RFBasics
RFBasics
Technology Guide
Contents
1 Introduction
Time
vs.
Frequency
domain
Superposition
2
The
Electromagnetic
Spectrum
3 Frequency
Measurement
Instrumentation
The
Oscilloscope
Real-‐Time
Spectrum
Analyzers
The
Swept
Superheterodyne
Receiver
Attenuator
Preamplifier
Preselection/Lo
Pass
Filter
Mixer
Local
Oscillator
(LO)
The
IF
Filter
The
Analog-‐to-‐Digital
Converter
(ADC)
Resolution
Bandwidth
(RBW)
Filter
The
Details
of
a
Sweep
The
Envelope
Detector
Video
Bandwidth
(VBW)
Noise
Sweep
Speed
The
Vector
Network
Analyzer
4
Spectrum
Monitoring
with
Real-‐time
analysis
Real
Time
Capture
and
Analysis
Probability
of
Intercept
Visualizations
Chapter
1:
Introduction
Radio
frequency
(RF)
waves
are
fascinating
natural
phenomena
that
have
allowed
humanity
to
create
tools
to
communicate
over
vast
distances,
observe
other
worlds,
and
gain
a
deeper
understanding
of
our
own
planet.
They
are
also
ubiquitous
here
on
Earth.
We
are
surrounded
by
RF
waves
created
by
human
technology,
like
the
signals
produced
by
radio
and
television
stations
and
mobile
phones,
as
well
as
signals
emitted
by
natural
sources
like
pulsars
and
supernovae
scattered
throughout
the
universe.
We
created
this
document
to
explain
the
basics
of
RF
signals,
help
you
understand
the
time
and
frequency
domains,
and
introduce
common
RF
measurement
instrumentation
and
measurement
techniques.
We
hope
you
find
this
information
helpful.
Time
vs.
Frequency
Domain
Events
are
often
measured
with
respect
to
time.
The
average
speed
of
a
car,
for
example,
can
be
calculated
by
dividing
the
distance
traveled
by
the
time
it
takes
to
travel
that
distance.
Time
domain
measurements—those
events
measured
with
respect
to
time—are
very
useful
to
our
understanding
of
the
physical
world
and
can
be
critical
to
building
something
that
operates
as
intended.
In
electronics,
time
domain
measurements
are
extremely
common.
The
point
in
time
at
which
a
certain
event
occurs
can
be
key
to
the
success
or
failure
of
a
design.
Unfortunately,
humans
don’t
have
the
ability
to
observe
some
elements
of
our
world.
Obviously,
electrons
are
extremely
useful,
but
they’re
notoriously
small
and
hard
to
catch.
However,
we
have
been
able
to
build
tools
that
can
help
us
observe
electrons
as
they
do
their
work.
The
oscilloscope
is
one
of
these
tools.
In
fact,
oscilloscopes
are
among
the
most
common
tools
used
to
perform
time
domain
measurements.
In
essence,
an
oscilloscope
plots
a
graph
of
the
voltage
at
its
input
with
respect
to
time.
Figure
1-‐1:
Oscilloscope
display
showing
two
waveforms.
The
display’s
horizontal
axis
shows
time;
the
vertical
axis
shows
amplitude.
The
upper
waveform
is
sinusoidal
and
the
lower
waveform
is
a
square
wave.
Note
that
they
contain
elements
that
repeat
with
respect
to
time.
An
oscilloscope
can
show
when
events
occur,
measure
the
amplitude
of
the
event,
and
measure
the
time
between
events.
When
discussing
time-‐varying
events,
we
often
use
terms
from
basic
wave
theory.
Let’s
take
a
look
at
a
common
wave
function—the
sine
wave—and
describe
these
basic
elements
in
more
detail.
The
sinusoidal
(sine)
wave
is
a
time-‐varying
waveform
with
smooth
transitions
that
occurs
quite
frequently
in
electronics.
Where
y(t)
=
A
is
the
amplitude,
f
is
the
frequency
of
oscillations
(cycles)
that
occur
per
unit
of
time,
and
j
is
the
phase,
specifies
(in
radians)
where
in
its
cycle
the
oscillation
is
at
t
=
0.
The
period
of
a
time-‐varying
signal
is
the
smallest
amount
of
time
that
defines
a
fundamental
repeating
element
of
the
waveform.
Figure
1-‐2
shows
a
sinusoidal
waveform
illustrating
the
amplitude
and
one
period
of
the
waveform.
Figure
1-‐2:
Sinusoidal
waveform
with
common
waveform
terminology.
The
frequency
is
the
number
of
such
periods
that
occur
during
a
specific
amount
of
time.
f = 1/T
Where
f
is
the
frequency
in
Hertz
(Hz)
and
T
is
the
waveform
period
in
seconds.
Hertz
is
a
secondary
unit
that
represents
the
inverse
of
the
waveform
period
(1/s).
Let’s
look
at
the
voltage
that
comes
from
a
wall
outlet.
In
the
United
States,
if
we
measured
the
voltage
from
a
wall
outlet
with
an
oscilloscope,
we
would
see
that
it
has
an
amplitude
of
approximately
110V
and
a
period
of
16.67ms.
That
means
that
every
16.67ms,
the
voltage
values
repeat.
Now, what is the frequency of the voltage from a wall outlet in the United States?
As you can see, a waveform can be described by its characteristics in both the time and frequency domains.
Superposition
Learning
the
basics
of
periodic
waveforms
like
the
sine
wave
offers
extremely
powerful
tools
for
explaining
and
understanding
more
complicated
waveforms.
Figure
1-‐2
showed
a
single
sine
waveform.
Figure
1-‐3
illustrates
what
happens
when
we
source
a
5V
sine
wave
into
an
oscilloscope:
Figure
1-‐3:
An
oscilloscope
displaying
a
sine
wave
with
a
frequency
of
10MHz.
You
can
see
that
the
frequency
is
10MHz.
Now, let's source a 20MHz sine wave at the same time and compare the two (Figure 1-‐4).
Figure
1-‐4:
An
oscilloscope
display
of
one
sine
wave
with
a
frequency
of
10MHz
(yellow)
and
another
with
a
frequency
of
20MHz
(light
blue).
So,
we
have
a
10MHz
sine
wave
and
a
20MHz
sine
wave.
What
happens
when
we
add
these
sine
waves
together?
The
waveform
changes.
Figure
1-‐5:
An
oscilloscope
displaying
a
sine
wave
with
a
frequency
of
10MHz
(yellow)
and
a
wave
that
combines
a
10MHz
and
a
20MHz
sine
wave
(blue).
This
is
known
as
the
superposition
principle.
You
can
add
sine
waves
together
and
the
resultant
wave
can
have
a
drastically
different
shape
than
the
original
waveforms.
To
put
it
another
way,
any
waveform
can
be
constructed
by
the
addition
of
simple
sine
waves.
Now,
let’s
discuss
some
basic
terms.
The
fundamental
frequency
of
the
new
waveform
is
the
lowest
repeated
frequency.
In
this
case,
the
fundamental
frequency
of
the
waveform
is
10MHz.
The
second
harmonic
is
a
waveform
with
a
frequency
that
is
twice
the
fundamental.
In
this
case,
the
second
harmonic
is
20MHz
(2*10MHz).
You
can
continue
on
in
this
way
to
create
any
waveform.
Let's
take
a
look
at
a
special
case.
If
you
continue
to
add
odd
harmonics
(1,
3,
5,
7,
9,
etc.),
you
will
build
a
square
wave.
The
lower
waveform
in
Figure
1-‐6
was
built
using
odd
harmonics.
Figure
1-‐6:
An
oscilloscope
display
showing
a
sine
wave
with
a
frequency
of
10MHz
(yellow)
and
a
square
waveform
with
a
frequency
of
10MHz
(light
blue).
Note
that
the
waveform
is
starting
to
look
more
“square,”
but
the
frequency
of
the
main
shape
is
still
at
10MHz.
What would these waveforms look like in the frequency domain?
A spectrum analyzer is an instrument that displays the amplitude vs. frequency for input signals.
If we source a 10MHz sine wave into a spectrum analyzer, the display looks like Figure 1-‐7.
Figure
1-‐7:
A
10MHz
sine
wave
displayed
on
a
spectrum
analyzer.
Note
the
peak
at
10MHz.
Now,
let's
look
at
the
square
waveform
on
a
spectrum
analyzer
(Figure
1-‐8).
You
can
see
the
fundamental
frequency
at
10MHz;
the
third
(3*10MHz
=
30MHz),
fifth
(5*10MHz
=
50MHz),
and
seventh
(7*10MHz
=
70MHz)
harmonics
are
also
shown.
By
visualizing
the
signal
in
frequency
domain,
it’s
easy
to
see
what
frequencies
we
are
sourcing,
as
well
as
the
power
distribution
for
each
frequency.
Spectral
analysis
is
critical
in
designing
and
troubleshooting
communications
circuits,
radio/broadcast,
transmitters/receivers,
as
well
as
Electromagnetic
Compliance
(EMC)
measurements.
The
following
chapters
will
offer
an
overview
of
spectrum
analyzer
design
and
techniques
for
using
these
instruments
properly.
Some
of
the
most
significant
contributions
to
our
understanding
of
waves
come
from
a
French
mathematician,
Jean-‐Baptiste
Joseph
Fourier
(1768–1830).
Fourier
was
investigating
a
solution
to
modeling
the
transfer
of
heat
across
a
metal
plate.
As
part
of
his
work,
he
created
a
method
of
adding
simple
sine
waves
to
create
a
more
complicated
waveform.
His
“Fourier
Transform”
has
been
used
to
solve
many
complex
physical
problems
in
thermodynamics
and
electronics.
It
also
provides
a
way
to
convert
signals
captured
in
the
time
domain
into
the
frequency
domain.
This
concept
has
had
far-‐reaching
effects
in
electronics,
communications,
and
the
physical
sciences.
The
superposition
principle
discussed
earlier
in
this
chapter
is
based
on
Fourier's
initial
research.
Electromagnetic
radiation
is
a
form
of
energy
that
is
carried
by
synchronized
oscillating
electric
and
magnetic
fields.
It
is
unique
in
that
its
actions
can
be
explained
by
theories
that
are
based
on
both
waves
and
particles.
Electromagnetic
radiation
also
travels
without
a
medium.
Waves
on
the
ocean
require
water
in
order
to
exist;
sound
waves
require
air
to
propagate.
Although
neither
of
these
waves
can
travel
through
a
vacuum,
electromagnetic
waves
can.
In
fact,
they
travel
through
the
vacuum
of
space
at
the
speed
of
light.
Recall
that
a
wave
can
be
described
by
its
frequency
of
oscillation.
Electromagnetic
waves
are
no
different
and
they
cover
quite
a
broad
range
of
frequencies.
In
fact,
nature
has
no
known
physical
limits
on
maximum
and
minimum
frequencies.
Frequencies
are
grouped
into
bands
based
on
similarities
in
their
physical
traits
or
specific
applications.
Some
frequency
bands
travel
through
the
Earth’s
atmosphere
with
less
loss;
others
are
more
useful
for
a
particular
application
and
are
“set
aside”
for
experimentation.
Some
bands
have
more
than
one
official
user.
Two common frequency bands to note are light and radio.
Visible
light
is
defined
as
electromagnetic
radiation
with
wavelengths
from
400nm
to
700nm
(1nm
is
1¥10-‐
9
m).
This
is
equivalent
to
frequencies
from
5¥1014
Hz
to
1¥1015
Hz,
although
wavelengths
are
traditionally
used
when
discussing
light.
Humans
can
see
electromagnetic
radiation
with
wavelengths
(or
frequencies)
in
this
band.
