), which is a simple statement
expressing the principle of conservation of energy.
Original Postulates
A series of exact postulate (space, energy, change) that have gone
into the construction of the physical universe. See primary
postulates.
Oscillator
An electronic device that generates alternating-current power at a
frequency determined by constants in its circuits.
Output
The current, voltage, power, or driving force delivered by a circuit or
device.

P
Paramagnetics
are materials (such as aluminum, beryllium, etc.), which accept
magnetism.
Particle
Any tiny piece of matter, so small as to be considered theoretically
without magnitude (i.e., zero size), though having mass, inertia and
the force of attraction. Knowing zero size is an absolute and thus
impossible in the physical universe, practical particles range in
diameter from a fraction of angstrom (as with electrons, atoms and
molecules) to a few millimeters (as with large rain drops).
Passive
A component that may control but does not create or amplify
electrical energy.
Perfect Conductor
Is a conductor having infinite conductivity or zero resistivity.

RF & Microwave Design Essentials

438

Personalized Application Mass (P.A.M.)

Is the category of application mass, which has been created and is
based solely upon the viewpoints own postulates and
considerations. Examples of this category include such things as
ones own customized possessions, any piece of artwork or music,
ones own body characteristics (such as hairdo, clothing, shape,
etc.), a books layout or cover design, so on and so forth. see also
application mass, Technical application mass, and Generalized
application mass.
Phase
The angular relationship of a wave to some time reference or other
wave.
Phase Constant
The imaginary component of the propagation constant for a traveling
wave at a given frequency.
Phasor
A result of a mathematical transformation of a sinusoidal waveform
(voltage, current, or EM wave) from the time domain into the
complex number domain (or frequency domain) whereby the
magnitude and phase angle information of the sinusoid is retained.
Physical Universe (Also Called Material Universe; The
Universe)
Is a universe based upon three postulates, called original postulates
(space, energy and change) and has four main components (matter,
energy, space and time).
Plane Wave
A wave whose wave fronts are plane surfaces and normal to the
direction of propagation.
Plating
See electroplating.
PN Junction
An abrupt transition between p-type and n-type semiconductor
regions within a crystal lattice. Such a junction possesses specific
electrical properties such as the ability to conduct in only one
direction, and is used as the basis for semiconductor devices, such as
diodes, transistors, etc.

Matthew M. Radmanesh, Ph.D. 439

Port
Access point to a system or circuit.
Postulate
a) (NOUN) is an assumption or assertion set forth and assumed to be
true unconditionally and for all times without requiring proof;
especially as a basis for reasoning or future scientific development;
b) (VERB) To put forth or assume a datum as true or exist without
proof.
Potential Difference (or Voltage)
The electrical pressure or force between any two points caused by
accumulation of charges at one point relative to another, which has
the capability of creating a current between the two points.
Potential Energy (P.E.)
Any form of stored energy that has the capability of performing
work when released. This energy is due to the position of particles
relative to each other.
Power
The rate at which work is performed; i.e., the rate at which energy is
being either generated or absorbed.
Primary Postulates
A series of four postulates derived from original postulates. These
postulates are responsible for the four basic components of the
physical universe: matter, energy, created space, and mechanical
time. See original postulates.
Principle
A rule or law illustrating a natural phenomenon, operation of a
machine, the working of a system, etc.
Processing
The act of converting material from one form into another more
desired form, such as in integrated circuit fabrication where one
starts with a wafer and through many steps ends up with a functional
circuit on a chip.
Propagation
The travel of electromagnetic waves through a medium.

