Antenna Lecture 1

Introduction to Antennas
One Quarter Graduate Course in ECE

First, let’s establish some expectations. A graduate course in Antennas, in an Electrical Engineering
Department is similar to a course in “Birds” in a Biology department. The topic is way too large to be
covered in one quarter, one academic career, or one human lifetime. Graduate courses in Antennas
tend to fall into several categories, depending on the department and the professor. They may be fo-
cused on EMag theory if that is what the professor understands, Field Simulations if the prof is adept
in numerical electromagnetic code, giant radiotelescopes, arrays, arrays of giant radiotelescopes...
If you really want to study antennas at the graduate level, find a university where the author of one of
the classic texts has left a legacy: Ohio State, (John Kraus), Virginia Tech (Warren Stutzman), or Ari-
zona State (Constantine Balanis). It is worth noting that Warren Stutzman and Constantine Balanis
both earned their Doctoral degrees at Ohio State. There you will be immersed in a culture of antenna
theory, practice, and ongoing research. Devoting 5 years to graduate school in a department with a
half dozen active contributors to the IEEE Antennas and Propagation Society on the faculty is a good
start, and a decade or so into a career as an antenna designer you may feel you are beginning to
grasp the basics in one of the specialized realms of antenna theory and practice.

We are not going to do that in one quarter. But we are going to set you at the trailhead, and how far
you want to follow the path is up to you. In any introductory course on a topic with too much material,
my goal is to get you started, give you enough basics that you can launch your own investigations,
and not make you learn anything you have to unlearn later in order to proceed further down the path.

I don’t know where to send you to study antennas in 2020, but I will recommend the books by Stutz-
man, Kraus, and Balanis, in the order of my familiarity with them. Warren Stutzman is an old friend,
and I have tremendous respect for him. I missed the opportunity to interact with John Kraus during
his lifetime, but as an Electromagnetics professor my early career was dominated by interactions with
his students. I’ve met and enjoyed a few conversations with Constantine Balanis, and many of you
are familiar with his work through both his antenna textbook and remarkably clear and accessible
treatment of electromagnetics at the graduate level. Kraus’s text on Radio Astronomy is truly clas-
sic, and a compact and interesting introduction to a fascinating way to explore the universe. Browse
the used bookstores. If you study Antenna textbooks written prior to 1984 you will be immersed in
the fundamentals, before numerical electromagnetic code spun off the branch of electromagnetics
where computing power takes off where human analysis ends. In 2020 it is common, and in practice
necessary, to go immediately to the work station and run electromagnetic code rather than attempting
analysis with pen and paper.

But in an Antennas graduate course we have the opportunity, and obligation in my opinion, to at least
study the basic fundamentals. The combination of fundamental theory and years of simulations and
practice result in intuitions that fill in the blanks where both theory and simulation fail.

Some Necessary Review

The language of Electromagnetics involves E and H fields in time and three dimensional space. In
the past 130 years or so, we’ve developed a compact way to describe the way fields vary across
those four dimensions, using vector calculus. A brief review of the del operator and how it works will
be the subject of Thursday’s lecture.
Antenna Course Topics and Outline

Before we dive into Vector Calculus, lets divide the Portland State University ECE510 Antennas
course into separate topics.

Topic 1 The Electromagnetic Foundation of Antennas and the A Vector

This topic will be spread over most of the quarter, so we can take it in small doses and give it time to
sink in. But let’s begin by saying a little bit about the magnetic Hertz vector and why it is such a use-
ful way to treat the E and H fields around antennas. Introductory courses on Electromagnetic Wave
Propagation start with Maxwell’s equations, and after a few mathematical operations, invoke the Far
Field approximation to greatly simplify the math. But we don’t have the luxury of operating in the far
field when studying antennas, particularly antennas made up of configurations of wires inducing cur-
rents in each other. By using the A vector, we can greatly simplify the math needed to specify A any-
where near or far from from a current on a bit of wire. Conveniently, A is pointed in the same direction
as the wire, so it’s easier to visualize. A satisfies the same wave propagation differential equation, so
our intuitions are good, and has a familiar solution from Far Field wave propagation.