The
radio
frequency
(RF)
band
of
electromagnetic
waves
has
frequencies
from
8.3kHz
(104
Hz)
to
300GHz
(1011
Hz).
Figure 2-‐1 illustrates the full electromagnetic spectrum and the RF band.
Figure
2-‐1:
The
full
electromagnetic
spectrum
and
the
RF
band.
The
RF
band
is
useful
for
many
industries
and
applications,
including
direct
audio
communications
(mobile
phones,
mobile
radios,
FM
radios),
device
communications
(wireless
keyboards,
Wi-‐Fi
hotspots,
game
controllers),
and
interplanetary
research
(like
the
giant
radio
telescope
at
the
Arecibo
Observatory
in
Puerto
Rico).
Within
the
RF
band,
specific
frequencies
are
dedicated
to
communication
and
broadcast
that
are
open
to
anyone
with
the
ability
to
transmit.
The
Citizens
Band
(CB)
as
well
as
Industrial,
Science,
and
Medical
band
(ISM)
are
examples
of
unlicensed
communications
bands.
Others,
like
FM
radio,
are
licensed
channels
that
are
specifically
allocated
or
rented
by
individuals
or
corporations
for
a
particular
use.
The
national
government
and
the
channel
licensee
monitor
licensed
broadcast
channels
very
closely
in
order
to
ensure
that
the
broadcasts
adhere
to
certain
content
and
physical
transmission
criteria.
In
the
United
States,
the
Federal
Communications
Commission
(FCC)
regulates
the
RF
spectrum.
Electromagnetic
interference
(EMI)
is
another
important
aspect
of
the
RF
story.
Devices
designed
to
transmit
and
receive
RF
signals
are
classified
as
intentional
radiators.
Examples
include
FM
radios,
Wi-‐Fi
routers,
and
wireless
keyboards.
However,
other
devices
that
are
not
specifically
intended
to
create
RF
signals
are
classified
as
unintentional
radiators;
they
represent
the
primary
source
of
EMI.
EMI
is
RF
noise.
An
unintentional
radiator
creates
RF
radiation
that
is
not
intended
to
communicate,
control,
or
deliver
any
relevant
information.
Therefore,
unintentional
radiators
are
RF
noise
sources.
Some
designs
exhibit
less
noise
than
others.
But,
just
imagine
if
every
electronic
device
emitted
a
large
amount
of
RF
noise!
What
if
your
radio-‐controlled
car
interfered
with
the
radar
at
a
nearby
airport?
In
order
to
control
and
maintain
a
safe
operating
environment,
governments
regulate
the
amount
of
acceptable
EMI
that
a
design
or
product
can
produce.
Products
that
exceed
the
limits
set
forth
by
the
regulations
can
lead
to
heavy
financial
penalties
for
offending
individuals
or
companies.
When
performing
experiments
and
development
with
RF,
it
is
very
important
to
understand
the
requirements
of
working
within
a
specific
frequency
band.
If
you
are
working
within
a
licensed
or
restricted
band,
make
sure
to
research
how
to
do
that
safely
and
work
within
the
regulations
for
that
band.
Our
previous
discussions
on
the
time/frequency
domains
and
the
electromagnetic
spectrum
have
provided
a
base
for
our
knowledge
of
RF.
In
the
following
sections,
we
will
introduce
basic
RF
measurement
instrumentation
and
techniques,
with
a
focus
on
typical
RF
component
tests,
broadcast/radio
monitoring,
and
EMI.
The
Oscilloscope
In
many
cases,
looking
at
a
signal
in
the
time
domain
can
provide
indications
about
the
performance
of
a
particular
design.
It
can
tell
you
how
quickly
a
signal
achieves
its
maximum
voltage
(rise
time)
or
its
lowest
(fall
time),
how
two
signals
compare
with
one
another
vs.
time,
or
the
duration
of
a
signal.
All
of
these
measurements
are
ideally
measured
in
the
time
domain.
The
oscilloscope,
introduced
in
chapter
1,
measures
voltage
with
respect
to
time,
then
displays
the
graph
of
voltage
(amplitude)
vs.
time
(Figure
3-‐1).
Figure
3-‐1:
Oscilloscope
display
showing
two
waveforms.
The
horizontal
axis
of
the
display
shows
time;
the
vertical
axis
displays
amplitude.
The
upper
waveform
is
sinusoidal
and
the
lower
waveform
is
a
square
wave.
Note
that
they
contain
elements
that
repeat
over
time.
Figure
3-‐2:
A
modern
digital
oscilloscope.
Early
oscilloscopes
were
strictly
analog
in
nature
and
used
a
cathode
ray
tube
(CRT)
as
a
display.
Much
like
an
analog
television
set,
these
scopes
would
“draw”
the
incoming
signal
on
the
display.
This
was
extremely
helpful
in
visualizing
the
input
signal
but
made
it
difficult
to
perform
any
direct
measurements;
the
only
way
to
save
data
was
to
take
a
picture
of
the
display
of
the
oscilloscope.
The
advancement
of
digital
technology
led
to
fully
digital
oscilloscopes
(Figure
3-‐2).
With
the
raw
voltage
and
time
data
digitized,
the
data
could
be
saved
and
used
to
perform
calculations
directly
within
the
scope
itself.
Modern
oscilloscopes
can
now
directly
calculate
rise
time,
duty
cycle,
maximum
voltage,
and
more.
Figure
3-‐3:
Oscilloscope
display
showing
all
measurements
for
a
1MHz
square
wave
input.
Some
digital
scopes
can
also
display
the
amplitude
of
the
incoming
signal
vs.
frequency
by
using
Fast
Fourier
Transform
(FFT)
calculations.
The
FFT
function
of
oscilloscopes
can
be
useful
in
identifying
the
fundamental
frequency,
as
shown
in
Figure
3-‐4.
Figure
3-‐4:
Oscilloscope
display
showing
an
FFT
of
a
1MHz
square
wave
input.
So,
with
a
scope,
we
can
read
the
phase
information
(in
the
time
domain)
and
gather
basic
amplitude
and
frequency
information
in
the
frequency
domain
by
using
FFTs.
Unfortunately,
oscilloscopes
tend
to
have
a
much
higher
noise
floor
than
traditional
frequency
measurement
instrumentation
like
spectrum
analyzers.
This
can
make
looking
for
small
amplitude
elements,
like
higher
ordered
harmonics,
difficult
if
not
impossible.
They
are
also
“wideband”
instruments,
which
means
that
they
detect
a
wide
range
of
frequencies
at
the
same
time.
This
raises
the
noise
floor
and
does
not
provide
for
an
easy
way
to
differentiate
between
signals
that
could
have
frequencies
that
are
close
together.
Real-‐Time
Spectrum
Analyzers
Real-‐time
spectrum
analyzers
are
similar
to
oscilloscopes
in
that
they
first
collect
data
in
the
time
domain,
then
calculate
the
frequency
using
FFT
algorithms.
In
this
way,
they
can
collect
a
large
number
of
data
points
over
a
broad
range
of
frequencies,
calculate
the
amplitude
vs.
frequency,
and
display
them
quickly
in
the
frequency
domain.
They
differ
from
oscilloscopes
in
that
they
tend
to
offer
lower
noise
floors,
as
well
as
special
filtering
that
can
differentiate
between
signals
that
are
close
together.
Real-‐time
spectrum
analyzers
are
very
useful
in
capturing
fast-‐changing
signals,
especially
when
working
with
digital
communications.
They
generally
can
capture
transients
and
fast
signals
more
quickly
than
a
swept
spectrum
analyzer,
but
they
also
have
a
higher
noise
floor
and
price
tag.
Although
real-‐time
systems
are
gaining
in
popularity,
they
are
still
significantly
outnumbered
by
the
swept
analyzer
design.
The
remainder
of
this
section
will
explore
the
inner
workings
of
the
most
popular
method
of
frequency
measurement,
the
swept
spectrum
analyzer.
The
Swept
Superheterodyne
Receiver
Spectrum
analyzers
based
on
swept
superheterodyne
designs
are
very
popular,
due
in
part
to
their
low
noise,
ease
of
use,
and
ability
to
differentiate
between
signals
that
have
very
close
frequencies.
In
basic
terms,
the
swept
superheterodyne
is
almost
identical
to
a
radio
receiver.
Both
can
be
set
to
a
particular
frequency
range
and
filter
out
other
frequencies
(like
tuning
to
particular
radio
station)
and
then
observe
the
incoming
signal.
Unlike
a
radio,
which
is
tuned
to
a
particular
frequency
and
then
the
signal
is
fed
to
a
speaker,
an
analyzer
is
not
set
to
a
fixed
frequency.
Instead,
the
analyzer
sweeps
across
frequencies
in
steps,
like
moving
the
radio
to
a
new
channel,
and
then
plots
the
signal
amplitude
on
a
display.
In
simple
terms,
this
design
takes
an
unknown
signal
(an
input
or
RFIN
signal)
and
mixes
(combines)
it
with
a
sweeping
signal,
or
swept
Local
Oscillator
(LO),
to
create
a
signal
that
is
a
combination
of
both.
The
LO
is
swept
from
a
start
frequency
to
a
stop
frequency
in
discrete
steps.
Each
step
in
the
sweep
defines
a
frequency
“bin”
on
the
spectrum
analyzer
display.
At
each
bin,
the
power
is
measured.
If
the
unknown
signal
has
a
frequency
component
within
the
bin,
the
display
will
place
a
data
point
at
the
equivalent
amplitude
of
the
unknown
signal.
After
the
sweep
is
completed,
the
resulting
display
will
represent
one
scan
across
the
span
defined
by
the
start
and
stop
values
of
the
instrument.
The
following
sections
will
examine
how
each
circuit
element
is
used
to
create
this
output.
First,
let’s
look
back
at
a
little
electrical
engineering
history.
The
term
superheterodyne
is
short
for
supersonic
heterodyne.
The
basic
design
was
created
by
Edwin
Armstrong,
an
American
electrical
engineer,
in
1918.
Supersonic
refers
to
waves
with
frequencies
higher
than
those
within
the
range
of
human
hearing
(31Hz
to
21kHz).
Heterodyne
is
a
contraction
of
the
Greek
words
hetero-‐,
which
means
“different”
and
–dyne,
which
means
“power.”
Figure 3-‐5 illustrates a basic design for a modern superheterodyne receiver used in a spectrum analyzer.
Figure
3-‐5:
A
modern
superheterodyne
receiver
used
in
a
spectrum
analyzer.
Let's
take
a
look
at
each
of
the
elements
as
a
signal
passes
through
each
of
them.
Attenuator
The
input
RF
signal
is
connected
to
the
spectrum
analyzer’s
RF
input
where
it
enters
the
attenuator
circuit.
The
attenuator
is
used
to
decrease
the
amount
of
power
delivered
to
the
circuit
elements
that
follow
it.
This
protects
the
sensitive
electronics
that
follow
and
decreases
the
effects
of
spurs
and
modulation
effects
in
the
mixer.
In
many
cases,
a
design
can
incorporate
an
integrated
attenuator
that
can
be
controlled
by
settings
on
the
analyzer.
External
attenuators
that
have
fixed
or
variable
attenuation
can
also
be
used.
Preamplifier
The
preamplifier
(PA)
is
a
low
noise
amplifier
that
increases
the
input
signal
amplitude
(Figure
3-‐6).
It
can
increase
the
signal-‐to-‐noise
ratio
and
helps
increase
the
measurement
sensitivity
to
low
power
elements
in
the
input
signal.
It
is
also
usually
controlled
by
the
spectrum
analyzer.