RF & Microwave Design Essentials

440

Propagation Constant
A number showing the effect (such as losses, wave velocity, etc.) a
transmission line has on a wave as it propagates along the line. It is a
complex term having a real term called the attenuation factor and an
imaginary term called the phase constant.
Proton
An elementary particle, which is one of the three basic subatomic
particles, with a positive charge equivalent to the charge of an
electron (q= +1.602x10-19 C) and has a mass of about 1.67x10-27 kg
with a spin of . Proton together with neutron is the building block
of all atomic nuclei.
Pulse
A variation of a quantity, which is characterized by a rise to a certain
level (amplitude), a finite duration, and a decay back to the normal
level.
Pyramid of Knowledge
Workable knowledge forms a pyramid, where from a handful of
common denominators efficiently expressed by a series of basic
postulates, axioms and natural laws, which form the foundation of a
science, an almost innumerable number of devices, circuits and
systems can be thought up and developed. The plethora of the mass
of devices, circuits and systems generated is known as the
application mass, which practically approaches infinity in sheer
number.

Q
Quantum Mechanics (Also Called Quantum Physics or
Quantum Theory)
Is the study of atomic structure which states that an atom or
molecule does not radiate or absorb energy continuously. Rather, it
does so in a series of steps, each step being the emission or
absorption of an amount of energy packet (E) called a quantum.
Quantum physics is the modern theory of matter, electromagnetic
radiation and their interaction with each other. It differs from

Matthew M. Radmanesh, Ph.D. 441

classical physics in that it generalizes and supersedes it, mainly in

the realm of atomic and subatomic phenomena.
Quark
A hypothetical basic particle having a fraction of charge of an
electron (such as 1/3 or 2/3) from which many of the elementary
particles (such as electrons, protons, neutrons, mesons, etc.) may be
built up theoretically. No experimental evidence for the actual
existence of free quarks has been found.

R
Radio Frequency (RF)
Any wave in the frequency range of a few kHz to 300 MHz, at
which coherent electromagnetic radiation of energy is possible.
Rankine (R)
The unit of measurement of temperature in the absolute scale (based
on Fahrenheit temperature scale), in which the absolute zero is at
-459.67 F. See also temperature.
Reactance
Is a parameter that is the measure of the opposition to the flow of
alternating current (Symbolized by X).
Reactive Element
Is an element, which impedes the flow of current in a wire. An
inductor or a capacitor are reactive elements. A purely reactive
element does not dissipate energy as does a resistor, but stores it in
the associated electric and/or magnetic fields.
Rectifier
Is a device having an asymmetrical conduction characteristic such
that current can flow in only one direction through the device.
Reflected Waves
The waves reflected from a discontinuity back into the original
medium, in which they are traveling.
Reflection Coefficient
The ratio of the reflected wave phasor to the incident wave phasor.

RF & Microwave Design Essentials

442

Resistance
A property of a resistive material that determines the amount of
current flow when a voltage is applied across it. The resistor value is
dependent upon geometrical dimensions, material, and temperature.
Resistor
A lumped bilateral and linear element that impedes the flow of
current, i(t), through it when a potential difference, V (t ), is imposed
between its two terminals. The resistor's value is found by:
R=V(t)/i(t).
Resonant Frequency
The frequency at which a given system or circuit will respond with
maximum amplitude when driven by an external sinusoidal force.
Right-Hand Rule
For a current-carrying wire, the rule that if the fingers of the right
hand are placed around the wire so that the thumb points in the
direction of the current flow, the finger curling around the wire will
be pointing in the direction of the magnetic field produced by the
wire.

S
Science
A branch of study concerned with establishing, systematizing, and
aligning laws, facts, principles, and methods that are derived from
hypothesis, observation, study and experiments.
Semiconductor
A material having a resistance between that of conductors and
insulators, and usually having a negative temperature coefficient of
resistance.
Signal
An electrical quantity (such as a current or voltage) that can be used
to convey information for communication, control, etc.
Silicon (Si)
A semiconductor material element in column IV of the periodic table
used as in device fabrication.