But here’s the thing. A still contains everything. We didn’t need to throw out any terms when making
a Far Field approximation, so it’s all still in there. By operating on A using vector calculus operations
like curl in spherical coordinates, we obtain both the near and far H field components (there are sev-
eral) and further vector calculus operations yeald all of the near and far E field components (there are
many). So you do the entire antenna problem using the A vector, finish it all up, obtain A somewhere
in space near, far or even inside an arrangement of wires...and then you do vector calculus opera-
tions on A to obtain H field components close to one of the wires, or the intense E field component
pointing straight off the end of a dipole.

E and H fields near, far, and inside antennas are the coolest ever application of 3 dimensional vector
calculus in rectangular, spherical and cylindrical coordinate systems, and that makes all those math
courses you had to take worthwhile. The first time you use vector calculus to predict some near field
antenna phenomenon you’ve observed in your backyard off to the side of a yagi antenna (where the
far field approximations say the fields should be zero), you feel like you can eat nails and spit them
point first into a target ten meters away.

So the bottom line is, Antenna EMAG theory is really, really cool...and really really hard. If you want
to impress the smartest people you know, tell them you are studying it. If you want to look like a fool,
tell them you understand it. Antenna theory is too ponderous to describe some things that are easy to
build and measure, and the underlying assumptions that we use to develop advanced EM simulators
are sometimes flawed, and often oversimplified. Approach EMAG antenna theory with curiousity and
humility, and you won’t be disappointed.

Topic 2 The language of antennas, definitions, analysis, and simulation

Advertisements for antennas are filled with numbers like gain, physical dimensions, operating fre-
quencies and bandwidths. Accurate antenna measurements are difficult, radio wave propagation in
random media is an entire discipline as messy as social science... but precise antenna measurments
are easy, and for decades marketing took advantage of the difference between precision and accu-
racy to fill antenna advertisements with gain numbers to tenths of a dB.
Then in 1957 the Soviet Union launched Sputnik I, and for the first time the world had an orbiting
antenna measurement facility in actual free space. In the satellite communications era, a 0.1 dB im-
provement in antenna gain results in a 0.1 dB increase in signal strength. It is no wonder that gifted
antenna folks like Warren Stutzman ended up in the Satellite Communications Group at Virginia Tech,
where we worked as colleagues in an organization called the Nasa Propagation Experimenters.

So if you want a quick handle on all the terminology of antennas: gain, effective aperture, antenna
noise temperature, beamwidth, sidelobes, aperture efficiency... an introductory book on Satellite Com-
munications is a good place to start. Satellite Communications people are engineers first and EMAG
theory is a tool they depend on but don’t automatically fall back on when they are trying to explain
something. The first paragraph in a Satellite Communications textbook is less likely to contain an
integral equation.

All of the above antenna terminology may be explained using geometry and basic Fourier transforms,
which might sound scary at first but engineering grad students may find that it is easier to quickly
review the expression for the surface area of a sphere or the fact that the Fourier transform of a rect-
angle is a sinc function.

So here’s a little applied Fourier Theory. If the Fourier transform of a rectangle is a sinc function, and
the far field antenna pattern of a near field distribution is a Fourier transform, then how big are the
sidelobes of a uniformly illuminated dish antenna? Aha!!! sinc function! To lower the sidelobes you
taper the dish illumination so it doesn’t just suddenly go to zero at the edges...like windowing a time
record in digital signal processing. To reduce the sidelobes to absolute zero, you could illuminate the
dish with Gaussian distribution, but that would require an infinitely large dish.

What if the dish has a hole in it? The hole is a rectangle that you have to subtract, so subtract a sinc
function representing the hole from the far field pattern without the hole, and voila! you have some
sidelobes and a little less gain.

Applying Fourier Theory to dish antenna illumination appears to the unenlightened observer as if you
are just intuitively pulling numbers out of thin air, right up until the point where your quantitative pre-
dictions agree with measurements in the anechoic chamber. Intuition based on deep understanding
and experience provides the gifted designer with a profoundly useful tool.