Figure
3-‐6:
The
effect
of
a
n
attenuator
and
a
preamplifier
on
an
input
signal.
Preselection/Lo
Pass
Filter
The
preselection
filter
is
a
bandpass
filter
that
only
allows
certain
frequencies
to
reach
circuit
elements.
Unwanted
signals
are
rejected
to
prevent
them
from
causing
measurement
errors.
Preselection
filters
may
or
may
not
be
included
in
a
particular
design.
Although
they
add
complexity
and
cost,
they
do
reduce
the
likelihood
of
false
peaks
in
the
scanned
spectra.
A
low-‐pass
filter
typically
prevents
frequencies
that
exceed
the
maximum
operating
frequency
from
entering
the
circuit.
This
stops
them
from
entering
the
next
stage
of
the
circuit,
where
they
could
be
more
difficult
to
remove.
A
DC
block
is
also
included
in
the
RF
input
circuit.
This
element
blocks
out
any
DC
components
of
the
input
that
can
cause
overloading
or
damage
to
the
remaining
circuit
elements.
Mixer
A
mixer
is
a
three-‐port
circuit
element
that
takes
two
input
signals
and
creates
an
output
signal
that
is
a
combination
of
both
the
RFIN
and
LO.
In
this
design,
the
mixer
multiplies
the
unknown
input
signal
(frequency
=
fsig)
with
the
known
local
oscillator
(frequency
=
fLO).
The
resultant
output
(Figure
3-‐7)
is
composed
of
the
original
RF
signal
(fsig),
the
local
oscillator
signal
(fLO),
and
both
the
sum
and
difference
of
the
RF
and
LO
inputs
(fLO
-‐
fsig
and
fLO
+
fsig,
respectively),
and
the
sum
and
difference
of
higher
harmonics,
such
as
(2fLO
-‐
fsig/2fsig
-‐
fLO
and
2fLO
+
fsig/2fsig
+
fLO).
Figure
3-‐7:
The
input
signal
(fsig)
is
mixed
with
the
local
oscillator
signal
(fLO).
The
result
is
the
IFout,
which
includes
the
fundamental
mixed
products.
The output frequency from the mixer is known as the intermediate frequency (IF).
The
IFout
components
that
include
only
the
difference
between
the
LO
and
the
input
frequencies
(fLO
-‐
fsig)
are
known
as
the
downconverted
signal.
The
IFout
components
that
include
only
the
sum
between
the
LO
and
the
input
frequencies
(fLO
+
fsig)
are
known
as
the
upconverted
signal.
Downconverted
signals
are
calculated
by
subtracting
the
fsig
components
from
the
intermediate
frequency
as
shown
here:
By
adding
multiple
IF
stages,
it’s
possible
to
maximize
the
frequency
resolution
and
extend
the
operating
range.
Multistage
IF
sections
can
upconvert
(step
up)
or
downconvert
(step
down)
the
fLO
to
a
different
range,
where
they
can
be
filtered
to
remove
unwanted
frequencies
at
each
step
as
shown
in
Figure
3-‐8.
Figure
3-‐8:
A
diagram
of
a
triple
conversion
IF
stage.
The
use
of
multiple
mixing
stages
allows
the
instrument
to
have
superior
sensitivity,
good
frequency
stability,
and
high
frequency
selectivity.
This
allows
the
instrument
to
have
a
wide
operating
frequency
range
and
the
ability
to
differentiate
signals
that
have
frequencies
that
are
close
together.
Local
Oscillator
(LO)
Local
oscillators
(LOs)
are
an
integral
part
of
the
mixer
network.
An
LO
is
a
circuit
element
that
provides
a
signal
at
a
known
frequency
and
amplitude.
An
LO
can
be
made
up
of
different
designs
and
materials,
but
its
main
purpose
is
to
provide
a
stable
frequency
reference
that
can
be
used
to
compare
unknown
RF
from
the
input
stages.
Many
designs
(Figure
3-‐9)
incorporate
a
swept
LO
as
the
first
stage
(1st
LO).
A
swept
LO
uses
a
Voltage
Controlled
Oscillator
(VCO).
As
the
voltage
to
a
VCO
is
increased,
the
output
frequency
also
increases.
Figure
3-‐9:Example
of
a
swept
Voltage
Controlled
Oscillator
(VCO)
The
sweep
function
is
also
tied
to
the
instrument
display.
As
the
LO
steps
from
the
start
to
the
stop
frequency,
the
frequency
step,
or
“bin”
of
the
display
is
also
stepped.
This
synchronization
ensures
that
the
IF
and
the
displayed
values
for
each
frequency
“bin”
are
matched.
The
local
oscillator
frequency
is
based
on
a
reference
oscillator
within
the
circuit.
This
reference
oscillator
is
generally
a
crystal
oscillator,
like
quartz,
that
vibrates
at
a
known
frequency.
A
perfect
reference
oscillator
would
have
an
infinitely
accurate
output
frequency
that
would
not
change
with
aging
or
temperature.
It
would
have
no
“spectral
width”
and
would
be
a
straight
line
at
the
oscillator
frequency,
as
shown
in
Figure
3-‐
10.
Figure
3-‐10:
Example
of
a
perfect
oscillator
output.
Note
there
is
only
one
single
frequency
with
no
width.
Unfortunately,
the
vibrations
(and
therefore
the
output
frequency)
for
oscillators
are
effected
by
environmental
conditions
like
aging,
temperature,
and
humidity.
This
leads
to
phase
noise.
Phase
noise
represents
the
change
in
the
phase
of
an
oscillators
output
signal
over
time.
On
a
spectrum
analyzer,
phase
noise
shows
up
as
a
widening
of
the
occupied
frequencies
of
an
input.
This
widening
can
be
described
as
a
wedge
or
“skirt”
near
the
bottom
of
measurement
as
highlighted
in
the
box
in
Figure
3-‐11.
Figure
3-‐11:
A
real
oscillator
measurement
showing
the
phase
noise
that
is
due
to
thermal
fluctuations
in
the
circuit
elements.
Phase
noise
effectively
increases
the
noise
floor
and
can
increase
the
difficulty
of
observing
small
signals
close
to
an
input
frequency
by
effectively
covering
the
signals
that
you
are
seeking.
Low
phase
noise
can
help
increase
the
low-‐level
signal
observation
near
measured
input
frequencies.
The
yellow
trace
in
Figure
3-‐11
is
a
100MHz
RF
input.
The
purple
line
is
the
noise
floor
of
the
analyzer
with
identical
measurement
settings.
Note
the
rise
in
the
noise
floor
approaching
the
center
frequency
of
the
100MHz
input.
This
phase
noise
is
coming
from
either
the
source
or
the
analyzer,
whichever
is
greater.
Phase
noise
can
be
minimized
by
selecting
quality
reference
oscillator
materials
as
well
as
environmental
control.
Temperature
Compensated
Crystal
Oscillators
(TCXO),
Microcontrolled
Crystal
Oscillators
(MCXO),
and
Oven
Controlled
Crystal
Oscillators
(OCXO)
are
some
common
designs
that
can
be
incorporated
to
help
limit
the
phase
noise
of
a
particular
design.
The
IF
Filter
The
IF
filter
follows
immediately
after
the
mixer.
It
filters
out
the
fsig,
fLO,
and
leaves
either
the
upconverted
(fLO
+
fsig)
or
downconverted
signal
(fLO
-‐
fsig),
as
shown
in
Figure
3-‐12.
Figure
3-‐12:
Output
of
the
IF
stage
after
the
IF
filter.
Now,
a
known
signal
(fLO)
has
been
subtracted
from
the
original
signal
of
an
unknown
frequency
to
create
the
IF.
Because
we
know
the
frequency
of
the
fLO,
we
could
apply
a
filter
with
a
known
center
frequency
and
known
bandwidth
to
the
IF
and
measure
the
output.
If
the
IF
does
not
exist
in
that
frequency
range,
we
could
step
to
another
filter
(with
a
different
center
frequency)
and
check
again.
With
enough
filters,
we
could
continue
to
step
through
center
frequencies
and
look
for
the
IF.
Once
we
find
the
center
frequency
of
the
IF,
we
can
subtract
the
original
fLO
and
the
result
would
be
our
previously
unknown
fsig.
In
practice,
designing
analog
filter
networks
with
enough
range
and
performance
to
cover
giga-‐hertz
of
frequency
ranges
can
be
difficult
and
expensive.
In
the
past,
fully
analog
designs
were
the
only
available
option.
They
worked
very
well
over
their
intended
ranges,
but
there
are
disadvantages
to
a
fully
analog
system.
Most
modern
designs
incorporate
a
different
strategy
to
isolate
the
unknown
signal.
They
use
a
final
IF
stage
with
a
fixed
center
frequency
and
the
LO
is
swept.
The
next
section
covers
how
this
swept
design
eases
the
design
burdens
and
increases
functionality
of
spectrum
analyzer
design.
The
Analog-‐to-‐Digital
Converter
(ADC)
The
first
few
generations
of
spectrum
analyzers
used
analog
components
throughout
their
design,
with
the
readout
being
an
analog
cathode
ray
tube
(CRT)
display.
Wider
operating
ranges
and
smaller
bandpass
filters
required
more
components.
Due
to
the
nature
of
the
available
analog
components,
a
fully
analog
spectrum
analyzer
was
substantial
in
size,
weight,
and
cost.
Some
older
spectrum
analyzers
could
be
as
large
as
24”
¥
24”
¥
24”
in
size,
weigh
over
40
pounds,
and
cost
more
than
$40,000.
Today,
many
of
the
filters
and
other
components
have
been
replaced
by
digital
signal
processors
(DSPs),
which
can
successfully
model
the
characteristics
of
their
analog
counterparts.
The
integration
of
digital
components
has
simplified
the
design
and
decreased
the
size,
weight,
and
cost.
A
modern
spectrum
analyzer
a
bit
bigger
than
a
shoebox
and
weighing
less
than
10
pounds
can
be
purchased
for
less
than
$10,000.
There
are
even
USB-‐controlled
instruments
that
are
as
small
as
a
pack
of
playing
cards
for
a
few
hundred
dollars.
Now,
back
to
our
original
discussion.
The
unknown
signal
has
been
converted
to
create
the
IF
(intermediate
frequency).
The
IF
signal
is
then
sent
to
an
analog-‐to-‐digital
converter
(ADC).
The
ADC
creates
a
digital
output
that
is
proportional
to
the
analog
input.
Once
the
signal
has
been
converted
to
a
digital
signal,
it
can
be
easily
manipulated.
We
can
apply
different
mathematical
algorithms
to
help
isolate
the
unknown
signal.
The
RBW
filter
is
a
bandpass
filter
that
allows
frequencies
within
its
envelope
to
pass
through
the
filter
but
rejects
frequencies
outside
of
the
envelope.
The
RBW
filter
commonly
implements
a
Gaussian
shape,
as
shown
in
Figure
3-‐13.
Figure
3-‐13:
Typical
Gaussian
RBW
filter
shape.
A
bandpass
filter
has
a
center
frequency
and
a
bandwidth.
Typically,
the
bandwidth
is
given
at
the
filter’s
3dB
point.
The
3dB
point
is
the
amplitude
at
which
the
area
of
the
curve
is
split
in
half.
Fifty
percent
lies
above
the
3dB
point,
and
50
percent
lies
below
the
3dB
point.
The bandwidth is defined as the frequency difference (f2 -‐ f1) at the 3dB point.
The
Gaussian
filter
shape
is
used
because
it
offers
the
greatest
degree
of
phase
linearity
balanced
with
good
selectivity.