Matthew M. Radmanesh, Ph.D. 443

Sinusoidal
Varying in proportion to the sine or cosine of an angle or time
function. For example, the ordinary AC signal is a sinusoidal.
Small Signal
A low-amplitude signal that covers such a small part of the operating
characteristic curve of a device that operation is nearly always
linear.
Solid-State Device
Any element that can control current without moving parts, heated
filaments, or vacuum gaps. All semiconductors are solid-state
devices, although not all solid-state devices (such as transformers,
ferrite circulators, etc.) are semiconductors.
Space (Also Called Created Space)
The continuous three-dimensional expanse extending in all
directions, within which all things under consideration exist.
Standing Wave
A standing, apparent motionless-ness, of particles causing an
apparent no out-flow, no in-flow. A standing wave is caused by two
energy flows, impinging against one another, with comparable
magnitudes to cause a suspension of energy particles in space,
enduring with a duration longer than the duration of the flows
themselves.
Standing Wave Ratio (SWR)
The ratio of current or voltage on a transmission line that results
from two waves having the same frequency and traveling in opposite
directions meeting and creating a standing wave.
Static
(Adjective) Pertaining to no-motion or no-change. (Noun)
Something which is without motion or change such as truth (an
abstract concept). In physics, one may consider a very distant star (a
physical universe object) a static on a short term basis, but it is not
totally correct because the distant star is moving over a long period
of time, thus is not truly a static but only an approximation, or a
physical analogue of a true static.

RF & Microwave Design Essentials

444

Subjective Time
Is the consideration of time in ones mind, which can be a nonlinear
or linear quantity depending on ones viewpoint.
Substrate
A single body of material on or in which one or more electronic
circuit elements or integrated circuits are fabricated.
Superposition Theorem
This theorem states that in a linear network, the voltage or current in
any element resulting from several sources acting together is the sum
of the voltages or currents resulting from each source acting alone,
while all other independent sources are set to zero; i.e.,
f(v1+v2++vn)=f(v1)+f(v2)+..+f(vn)
Supplemental Discovery
Any of the eight subordinate discoveries (along with their magnetic
duals), which fill in the gaps left behind by the six monumental
discoveries of electricity.
Switch
A mechanical or electrical device that completes or breaks the path
of the current or sends it over a different path.
Switching
Is the making, breaking, or changing of connections in an electronic
or electric circuit.
Symbiont
An organism living in a state of association and interdependence
with another kind of organism, especially where such association is
of mutual advantage, such as a pet. Such a state of mutual
interdependence is called "symbiosis."

T
Technical Application Mass (T.A.M.)
Is the category of man-made application mass that is produced
directly as a result of application of a science using its scientific
postulates, axioms, laws and other technical data. Examples include
such things as a television set, a computer, an automobile, a power

Matthew M. Radmanesh, Ph.D. 445

generator, a telephone system, a rocket, etc. See also Application

mass, Personalized application mass, and Generalized
application mass.
Technology
The application of a science for practical ends.
Temperature
The degree of hotness or coldness measured with respect to an
arbitrary zero or an absolute zero, and expressed on a degree scale.
Examples of arbitrary-zero degree scales are Celsius scale (C) and
Fahrenheit scale (F); and examples of absolute-zero degree scales
are Kelvin degree scale (based on Celsius degree scale) and Rankine
degree scale (based on Fahrenheit degree scale).
Tesla (T)
The unit of magnetic field in the MKSA system of units equal to one
Weber per square meter.
TEM (Transverse Electro-Magnetic) Wave
Waves having the electric and magnetic fields perpendicular to each
other and to the direction of propagation. These waves have no field
components in the direction of propagation.
Theorem
A proposition that is not self-evident but can be proven from
accepted premises and therefore, is established as a principle.
Theory
An explanation based on observation and reasoning, which explains
the operation and mechanics of a certain phenomenon. It is a
generalization reached by inference from observed particulars and
implies a larger body of tested evidence and thus a greater degree of
probability. It uses a hypothesis as a basis or guide for its
observation and further development.
Thermal Noise (Johnson Noise or Nyquist Noise)
The most basic type of noise that is caused by thermal vibration of
bound charges and thermal agitation of electrons in a conductive
material. This is common to all passive or active devices.