There are full 3D electromagnetic simulators that offer remarkably impressive maps of E and H fields
that are always interesting, often useful, and perhaps quite accurate. Accuracy must be assumed,
since 3D EMag simulators are invariably used to map fields that haven’t been accurately measured.
For the first decade or so, the simulator was the greater experiment, and the talented folks working on
numerical electromagnetic code were delighted when a physical experiment agreed with the output of
their algorithms. But alas, that generation has moved on to other pursuits, and been replaced with a
younger generation that has grown up as true believers. Electromagnetic Simulators are profoundly
useful tools, capable of providing answers and insights where analysis, experience, and intuition fail.
But I expect that any working electromagnetic engineer with access to both HFSS and an anechoic
chamber has a notebook of unanswered questions.

Simulation, measurement and analysis are three legs of a stool. If one of the legs is missing, the
stool will fall over. Trust, but verify in the anechoic chamber..
There is one area where simulators have evolved to the point where the ability to design an antenna,
build it as accurately as possible, and compare it with the simulations yields remarkable agreement:
Yagi-Uda antenna design. The yagi antenna is an array of parallel wire elements with the transmis-
sion line only connected to one element. The rest of the elements in the near field of the “driven
element” all receive induced currents, and the resulting far field is the sum of the contributions from
the antenna currents in the driven element and the “parasitic elements.” Yagi and Uda were an Emag
prof and student in Japan a century ago. Legend has it that Uda did the work but Yagi wrote the
papers in English. So there are always folks who think that yagi antennas should be uda antennas,
just as it should be the Vail code instead of being named after his boss, Samuel Morse, and North
America should be named after St Brendan or someone who was already here even then. But Yagi’s
name was on the paper, the antennas he described were referenced in some classified radar antenna
documents aboard a US Navy ship that was captured by the Japanese during WWII, and Yagi was
arrested for assisting the enemy... So I’m comfortable referring to parallel arrangements of driven and
parasitic elements as Yagi antennas.

A good way to design Yagi antennas is a program called EZNEC, available on-line at modest cost.
When I design UHF antennas using EZNEC and build them to the limits of my construction abilities,
the pattern measurements and simulations match to tenths of a dB, and even the impedance match
is very close. I strongly encourage the design and construction of VHF and UHF yagi antennas using
EZNEC as a project exercise in the antennas course. Which leads to the third topic, practical design
and construction.

Topic 3 Design and Construction of Antennas

This is also a huge topic, and a delightful one. Here we are talking about real antennas that con-
nect to real radios that we might hike up into the mountains, put on the roof of a car, or set up on the
beach aimed at a distant spit of land far down the Pacific Coast. So the first consideration is entirely
practical. What materials do we use? Kite sticks and flexible wire? Tape measure elements that fold
up for travel? A stressed parabolic dish antenna that may be folded up like an umbrella? 63’ of wire
lifted up in trees with the aid of fishing line and a sling shot?

Then there is the basic problem of how do we connect a transmission line to the antenna so that at
one end we have a connector that looks like a 50 ohm load, and at the other end are some open
circuited wires that radiate signals off into space. What about reflections and standing waves on the
transmission line and are they bad? No, not necessarily. What about differential and common mode
currents on the transmission lines? When to use a balun...

Until a few months ago the most desirable tool for an antenna engineer studying reflections from the
antenna connector and transmission lines was a Vector Network Analyzer that weighed a ton and
cost an arm and a leg. Beginning in January 2020 the most desirable tool for an antenna engineer
studying reflections from the antenna connector and transmission lines is a shirt-pocket NanoVNA
that costs $50. The NanoVNA displays a Smith Chart, and that’s another practical tool for the design,
construction, and evaluation of antennas.

That’s enough for lecture 1. Thursday’s lecture 2 will be a review of vector calculus, with an empha-
sis on building up the intuitions that help us navigate antenna electromagnetic theory. In preparation
for that, please review Reflections on the Campbell web page:

[Link]