Selectivity
is
a
measure
of
how
well
a
filter
rejects
signals
that
are
near
the
operating
frequency
range
of
the
filter.
Other
filter
shapes
(flat
top,
for
example)
have
better
selectivity,
so
they
can
reject
out-‐of-‐
band
signals
more
precisely,
but
they
do
not
have
good
phase
linearity,
which
can
cause
undesirable
effects
like
ringing
that
cannot
be
filtered
out
easily.
Although
very
popular,
there
are
also
other
filter
types
that
can
be
implemented
using
analog,
digital,
or
a
combination
of
both
in
a
hybrid
design.
Digital filters can be designed to have better selectivity than analog filters (Figure 3-‐14).
Figure
3-‐14:
Analog
(black)
vs.
digital
(blue)
filter
shapes.
Note
that
digital
filters
can
achieve
higher
directivity
than
analog
filters.
Digital
filters
also
offer
other
advantages.
For
example,
many
electromagnetic
compliance
(EMC)
tests
typically
use
a
6dB
Gaussian
filter.
This
filter
bandwidth
is
defined
at
the
6dB
point,
where
approximately
75
percent
of
the
area
of
the
filter
is
higher
than
the
6dB
point.
This
type
of
filter
has
a
more
desirable
response
to
impulse
and
short
duration
burst
RF
signals,
which
can
be
a
major
contributor
to
electromagnetic
interference
(EMI).
Figure
3-‐15:
Block
diagram
of
a
swept
superheterodyne
design
showing
the
location
of
the
final
stage
of
filtering
and
detector
(RBW,
Envelope,
and
VBW).
Figure
3-‐16
shows
a
screen
shot
from
a
swept
spectrum
analyzer.
The
input
signals
are
40MHz
and
40.1MHz
sine
waves.
Figure
3-‐16:
Two
input
signals
scanned
with
RBWs
of
1MHz
(yellow),
100kHz
(purple),
and
10kHz
(blue).
The
yellow
trace
was
captured
using
an
RBW
of
1MHz.
The
purple
trace
was
at
an
RBW
of
100kHz,
and
the
light
blue
trace
was
at
10kHz.
Because
the
signals
are
only
separated
by
100kHz,
the
RBW
of
1MHz
shows
just
a
single
bump
because
the
RBW
is
greater
than
the
input
signal.
This
shape
is
characteristic
of
the
swept
IF
frequency
tracing
out
the
envelope
filter
that
follows
the
RBW
filter.
As
the
RBW
decreases,
we
get
higher
frequency
resolution
and
lower
noise
floor,
as
you
can
see
in
the
purple
(RBW
=
100kHz)
and
blue
(RBW
=
10kHz)
traces.
The
Details
of
a
Sweep
As
mentioned
earlier,
we
could
implement
a
design
that
uses
a
filter
for
each
frequency
step
of
interest.
This
approach
might
be
acceptable
for
analyzing
a
few
hundred
kilohertz
of
range,
but
it
would
be
difficult
and
expensive
if
it
had
to
work
over
a
larger
frequency
range.
For
example,
if
we
wanted
to
have
the
100Hz
of
frequency
resolution
over
100MHz
of
operating
range,
that
would
require
a
million
filters,
which
would
not
be
very
practical
to
implement
in
hardware.
As
an
alternative,
the
swept
superheterodyne
resolution
bandwidth
filter
(RBW)
has
a
fixed
center
frequency
that
does
not
change
and
also
incorporates
a
user-‐defined
range
of
bandwidths
that
provide
resolution
selectivity.
This
allows
the
designer
to
use
a
filter
with
better
performance
and
a
wider
operating
range
to
maximize
the
usefulness
of
the
analyzer.
Instead,
the
LO
is
swept
across
a
frequency
range
in
steps
that
are
synchronized
with
the
display
steps.
The
actual
sweep
range
frequencies
for
the
1st
LO
(swept
LO
stage)
are
typically
not
the
same
as
the
sweep
range
on
the
display.
The
LO
sweep
range
is
selected
to
be
out
of
the
measurement
range
of
the
instrument
to
help
minimize
noise
and
harmonics
from
the
RF
input.
Let’s
consider
an
example
where
an
analyzer
is
configured
to
sweep
from
a
start
frequency
of
1GHz
to
1.5GHz.
The
RBW
is
fixed
at
500MHz,
and
the
LO
sweeps
from
500MHz
to
1GHz
as
shown
in
Figure
3-‐17.
At the first step in the sweep, the RFIN is 1GHz, LO is 500MHz, and the resultant IF is 500MHz.
Figure
3-‐17:
Block
diagram
showing
the
unknown
inputs,
LO,
and
the
IF
output.
Note
the
RBW
filter
at
a
fixed
frequency
(green).
Note:
The
LO
sweep
spans
the
same
frequency
range
as
the
display
range
on
the
analyzer.
The
LO
continues
to
sweep
with
the
display.
When
the
LO
is
at
750MHz,
the
IF
has
signals
present
near
the
RBW/IF
filter
center
wavelength,
as
shown
in
Figure
3-‐18.
Figure
3-‐18:The
unknown
RF
input
signals
are
mixed
with
the
LO
to
create
the
IF.
The
IF
is
then
compared
with
the
RBW
filter.
This
process
continues
through
the
entire
scan
range.
Figures 3-‐19 and 3-‐20 show the same steps, as well as the actual displayed spectrum on the analyzer.
Note: In frequency bins where there are no RFIN signals, the display shows the noise floor of the analyzer.
Figure
3-‐19:
The
displayed
spectrum
as
the
analyzer
scans
across
the
operating
frequency
range.
Figure
3-‐20:
A
completed
sweep
showing
the
RBW
filter’s
effect
on
the
displayed
data
(green).
Narrower
RBW
bandwidths
provide
higher
frequency
resolution
and
a
lower
noise
floor,
but
they
increase
the
sweep
time
for
a
given
span
because
they
increase
the
number
of
steps
required
to
cover
that
span.
If
the
IF
is
a
single
frequency
sine
wave,
the
output
will
be
a
single
DC
voltage.
This
is
the
situation
when
the
IF
is
simply
the
LO,
with
no
mixed
frequencies
present.
If
the
IF
is
comprised
of
multiple
frequencies,
which
is
the
case
if
there
are
mixing
products
in
the
IF
(fLO
-‐
fsig
or
fLO
+
fsig),
then
the
output
voltage
will
match
the
envelope
of
the
IF
in
time,
as
shown
in
Figure
3-‐21.
Figure
3-‐21:The
intermediate
frequency
after
the
envelope
detector.
Now,
let's
look
at
how
the
sweeping
LO,
RBW
filter,
and
envelope
work
together
to
convert
the
signal
from
the
time
domain
to
the
frequency
domain.
Figure
3-‐22
shows
an
example
of
an
unknown
signal
(fs)
as
it
is
mixed
with
three
steps
(flo1,
flo2,
and
flo3)
of
a
sweeping
LO.
Figure
3-‐22:
An
unknown
signal
is
mixed
with
three
LO
steps
showing
how
the
data
from
the
envelope
detector
data
is
selected
for
display.
The
envelope
detector
returns
the
voltage
value
of
each
of
steps
(flo1,
flo2,
and
flo3)
in
this
sweep.
At
each
step
in
the
LO,
the
amplitude
(voltage)
is
plotted
vs.
the
step
frequency.
Many
analyzers
have
a
number
of
detectors
available
that
allow
selecting
the
voltage
values
that
are
displayed
from
each
frequency
bin
that
has
been
collected.
Figure
3-‐23
shows
the
video
voltage
of
a
signal
for
two
frequency
bins,
f0
and
f1.
The
top
plot
shows
the
video
voltage
(green
trace)
of
this
incoming
signal
with
respect
to
time.
The
bottom
plot
shows
the
value
that
each
detector
selection
would
report
to
the
display.
Figure
3-‐23:The
video
voltage
is
selected
based
on
the
detector
type.
This
diagram
shows
the
data
displayed
based
on
a
Positive
Peak,
Sample,
and
Negative
Peak
detector
selection
for
a
given
input
signal.
The
positive
peak
detector
displays
the
maximum
amplitude
value
in
each
bin.
It
is
useful
for
sinusoidal
inputs,
but
it
is
not
recommended
for
noise
because
it
will
not
show
the
true
random
nature
of
a
noise
source.
The negative peak detector displays the minimum amplitude value in each bin.
The sample detector will select a random value from the video voltage. It is most useful for observing noise.
The
normal
or
“rose-‐and-‐fell”
detector
will
select
either
the
positive
or
negative
peak
values,
depending
on
the
trend
within
that
bin.
If
the
signal
both
rose
and
fell
within
that
bin,
it
will
assume
that
the
input
is
noise
and
will
alternate
between
positive
and
negative
peaks
to
provide
a
more
appropriate
response
to
the
input.
If
it
rose
within
the
bin,
it
will
select
positive;
if
it
fell,
negative.
It
is
most
useful
for
noise
and
sinusoidal
signals.
There
can
also
be
averaging
detectors,
such
as
RMS
average,
voltage
average,
and
Quasi-‐Peak
average
detectors.
The
RMS
detector
reports
the
root-‐mean-‐square
(RMS)
value
of
the
incoming
signals.
It
is
useful
when
measuring
the
power
of
communication
and
other
complex
signals.
The
voltage
average
reports
the
average,
or
mean,
voltage
of
the
bin.
It
can
be
helpful
for
electromagnetic
interference
(EMI)
measurements.
Some
specialized
analyzers
also
come
with
a
Quasi-‐Peak
(QP)
detector,
which
provides
a
special
weighted
average
of
the
data
in
each
bin
and
is
specifically
used
for
electromagnetic
compliance
(EMC)
testing.
It
should
be
noted
that
the
QP
detector
has
a
fixed
integration
time
that
is
significantly
longer
than
the
standard
detector
types.
It
will
increase
sweep
time
considerably.
Figure
3-‐24:
Suggested
detectors
for
specific
applications.
Figures
3-‐25
and
3-‐26
compare
a
20MHz
sine
wave
input
captured
at
a
VBW
of
100kHz
and
1kHz
respectively.
Note
how
the
VBW
smoothes
the
trace
and
makes
it
easier
to
see
the
true
signal.
Figure
3-‐25:
20MHz
input
on
a
spectrum
analyzer
with
VBW
set
to
100kHz.
Figure
3-‐26:
Identical
20MHz
signal
with
the
VBW
decreased
to
1kHz.
Note
smoothing
of
the
trace.
Note
that
decreasing
the
VBW
setting
can
increase
the
sweep
time.
Many
analyzers
also
provide
the
ability
to
average
a
number
of
traces
when
using
an
Averaging
Detector
like
RMS
or
Voltage.
Averaging
traces
has
a
similar
effect
to
lowering
the
VBW
setting,
as
shown
in
Figures
3-‐27
and
3-‐28,
but
this
technique
can
require
successive
scans,
which
can
lead
to
a
longer
total
test
time.
Figure
3-‐27:
A
20MHz
input
signal
with
no
averaging
applied.
Figure
3-‐28:
A
20MHz
input
signal
with
100
averages
applied.
Note
that
the
noise
on
the
trace
has
decreased
significantly.
Noise
When
making
measurements,
it’s
important
to
limit
the
error
and
decrease
the
effects
of
unwanted
signals
in
order
to
get
the
most
accurate
representation
possible.
Some
sources
of
noise
are
part
of
the
instrument
itself
(internal);
others
are
external.
This
discussion
will
address
only
internal
noise
sources.
One
of
the
main
sources
of
noise
is
thermal
effects
in
the
circuit
elements
within
the
instrument.