RF & Microwave Design Essentials

446

Time (Also Called Mechanical Time or Objective Time)

That characteristic of the physical universe at a given location that
orders the sequence of events on a microscopic or macroscopic
level. It proceeds from the interaction of matter and energy and is
merely an "index of change," used to keep track of a particle's
location. The fundamental unit of time measurement is supplied by
the earth's rotation on its axis while orbiting around the sun. It can
also alternately be defined as the co-motion and co-action of moving
particles relative to one another in space. See also subjective time.
Torque
A force that tends to produce rotation or twisting.
Transformer
An electrical device that, by electromagnetic induction, transforms
electric energy from one (or more) circuit(s) to one (or more) other
circuit(s) at the same frequency, but usually at a different voltage
and current value.
Transmission Line (T.L.)
Any system of conductors suitable for conducting electric or
electromagnetic energy efficiently between two or more terminals.
Transmitted wave
That portion of an incident wave that is not reflected at the interface,
but actually travels from one medium to another.
Two-port network
A network that has only two access ports, one for input or excitation,
and one for output or response.

U
Unidirectional
Flowing in only one direction (e.g., direct current).
Unilateral
Flowing or acting in one direction only causing a non-reciprocal
characteristic.

Matthew M. Radmanesh, Ph.D. 447

Universal Communication Principle (Also Called

Communication Principle)
A fundamental concept in life and livingness that is intertwined
throughout the entire field of sciences that states for communication
to take place between two or more entities, three elements must be
present: a source point, a receipt point, and an imposed space or
distance between the two.
Universe (Derived From Latin Meaning "Turned Into One", "A
Whole)
Is the totality or the set of all things that exist in an area under
consideration, at any one time. In simple terms, it is an area
consisting of things (such as ideas, masses, symbols, etc.) that can be
classified under one heading and be regarded as one whole thing.

V
Viewpoint
Is a point on a mental plane from which one creates (called
postulating viewpoint) or observes (called observing viewpoint) an
idea, an intended subject or a physical object.
Volt (V)
The unit of potential difference (or electromotive force) in the
MKSA system of units equal to the potential difference between two
points for which one Coulomb of charge will do one joule of work in
going from one point to the other.
Voltage
Voltage or potential difference between two points is defined to be
the amount of work done against an electric field in order to move a
unit charge from one point to the other.
Voltage Source
The device or generator connected to the input of a network or
circuit.
Voltage Standing Wave Ratio (VSWR)
The ratio of maximum voltage to the minimum voltage on a
transmission. The standing wave on a line results from two voltage

RF & Microwave Design Essentials

448

(or current) waves having the same frequency, and traveling in

opposite directions.

W
Wafer
A thin semiconductor slice of silicon or germanium on which
matrices of microcircuits or individual semiconductors can be
formed using manufacturing processes. After processing, the wafer
is separated into chips (or die) containing individual circuits.
Watt (W)
The unit of power in MKSA system of units defined as the work of
one joule done in one second.
Wave
A disturbance that propagates from one point in a medium to other
points without giving the medium as a whole any permanent
displacement.
Wave Propagation
The travel of waves (e.g., electromagnetic waves) through a
medium.
Waveguide
A transmission line comprised of a hollow conducting tube within
which electromagnetic waves are propagated.
Wavelength
The physical distance between two points having the same phase in
two consecutive cycles of a periodic wave along a line in the
direction of propagation.
Weber (Wb)
The unit of magnetic flux in the MKSA system of units equal to the
magnetic flux, which linking a circuit of one turn, produces an
electromotive force of one volt when the flux is reduced to zero at a
uniform rate in one second.
Work
The advancement of the point of application of a force on a particle.