Essentially,
temperature
is
a
measure
of
the
average
kinetic
energy
of
a
system.
On
a
molecular
level,
higher
temperatures
lead
to
higher
average
kinetic
energy
due
to
an
increase
in
molecular
vibrations.
These
vibrations
create
electrical
potentials
(voltage)
that
can
effect
measurements.
Thermal
(also
known
as
Johnson)
noise
is
defined
by
this
equation:
Pn
=
kTB
What
is
the
noise
power
of
a
system
with
T
=
293°K
(equivalent
to
room
temperature
of
20°C
or
68°F)
and
a
measurement
bandwidth
of
1MHz
(1¥10+6
Hz)?
Let’s say we cut the bandwidth by 10: B = 100kHz (1¥10+5)
Now, what if we compared these two powers in dB?
The equation for power in dB is Lp = 10 log10 (P/Po)dB.
As
you
can
see,
the
bandwidth
of
the
measurement
directly
affects
the
thermal
noise
that
can
influence
the
measurement.
In
order
to
increase
sensitivity
(synonymous
with
decreasing
noise),
we
can
l
The
Displayed
Average
Noise
Level
(DANL)
of
a
spectrum
analyzer
is
a
term
that
describes
the
expected
noise
level
of
the
analyzer
and
it
determines
the
lowest
signal
level
that
can
be
measured
by
the
instrument.
The
DANL
represents
the
noise
floor
of
the
instrument.
The
DANL
value
is
heavily
influenced
by
the
frequency
span
of
the
measurement,
RBW,
VBW,
preamplifiers,
and
detector
settings
but
can
also
be
affected
by
factors
such
as
the
number
of
trace
averages
being
used.
It’s
possible
to
lower
the
DANL
quickly
by
decreasing
the
RBW
setting.
Decreasing
the
RBW
by
10
times
will
decrease
the
DANL
by
10dB
as
shown
in
Figure
3-‐29.
Figure
3-‐29:
Noise
floor
differences
on
an
analyzer
with
three
RBW
settings.
RBW=
1MHz
(yellow),
100kHz
(purple)
and
10kHz
(blue).
In
this
experiment,
there
was
no
input
signal
to
the
analyzer,
just
the
noise
floor
of
the
instrument.
The
yellow
trace
represents
a
scan
from
9kHz
to
1.5GHz
with
an
RBW
of
1MHz.
The
purple
trace
was
acquired
using
an
RBW
of
100kHz
and
the
light
blue
trace
used
an
RBW
of
10kHz.
Note
that
each
decade
(10)
decrease
in
RBW
resulted
in
a
10dB
drop
in
the
DANL.
Sweep
Speed
Decreasing
the
RBW
has
a
dramatic
effect
on
the
noise
floor,
but
it
also
increases
the
time
that
it
takes
for
the
instrument
to
scan
over
a
specific
frequency
span.
That’s
because
by
decreasing
the
bandwidth
of
each
step,
we
increase
the
number
of
steps
we
must
perform
to
cover
that
span.
The
sweep
speed
is
determined
by
the
span,
RBW,
and
detector
settings.
Many
analyzers
will
have
default
settings
that
will
automatically
set
the
sweep
time
to
provide
the
best
balance
between
sweep
speed
and
amplitude
accuracy.
Short
sweep
times
could
be
too
fast
for
the
IF
stage
to
respond
to
the
input
and
result
in
additional
measurement
error.
The
first
sweep
(Figure
3-‐30)
is
a
scan
of
a
20MHz
sine
wave
with
an
amplitude
of
-‐10dBm.
The
automatic
settings
were
used
on
the
instrument
and
the
resultant
sweep
time
was
101.10ms.
The
second
sweep
(Figure
3-‐31)
is
a
scan
of
exactly
the
same
input
signal
but
with
a
different
sweep
time.
It
was
forced
to
10ms.
Note
that
the
amplitude
error
increased
significantly
and
the
spectral
profile
is
quite
different.
The
analyzer
actually
indicates
that
the
sweep
time
is
too
fast
for
the
settings
with
an
UNCAL
label
in
the
top
portion
of
the
display.
Figure
3-‐30:
20MHz
sine
input
collected
using
the
automatic
settings
on
the
analyzer.
Figure
3-‐31:
The
same
20MHz
sine
input,
but
this
time,
the
sweep
speed
was
increased
manually.
Note
that
the
instrument
is
notifying
the
user
of
UNCAl
and
there
are
amplitude
and
frequency
errors
due
to
the
increased
sweep
speed.
To
ensure
the
greatest
accuracy,
use
the
instrument’s
recommended
automatic
settings.
The
Vector
Network
Analyzer
The
swept
superheterodyne
design
described
previously
captures
amplitude
and
frequency
information.
These
measurements
are
also
known
as
scalar
measurements
because
they
do
not
capture
phase
information
for
the
incoming
signal.
Instruments
that
capture
amplitude
vs.
frequency
are
therefore
known
as
Scalar
Network
Analyzers.
A
host
of
applications
require
capturing
the
incoming
signal
phase
information
in
addition
to
the
amplitude
and
frequency.
This
is
especially
important
for
proper
demodulation
of
digital
communications
and
for
characterizing
components
used
in
digital
systems.
The
Vector
Network
Analyzer
(VNA)
is
widely
used
for
these
applications.
The
VNA
is
based
on
a
swept
superheterodyne
design.
The
main
difference
is
that
the
VNA
uses
an
extra
stage
in
the
signal
path
that
collects
and
stores
phase
information
for
the
incoming
signal.
VNAs
are
useful
for
measuring
the
performance
of
RF
components,
commonly
called
the
scattering
or
S-‐
parameters,
as
well
as
measuring
the
performance
of
digital
communications
signals.
For
example
while
a
fast
changing
frequency
hopping
signal
like
Bluetooth
can
be
measured
with
SA
it
is
severely
limited.
One
trace
can
be
set
to
maximum
hold.
A
second
trace
can
be
set
to
clear
write.
With
one
sweep
it
is
not
possible
to
capture
all
signal
components.
Multiple
sweeps
are
necessary
to
build
up
an
image
of
the
spectrum
(see
figure
2).
So
while
you
have
built
up
a
view
of
the
spectrum
over
time
not
all
frequency
components
are
visible,
there
is
no
time
information
available
and
it
is
not
possible
to
detect
that
this
signal
is
a
frequency
hopping
spread
spectrum
signal.
Figure
2:
Bluetooth
signals
are
only
visible
via
max
hold
function
with
SA
Certain
signals
which
are
only
randomly
available
and
very
fast
may
never
be
detected.
Additionally,
Frequency,
Span
and
RBW
have
a
direct
influence
to
sweep
time
on
a
swept
spectrum
analyzer.
So
if
a
finer
frequency
resolution
is
required,
then
RBW
needs
to
be
decreased.
This
results
in
a
lower
sweep
time
making
the
capture
of
those
fast
changing
signals
even
less
likely.
and
Display
the
FFT.
This
creates
an
acquisition
GAP
and
some
parts
of
the
signal
will
be
lost
(see
figure
3,
below).
This
means
a
standard
scope
based
FFT
also
can’t
be
used
for
measurement
of
pulsed
signals
because,
as
with
swept
analyzers
signals
can
be
missed
in
the
gaps
between
FFT
acquisition.
A
t
FFT
Lost
FFT
Lost
FFT
Lost
Calculation
Calculation
Calculation
Figure
3:
FFT
with
gaps
in
between
based
on
slow
calculation
process
In
real
time
analysis
multiple
FFT
acquisitions
and
calculations
are
performed
in
parallel
and
overlapping.
The
Real-‐time
analyzer
contains
additional
processing
capability
to
handle
this
additional
workload
to
capture,
calculate
and
display
the
results
results
in
real-‐time.
The
result
is
that
time
acquisition
of
different
FFT
blocks
is
Gap
Free
(see
figure
4
below)
and
performance
will
be
not
be
affected
by
choosing
using
different
RBW
settings.
A
t
A
fixed
number
of
1024
samples
are
used
for
one
FFT
time
acquisition.
Each
FFT
calculation
is
using
a
window
function.
Windowing
is
important
to
define
a
discrete
number
of
time
points
for
calculation.
Size
of
window
can
be
varied
and
is
not
fixed
in
time
domain.
A
variation
of
window
size
will
have
a
direct
influence
of
real
time
resolution
bandwidth
[RBW]
or
the
other
way
around:
with
changing
the
RBW,
size
of
window
will
be
changed.
The
position
of
a
time
signal
like
a
pulse
needs
to
be
in
the
center
of
FFT
window
to
transform
it
correctly
into
frequency
range.
In
case
that
a
pulse
is
in
between
two
FFT
events,
then
amplitude
is
suppressed
by
filter
side
loops
and
is
no
longer
correct
(see
figure
6).
A
FFT 1 FFT 2
t
Figure
5:
Amplitude
is
wrong
if
signal
is
located
in
between
of
two
FFT
blocks
An
overlapping
process
of
FFT
events
will
be
used
in
Real-‐time
analyzers
to
avoid
losing
signal
information.
Overlapping
has
the
effect
that
more
spectrums
are
available
over
a
time
period
and
time
resolution
is
higher.
Smaller
events
can
be
measured
(see
figure
7)
and
signal
suppression
of
single
FFT
acquisition
occurred
due
windowing
is
eliminated
with
overlapping.
A
FFT
5
FFT
4
FFT
3
FFT
2
FFT
1
t
Figure
6:
Overlapping
process
in
real
time
spectrum
analyzer
In
other
words,
overlapping
process
of
FFT
events
has
a
direct
influence
of
smallest
pulse
width
which
can
be
measured
with
a
real
time
spectrum
analyzer.
The
RT-‐SA
RSA5000
is
working
with
a
FFT
rate
of
146.484
FFT/sec.
which
results
into
a
calculation
speed
(Tcalc)
of
6,82
µsec.:
1 1
𝑇!"#! = = = 6,82 µμ𝑠𝑒𝑐.
𝐹𝐹𝑇 𝑅𝑎𝑡𝑒 𝐹𝐹𝑇
146.484
𝑠𝑒𝑐.
Depending
on
real
time
span
there
are
4
different
sample
rates
available.
The
maximum
sample
rate
is
51,2
MSa/sec.
With
that
sample
rate
and
the
fixed
number
of
samples
(NFix
=
1024),
used
for
one
FFT
acquisition,
the
duration
can
be
calculated
as
follow:
𝑁!"# 1
𝑇!"# = = 1024 𝑆𝑎.∗ = 20 µμ𝑠𝑒𝑐.
𝑆𝑎𝑚𝑝𝑙𝑒 𝑅𝑎𝑡𝑒 𝑀𝑆𝑎.
51,2
𝑠𝑒𝑐.
An
overlap
of
FFT
frames
is
not
possible
during
calculation
progress.
Therefore
the
overlapping
time
of
FFT
frames
can
be
calculated
with
that
formula:
For example with sample rate of 51,2 MSa./sec. the overlap time is 13,18 µsec or 65,86% which results into
Figure
7:
Measurement
of
a
pulse
of
7.45
µsec.
(period:
1
sec.)
with
amplitude
of
-‐35
d Bm.
Each
pulse
is
captured
with
correct
amplitude.
POI
calculation
depends
on
FFT
rate,
chosen
RBW
and
adjusted
Span.
The
principle
of
POI
is
described
with
a
span
of
40
MHz
(=51,2
MSa/sec.)
and
RBW
of
3.21
MHz
(Kaiser
Window)
in
figure
10.
Due
to
calculation
time,
second
FFT
acquisition
starts
after
6.82
µsec.
Window
size
is
depending
of
RBW
in
real
time
mode:
1
𝑇!"#$%& =
𝑅𝐵𝑊
Start position of first FFT acquisition and End position of second FFT acquisition defines POI time value
A
17,13
µsec.
FFT
2
(20
µsec.)
6,82
µsec.
9,68
µsec.
𝟏
𝑹𝑩𝑾∗
FFT
1
(20
µsec.)
t
7,45
µsec.
RIGOL
Figure
8–:
EUxample
ncompromised
Performance...
with
RT-‐Span
Unprecedented
of
40
MHz,
sample
Value
rate
of
51.2
MSa/s
PAGE
50
and
RBW
of
3.21
MHz
(Kaiser
Filter)
With
that
POI
and
speed
it
is
now
possible
to
measure
a
Bluetooth
signal
with
the
RT-‐SA
mode
of
RSA5000
series.
Usage
of
maximum
hold
is
no
longer
needed.
It
is
possible
to
set
6
different
RBW
settings
in
RT-‐SA
mode
and
speed
is
not
affected.
Real-‐time
Visualization:
Since
Real-‐time
analysis
adds
the
dimension
of
time
to
the
acquisition
the
user
can
choose
from
multiple
rich
data
views
helping
to
quickly
debug
their
designs
• Normal
Trace
Analysis
• Density
Analysis
• Spectrogram
• Power
vs
Time
In
Normal
Mode
the
trace
information
of
current
time
is
visible.
It
looks
like
a
trace
of
a
SA
but
due
to
the
real
time
sweep
more
information
is
visible
at
the
same
time
compare
to
SA.
Normal
trace
analysis
is
a
2D
measurement
(power
over
frequency).
With
Density
Analysis
it
is
possible
to
analyze
the
repetition
rate
of
a
signal.
Density
Mode
puts
data
into
a
color
map
based
on
the
probability
of
that
frequency
(from
blue
=
0%
to
red
=
100%,
see
figure
8).
As
more
often
the
signal
hits
a
single
pixel
point
(power
vs
frequency)
within
a
certain
time
it
increases
the
probability
and
changes
the
color
of
this
pixel.
For
example
a
constant
wave
[CW]
signal
would
be
visible
in
red
color.
A
very
short
single
signal
event
would
be
visible
in
blue
color.
The
color
percentage
can
be
calculated
as
follow:
𝑇!!"#
𝑛= ∗ 100%
𝑇!"#$%&'
Density
View
is
particularly
useful
for
identifying
superimposed
signals
and
separating
signals
that
share
the
same
frequency
and
for
seeing
time
varying
and
frequency
hopping
signals.
In
normal
and
density
mode
it
is
possible
to
activate
a
Spectogram
measurement.
Spectrogram
is
a
waterfall
displaying
frequency
over
time
allowing
users
to
measure
out
duration
of
pulses
(like
for
Bluetooth
signals),
to
clearly
view
frequency
changes
over
time
and
to
identify
interference.
A
spectrogram
also
works
with
a
color
scheme
for
signal
level
(DANL:
0%
=
blue,
Reference
level:
100%
=
red).
Density
in
combination
with
spectrogram
is
a
4D
measurement
(power
over
frequency
over
repetition
rate
and
power
over
time,
see
figure
12
with
a
Bluetooth
example).
Spectogram Display
Figure
9:
Bluetooth
signal
measured
with
density
spectrogram
In
Power
vs
Time
(PvT)
mode
it
is
possible
to
display
the
time
domain
of
a
signal
within
adjusted
real
time
bandwidth.
The
acquisition
time
can
be
changed
in
this
measurement.
The
Power
vs
Time
analysis
is
displayed
for
the
used
real
time
bandwidth
and
not
to
RBW
like
in
SA
with
zero-‐span
configuration.
Signal
bursts
of
modulated
signals
and
pulses
can
be
displayed
to
measure
duty
cycle
and
amplitude
of
a
pulse
or
to
display
pulse
trains
over
certain
time.
PvT
can
be
used
in
combination
of
normal
trace
analysis
(frequency
spectrum)
and
spectrogram
(see
figure
13).
Comparing
the
measurement
result
of
Bluetooth
signal
in
figure
12
and
figure
13
and
the
result
of
traditional
swept
spectrum
analyzer
(figure
2)
shows
the
engineer
has
much
more
information
available
with
Real-‐time.
Within
the
adjusted
real
time
bandwidth
all
frequency
components
can
be
measured.
Time
information
can
be
displayed
in
parallel
of
spectrum
measurement,
it
is
visible
that
this
signal
is
a
frequency
hopping
spread
spectrum
signal
and
the
length
of
data
block
can
be
analyzed.
The
Power
vs
Time
is
no
longer
depended
on
RBW
bandwidth
like
in
SA
and
frequency
domain
and
time
domain
can
be
displayed
in
one
time.
PvT Display
Figure
10:
normal
trace
vs
spectrogram
vs
PvT
of
a
Bluetooth
signal
A
tracking
generator
is
an
extension
of
the
sweep
circuit.
It
is
a
programmable
RF
source
with
the
ability
to
synchronize
the
output
frequency
with
the
sweep
steps
of
the
spectrum
analyzer.
In
this
way,
the
source
and
measurement
frequencies
are
locked.
If
measuring
from
1MHz
to
100MHz
on
the
spectrum
analyzer,
the
tracking
generator
outputs
a
continuous
sine
wave
with
a
frequency
that
will
sweep
from
1MHz
to
100MHz
in
full
synchronization
with
the
measurement.
Figure 4-‐1: Example of a spectrum analyzer with tracking generator sweeping a DUT.
Testing
Filters
A
filter
is
a
useful
component
in
many
designs.
The
primary
goals
of
a
filter
are
to
remove
unwanted
frequencies
and
to
enhance
desired
frequencies
from
an
input
signal.
Low
pass:
Allows
frequencies
lower
than
a
certain
value
to
pass
through,
and
rejects
higher
frequencies.
This
can
be
used
to
remove
high
frequency
noise
from
a
signal.
High
pass:
Allows
frequencies
higher
than
a
certain
value
to
pass
through,
and
rejects
lower
frequencies.
This
can
be
used
to
remove
low
frequency
components
from
the
input
signal.
Bandpass:
Allows
a
certain
frequency
range,
or
band,
to
pass
through
the
filter
and
rejects
those
frequencies
that
are
outside
of
the
operating
frequency
band.
This
type
of
filter
will
allow
a
band
of
frequencies
to
pass
through
the
filter
with
little
to
no
changes
while
drastically
lessening
the
amplitude
of
signals
having
frequencies
outside
of
the
operating
band.
Notch:
Rejects
those
frequencies
that
are
inside
of
the
operating
frequency
band
and
rejects
those
within
the
operating
band
of
the
filter.
This
type
of
filter
will
allow
all
frequencies
outside
of
the
operating
band
to
pass
through
the
filter
with
little
to
no
changes
while
drastically
lessening
the
amplitude
of
signals
having
frequencies
inside
of
the
operating
band.
This
is
the
exact
opposite
of
a
bandpass
filter.
In
all
cases,
testing
a
filter
provides
information
about
its
quality,
such
as
how
well
it
decreases
unwanted
signals
and
how
well
it
allows
wanted
signals
through.
Testing a filter using a spectrum analyzer has a number of common steps.
Required Hardware:
1. Connect the tracking generator output to the RF input using the same cabling that will be used to
test
the
device.
Any
element,
like
an
adapter,
used
during
normalization
should
also
be
used
during
device
measurement
because
any
changes
to
the
RF
signal
path
could
affect
the
accuracy
of
the
measurement.
NOTE:
Clean
the
surfaces
of
the
adapters
and
input
with
a
lint-‐free
cloth
or
swab
and
electronics
contact
cleaner
to
prevent
damage
and
ensure
repeatability.
2. Enable
the
tracking
generator.
3. Store
the
reference
trace.
4. Enable
the
normalization.
Now,
the
displayed
trace
will
more
accurately
represent
the
filter
by
removing
cable
and
adapter
losses.
• Measure
the
filter
1. Connect
the
tracking
generator
output
to
the
filter
input
using
the
appropriate
cabling
and
connectors.
2. Connect
the
filter
output
to
the
instrument
RF
input.
NOTE:
Clean
the
surfaces
of
the
adapters
and
input
with
a
lint-‐free
cloth
or
swab
and
electronics
contact
cleaner
to
prevent
damage
and
ensure
repeatability.
Figure
4-‐2:
Front
panel
of
a
spectrum
analyzer
equipped
with
a
tracking
generator.
This
view
shows
connections
to
the
RF
output
of
the
tracking
generator
and
the
RF
input
of
the
analyzer.
3. Configure
the
stop,
start
frequency
for
the
span
of
interest.
4. Set
the
tracking
generator
amplitude.
NOTE:
If
the
instrument
is
equipped
with
a
preamplifier,
enabling
it
can
lower
the
displayed
noise
floor.
5. Enable
the
tracking
generator.
Note
the
small
bump
in
Figure
4-‐3.
Figure
4-‐3:
A
bandpass
filter
trace.
Adjust
the
amplitude,
start,
and
stop
frequency
to
zoom
in
on
the
trace.
Some
analyzers
have
a
convenient
auto-‐scale
feature
that
can
automatically
configure
the
analyzer
to
display
a
full
view
of
the
area
of
interest
on
the
trace.
Figure
4-‐4:
After
Auto.
NOTE:
Some
analyzers
also
have
markers.
These
are
cursors
that
can
show
the
frequency
and
amplitude
of
specific
points
on
the
trace,
as
well
as
the
ability
to
measure
bandwidth
at
a
particular
dB
level.
Figure
4-‐5:
3dB
bandwidth
measurement
of
a
bandpass
filter.
Cable/Connector
Loss
Cables
and
connectors
can
have
a
dramatic
effect
on
the
accuracy
and
validity
of
measurements
on
additional
components.
They
also
wear
with
time
and
use.
This
wear
can
show
up
as
an
increase
in
attenuation
over
particular
frequency
ranges.
A
spectrum
analyzer
and
a
tracking
generator
allow
testing
the
insertion
loss
(loss
vs.
frequency)
of
the
cables
and
adapters
easily.
Required Hardware:
• Two
N-‐type
to
BNC
adapters
(Figure
4-‐6).
Select
adapters
that
convert
N-‐type
(in/out
connectors
on
most
spectrum
analyzers)
to
the
cable
type
being
tested.
Also
note
that
higher
quality
connectors
(silver-‐plated,
beryllium
copper
pins,
etc.)
equal
better
longevity
and
repeatability.
Figure
4-‐6:
N-‐type
to
BNC
adapter.
• A
short
reference
cable
with
terminations
that
match
the
adapters
being
used
and
the
cable-‐under-‐test.
• An
adapter
to
go
between
the
reference
cable
and
the
cable-‐under-‐test.
This
experiment
will
use
a
BNC
“barrel
connector”
(Figure
4-‐7).
Note
that
higher
quality
connectors
(silver-‐plated,
beryllium
copper
pins,
etc.)
equal
better
longevity
and
repeatability.
Figure
4-‐7:
BNC
barrel
adapter.
• Alternately,
use
two
adapters
and
a
short
cable
as
a
reference
assembly
to
normalize
the
display
before
making
cable
measurements.
This
removes
the
need
to
have
the
cable-‐to-‐cable
adapter.
• A
spectrum
analyzer
with
tracking
generator
(TG).
Test Steps:
1. Attach
the
adapters
to
the
tracking
generator
(TG)
output
and
RF
input.
NOTE:
Clean
the
surfaces
of
the
adapters
and
input
with
a
lint-‐free
cloth
or
swab
and
electronics
contact
cleaner
to
prevent
damage
and
ensure
repeatability.
2. Connect
the
reference
cable
to
the
TG
out
and
RF
In
on
the
analyzer.
Figure
4-‐8:
Measuring
the
reference
cable.
3. Adjust
the
span
of
the
scan
for
the
frequency
range
of
interest.
4. Adjust
the
tracking
generator
output
amplitude
and
spectrum
analyzer
display
to
view
the
entire
trace.
5. Enable
the
tracking
generator
output.
Figure
4-‐9:
Reference
cable
insertion
loss
before
normalization.
6. Normalize
the
reference
insertion
loss.
This
mathematically
subtracts
a
reference
signal
(stored
automatically)
from
the
input
signal.
Figure
4-‐10:
Reference
cable
insertion
loss
after
normalization.
7. Disconnect
the
reference
cable
from
the
RF
input.
Place
cable-‐to-‐cable
adapter
(BNC
barrel
or
other)
and
connect
to
the
cable
to
test.
8. Connect
the
cable-‐under-‐test
to
test
to
RF
input
and
enable
the
tracking
generator.
Figure
4-‐11:
Cable-‐under-‐test
connected.
The screen displays the cable-‐under-‐test losses plus the error of the cable-‐to-‐cable adapter.
Figure
4-‐12:
Zoomed
view
of
cable-‐under-‐test
loss
vs.
frequency.
With
the
bucket
analogy,
the
water
and
bucket
have
different
densities.
This
difference
affects
the
speed
of
the
wave
and
causes
a
portion
of
the
incoming,
or
incident,
wave
to
be
reflected
directly
opposite
to
the
original
path
of
motion.
As
the
waves
travel
back-‐and-‐forth
across
the
surface,
they
will
add
and
subtract
depending
on
their
relative
phases.
When
an
electromagnetic
wave
travels
from
one
material
to
another,
it
can
also
undergo
reflection.
When
an
electromagnetic
wave
encounters
a
material
with
a
different
impedance,
some
or
all
of
the
wave
can
be
reflected.
If
the
impedance
of
the
new
material
is
much
greater
than
the
original
impedance,
a
large
amount
of
the
original
wave
can
be
reflected.
Large
amounts
of
reflected
energy
can
be
inefficient
and
could
cause
damage
to
sensitive
components
in
the
circuit.
The
Voltage
Standing
Wave
Ratio
(VSWR)
is
a
measurement
that
can
be
used
to
judge
the
quality
of
the
impedance
matching
for
cables
and
other
passive
devices.
It
is
also
mainly
used
to
determine
the
efficiency
of
antennas
to
radiate
power.
A
perfectly
matched
antenna/cable
system
will
have
a
VSWR
of
1.
In
reality,
the
best
systems
will
have
VSWR
values
of
1.1
or
1.2.
Higher
VSWRs
indicate
a
greater
degree
of
impedance
mismatch.
For
antenna
users,
this
indicates
that
more
of
the
original
signal
is
reflected
back
towards
the
transmitter
and
it
indicates
a
system
that
has
low
efficiency.
Required
Hardware:
A
spectrum
analyzer
with
a
tracking
generator,
such
as
the
Rigol
DSA815-‐TG
• Coaxial
Directional
Coupler
or
VSWR
Bridge
with
an
impedance
match
for
the
component
being
tested.
This
is
typically
50Ω
for
most
antenna
and
RF
networks.
• An
antenna
or
other
component
to
test
• Impedance-‐matched
coaxial
cable
to
connect
the
coupler
to
the
tracking
generator.
Depending
on
the
Directional
Coupler
design,
you
may
need
another
cable
or
adapter
to
connect
it
to
the
DSA.
Figure
4-‐13:
A
Rigol
VB1032
VSWR
bridge.
Test Steps:
1. Clean
all
connectors
with
a
lint-‐free
cloth
or
swab
and
electronics
contact
cleaner.
Make
sure
to
remove
any
contamination,
dirt,
or
metallic
flakes
in
and
around
mating
surfaces.
2. Connect
the
RF
coupler
coupled
(CPL)
connection
to
the
RF
input
of
the
DSA.
3. Connect
the
cable
from
the
TG
output
on
the
front
panel
of
the
DSA
to
the
output
(OUT)
of
the
coupler.
NOTE:
Leave
the
input
(IN)
of
the
coupler
open.
This
will
provide
100%
reflection
and
be
used
to
minimize
the
effects
of
the
cabling,
adapters,
and
coupler.
4. Configure
the
DSA
frequency
span
for
the
DUT.Correct
for
the
cabling,
adapters,
and
coupler
effects
by
storing
the
open
circuit
(no
DUT
=
Max
Reflection)
conditions
by
normalizing
the
current
setup.Enable
the
tracking
generator.
5. Connect
the
DUT
of
interest
to
the
coupler
input
(IN).
6. If
the
analyzer
is
equipped
with
markers,
select
a
marker
to
identify
your
location
on
the
trace.
The
minimum
value
displayed
on
the
DSA
is
the
return
loss
of
the
DUT.
Record
this
value
in
dB.
s = VSWR
r = 10^(-‐0.05 * a)
For example, if measuring an antenna with the following parameters:
a = -‐26.22 dB
Whether
wireless
or
wired,
the
main
requirements
remain
the
same.
The
transmitted
signal
must
have
the
proper
amplitude
in
the
proper
frequency
band
to
be
picked
up
by
the
receiver.
Measuring
transmitters
is
common
practice
throughout
the
design
process
and
is
also
used
to
monitor
existing
transmitters
to
ensure
that
the
signals
remain
in
their
specific
operating
bands.
For
example,
AM
and
FM
radio
stations
commonly
monitor
their
transmitters
to
ensure
that
they
are
operating
within
their
licensed
frequency
band.
This
section
will
describe
some
common
tests
that
can
be
used
to
verify
transmitter
performance,
then
build
on
those
ideas
by
citing
some
specific
tests
used
within
a
specific
transmission
type.
Output
Power
The
strength
of
an
RF
transmission
can
be
affected
by
many
outside
factors.
Imagine
all
the
different
materials
that
an
FM
radio
transmission
will
encounter
as
the
signal
leaves
the
radio
station
antenna
and
arrives
at
a
receiver.
The
signal
may
have
to
pass
through
glass,
drywall,
furniture,
trees,
and
even
people
before
it
reaches
its
intended
target.
Even
weather
conditions,
like
air
density,
humidity,
and
storms,
can
affect
the
transmission.
By
measuring
the
output
power,
it’s
possible
to
verify
the
signal
is
present
and
has
enough
power
to
be
picked
up
at
the
receiver.
The
output
power
is
simply
a
measure
of
the
transmitted
signal’s
strength.
Measuring
the
output
power
directly
at
the
transmitter
allows
verifying
if
the
transmitter
is
working
correctly.
If
all
is
well,
it’s
time
to
move
down
the
transmission
path
and
perform
remote
measurements
at
a
distance
from
the
transmitter,
using
antennas
in
place
of
the
cable
from
the
transmitter
to
the
measurement
instrument.
Figure
5-‐1:
Transmitter
direct
output
measurement
using
a
cable
and
measurement
instrument.
Figure
5-‐2:
Transmitter
remote
output
measurement
using
antennas
and
a
measurement
instrument.
A
simple
power
measurement
can
be
performed
using
a
dedicated
RF
power
meter
or
a
spectrum
analyzer.
Although
power
meters
tend
to
be
more
accurate,
they
often
have
a
longer
measurement
time
than
spectrum
analyzers.
Transmission
Band
The
transmission
frequency
is
another
important
measurement
that
must
be
performed
to
characterize
a
transmitter.
Testing
the
frequency
band
means
directly
measuring
the
frequency
(or
frequencies,
in
some
cases)
that
a
signal
is
occupying
in
the
RF
spectrum.
It’s
important
to
be
sure
that
the
transmission
signal
has
the
proper
frequency
to
be
detected
by
the
receiver
and
that
the
signal
is
not
interfering
in
frequency
bands
near
the
desired
frequency.
Signals
that
bleed
into
or
occupy
adjacent
bands
can
cause
interference
and
disturb
the
reception
of
the
signals
that
are
supposed
to
be
occupying
that
band.
AM
Transmission
Test
Amplitude
modulation
(AM)
is
a
common
method
of
adding
information
to
an
RF
signal.
The
amplitude
of
the
carrier
waveform
changes
proportionally
to
the
input
signal.
AM
is
typically
used
to
transmit
voice
information
and
was
the
primary
modulation
scheme
for
initial
research
in
radio
communications
in
the
early
twentieth
century.
In the time domain, a typical AM signal will look something like Figure 5-‐3.
Figure
5-‐3:
10MHz
carrier,
1kHz
modulation
AM
signal
on
a
scope.
In
this
case,
the
carrier
signal
has
a
frequency
of
10MHz
and
the
amplitude
modulation
is
set
to
1kHz.
Note
that
the
base
period
in
the
oscilloscope
display
in
Figure
5-‐3
shows
periodic
beats,
or
nulls,
in
the
time-‐based
waveform.
These
are
areas
where
the
carrier
amplitude
is
near
zero.
Note
that
the
beat
frequency
matches
the
amplitude
modulation
of
1kHz.
By zooming in to a single beat, notice that the carrier waveform is still present at 10MHz (Figure 5-‐4).
Figure
5-‐4:
10MHz
carrier,
1kHz
modulation
AM
signal
zoom
view
on
scope.
Although
it’s
possible
to
use
an
oscilloscope
to
verify
the
frequency
and
amplitude
of
an
AM
signal,
it
is
more
common
to
use
a
spectrum
analyzer.
The
analyzer
can
provide
a
better
impedance
match
and
obtain
a
more
accurate
measurement
of
the
performance
of
the
transmitter,
especially
with
signals
that
have
low
power
and
high
carrier
frequencies.
Here’s how to perform a basic AM transmitter test using a spectrum analyzer.
Required Hardware:
• Spectrum
analyzer
• Transmitter
• Cables,
adapters,
or
antennas
to
connect
to
the
transmitter
and
analyzer
• An
external
attenuator
(optional)
may
be
required
to
limit
the
signal
power
that
is
directed
to
the
analyzer.
Test Steps:
Figure
5-‐5:
10MHz
carrier,
1kHz
modulation
AM
signal
on
a
spectrum
analyzer.
Delta
markers
are
enabled
to
show
the
frequency
and
amplitude
differences
between
the
carrier
and
modulation
peaks.
Note
that
the
center
peak
matches
the
carrier
frequency
of
900MHz
and
the
two
additional
peaks
are
1kHz
away
from
the
carrier.
It’s
also
possible
to
measure
the
power
of
the
signal.
In
this
case,
the
carrier
is
-‐50dBm.
The
input
signal
in
this
case
was
modulated
with
a
fixed
1kHz
input.
With
an
AM
signal
modulated
by
audio
or
voice
information,
the
modulated
peaks
will
actually
vary
with
time.
Many
analyzers
have
a
few
features
that
can
help
with
these
real-‐world
measurements.
If
the
instrument
is
so
equipped,
consider
enabling
the
Max
Hold
trace
type
(Figure
5-‐6).
Max
Hold
traces
are
similar
to
histograms.
Each
frequency
bin
value
will
only
display
the
maximum
value.
This
value
will
remain
for
successive
scans
until
a
greater
value
is
measured
for
that
frequency
bin.
This
allows
the
analyzer
to
“build
up”
the
full
modulation
envelope
over
a
series
of
scans.
Figure
5-‐6:
10MHz
carrier
with
varying
modulation
AM
signal
on
a
spectrum
analyzer.
The
yellow
trace
is
a
“Clear
Write”
trace
type.
The
pink
trace
is
a
“Max
Hold”
trace
type
that
was
built
over
successive
scans.
If
the
spectrum
analyzer
is
equipped
with
pass/fail
masking,
set
up
a
limit
mask
to
indicate
quickly
whether
a
particular
signal
is
within
the
test
limits,
as
shown
in
Figure
5-‐7:
Figure
5-‐7:
Spectrum
analyzer
showing
a
pass/fail
mask
on
an
AM
Max
Hold
trace
FM
Deviation
Frequency
modulation
(FM)
is
another
common
method
of
adding
information
to
an
RF
signal.
The
frequency
of
the
carrier
waveform
changes
proportionally
to
the
input
signal.
FM
is
typically
used
to
transmit
voice
information.
In the time domain, a typical FM signal will look like Figure 5-‐8:
Figure
5-‐8:
10MHz
carrier,
1kHz
modulation
FM
signal
on
a
scope
with
persistence
on
to
help
show
frequency
modulations.
In
this
case,
the
carrier
signal
has
a
frequency
of
10MHz
and
the
frequency
modulation
is
set
to
1kHz.
Frequency
modulation
on
an
oscilloscope
can
be
difficult
to
capture
due
to
the
triggering
model
most
scopes
use.
In
order
to
visualize
the
modulations
of
the
frequency
components
of
the
incoming
signal,
it’s
necessary
to
lengthen
the
persistence
time
of
the
display.
Persistence
determines
the
length
of
time
that
a
waveform
is
held
on
the
display.
Longer
persistence
times
will
hold
waveforms
on
the
display
for
a
longer
period
of
time
and
allow
for
a
direct
comparison
of
waveforms
over
a
period
of
time.
As
shown
in
Figure
5-‐8,
the
frequency
of
the
sine
wave
is
changing
with
time.
This
is
shown
by
the
wider
waveform
thickness
near
the
edges
of
the
displayed
waveform.
Due
to
their
constantly
changing
frequency,
FM
signals
are
difficult
to
analyze
on
an
oscilloscope,
even
one
with
FFT
analysis
capabilities.
Because
a
spectrum
analyzer
displays
frequency
information,
it
can
be
very
handy.
Follow these steps to perform a basic FM transmitter test using a spectrum analyzer.
Required Hardware:
• Spectrum
analyzer
• Transmitter
• Cables,
adapters,
or
antennas
to
connect
to
the
transmitter
and
analyzer
• An
external
attenuator
(optional)
may
be
required
to
limit
the
signal
power
that
is
directed
to
the
analyzer.
Test
Steps:
1. Connect
the
transmitter
to
the
cable/adapters
or
transmission
antenna.
NOTE:
Clean
the
surfaces
of
the
adapters
and
input
with
a
lint-‐free
cloth
and
electronics
contact
cleaner
to
prevent
damage
and
ensure
repeatability.
2. Connect
the
other
end
of
the
cable/adapters
or
receiver
antenna
to
the
RF
input
of
the
spectrum
analyzer.
NOTE: If using an external attenuator, place it on the RF input of the analyzer.
3. Set
the
center
frequency
of
the
analyzer
to
the
carrier
frequency
of
the
input
signal.
In
the
example
in
Figure
5-‐9,
a
100MHz
FM
transmission
is
being
monitored.
4. Set
the
analyzer
frequency
span
to
10¥
the
expected
modulation
frequency.
In
this
example,
the
carrier
is
being
modulated
at
1kHz,
so
the
span
is
set
to
10kHz.
5. Set
the
resolution
bandwidth
(RBW)
to
a
value
less
than
the
modulation
frequency.
In
the
Figure
5-‐9,
the
RBW
is
set
to
100Hz.
If
an
RBW
close
to
the
modulation
frequency
is
used,
there
might
not
be
sufficient
resolution
to
see
the
modulation.
Figure
5-‐9:
100MHz
carrier,
1kHz
modulation
FM
signal
on
a
spectrum
analyzer
The
input
signal
in
Figure
5-‐9
was
modulated
with
a
fixed
1kHz
input.
A
real-‐world
FM
signal
would
be
modulated
by
audio
or
voice
information
that
could
have
a
non-‐linear
change
in
frequency.
This
would
cause
the
frequency
to
vary
with
time.
If
the
Max
Hold
trace
type
is
available,
it
can
be
enabled,
just
as
suggested
with
the
AM
signal.
The
FM
Frequency
Deviation
of
the
signal
can
be
measured
in
this
way
by
collecting
sweep
data
over
a
period
of
time.
Figure
5-‐10:
100MHz
carrier
with
varying
modulation
FM
signal
on
a
spectrum
analyzer.
The
yellow
trace
is
a
“Clear
Write”
trace
type.
The
pink
trace
is
a
“Max
Hold”
trace
type
that
was
built
over
successive
scans.
FM
deviation
measurements
are
important
because
they
allow
visualizing
the
frequencies
being
used
for
transmission.
If
the
deviation
is
too
large,
the
transmission
may
interfere
with
nearby
channels.
By
monitoring
the
transmission,
adjustments
can
be
made
to
maintain
proper
transmission
characteristics
and
stay
within
the
proper
band.
Harmonics
and
Spurs
An
ideal
transmitter
would
deliver
the
exact
signal
intended,
with
no
additional
unwanted
components.
Unfortunately,
there
are
no
ideal
transmitters.
In
reality,
a
transmitter
can
have
undesirable
signal
components
like
excessive
harmonics
and
spurs.
Luckily,
there
are
a
few
ways
to
identify
and
minimize
them.
Transmitters
commonly
use
amplifiers
to
boost
the
signal
strength.
Unfortunately,
most
amplifier
designs
will
add
and
amplify
the
harmonics
of
the
output
signal.
Chapter
1
discussed
superposition
of
sinusoidal
waveforms
and
how
harmonics
of
a
sine
wave
can
be
built
up
to
create
different
waveform
shapes.
A
harmonic
is
simply
a
waveform
with
a
frequency
that
is
an
integer
value
of
the
intended
signals
frequency.
For
example,
for
a
sine
wave
with
a
fundamental
frequency
of
10MHz,
the
2nd
harmonic
is
a
20MHz
sine
wave
(Figure
5-‐11).
The
second
harmonic
is
two
times
the
fundamental
frequency,
the
third
is
three
times,
etc.
Figure
5-‐11:
An
oscilloscope
displaying
a
sine
wave
with
a
frequency
of
10MHz
(yellow)
and
another
sine
wave
with
a
frequency
of
20MHz
(light
blue).
Figure 5-‐12 is a screen capture of a 10MHz sinewave from a high quality RF source.
st nd
Figure
5-‐12:
10MHz
sine
input
into
a
spectrum
analyzer.
Note
1
and
2
harmonics.
Note
that
the
2nd
harmonic
has
a
frequency
that
is
twice
the
fundamental
(20MHz)
and
the
third
peak
is
at
30MHz,
or
three
times
the
fundamental.
Even
though
the
source
was
programmed
to
source
an
output
at
10MHz,
there
are
still
some
additional
components
to
the
output
sinewave.
When
searching
for
harmonics,
it
is
important
to
widen
the
frequency
span
on
the
analyzer
in
order
to
capture
them.
If
the
fundamental
frequency
of
the
transmitter
is
100MHz,
it
may
be
wise
to
look
at
a
span
from
100MHz
to
500MHz
or
more,
in
order
to
capture
a
larger
span
of
potential
harmonics.
Harmonics
also
tend
to
be
significantly
lower
in
power
than
the
fundamental
frequency.
Note
how
the
power
level
drops
significantly
between
the
fundamental
(-‐10dB)
and
the
harmonics
(-‐64dBm,
-‐73dBm)
in
Figure
5-‐12.
This
can
make
them
difficult
to
capture
using
an
oscilloscope.
Lowering
the
RBW
value
and
using
preamplification
(if
available)
will
lower
the
noise
level
of
the
analyzer
and
help
to
isolate
these
low
powered
signals.
If
issues
with
excessive
harmonics
arise,
note
that
many
can
be
minimized
by
using
filters
or
alternative
transmitter
designs.
Spurious
emissions,
or
spurs,
can
also
be
problematic.
A
spur
is
typically
the
unwanted
result
of
nonlinear
components
in
a
circuit
or
transmission
path.
Nonlinear
components
include
amplifiers,
mixers,
and
diodes,
but
they
can
also
be
created
by
oxide
layers
on
the
mating
surfaces
of
cables
and
adapters.
Hunting
for
spurs
is
very
similar
to
hunting
for
harmonics.
Configure
the
spectrum
analyzer
span
to
cover
a
frequency
range
wide
enough
to
cover
the
expected
location
of
the
spurs
and
lower
the
noise
floor
by
using
the
RBW
and
preamplifier
(if
available).
Unfortunately,
spurs
can
be
caused
by
different
events.
This
leads
to
their
location
at
unexpected
frequencies,
not
even
multiples
of
a
fundamental
like
harmonics.
Many
spurs
are
products
of
intentional
or
unintentional
mixing
of
signals.
Investigating
areas
where
mixing
products
from
known
signals
is
a
good
starting
spot.
Chapter
3
addressed
mixers
and
mixing
products.
In
the
simplest
case,
a
mixer
takes
two
signals
as
inputs
and
the
resultant
output
contains
the
original
signals,
as
well
as
the
addition
and
subtraction
of
the
inputs.
Figure
5-‐13:
A
simple
mixer.
The
best-‐case
scenario
is
to
identify
the
cause
of
the
spurs
and
minimize
the
unwanted
components
by
filtering,
but
investigation
into
connector
torque
(the
“tightness”
of
a
connection)
and
the
cleanliness
of
physical
connections
can
also
be
helpful.
Receivers
are
available
in
both
analog
(AM/FM
radio)
and
digital
(WiFi,
Bluetooth,
Zigbee)
types,
but
they
all
operate
on
similar
base
principles.
Figure
6-‐1
is
a
generic
block
diagram
of
a
typical
receiver.
Figure
6-‐1:
Block
diagram
of
a
typical
radio
receiver.
Note
that
other
receiver
types
are
similar
in
their
block
diagrams.
The
major
differences
are
the
demodulation
(analog
vs.
digital)
and
output
(speaker
vs.
data).
This
chapter
provides
a
brief
overview
of
each
element
in
a
typical
audio
receiver
and
presents
some
common
test
procedures
for
each,
moving
backward,
starting
at
the
speaker
and
ending
with
the
antenna.
In
real
life,
it’s
possible
to
jump
to
test
any
section
on
its
own,
but
this
backward
technique
allows
using
the
speaker
to
provide
instant
audible
feedback
when
stepping
through
each
design
area.
Receivers
typically
contain
filters
and
amplifiers
that
can
also
be
tested
individually
using
techniques
presented
in
chapter
4.
Review
those
sections
for
more
specific
tests
on
each
component,
if
desired.
Required Hardware:
• Function
or
arbitrary
waveform
generator
like
the
Rigol
DG1022
(Figure
6-‐2)
• Cabling
to
connect
generator
to
the
speaker.
This
is
typically
a
BNC-‐to-‐alligator
connection.
Figure
6-‐2:
An
arbitrary
waveform
function
generator.
Test Steps:
1. Study
the
schematics
for
the
design.
Clearly
identify
high
voltages
and
ensure
adequate
shielding
from
any
high
voltage
areas.
2. Disconnect
the
speaker
wires
from
the
receiver.
3. Check
to
make
sure
the
speaker
connections
are
clear
of
any
contamination
or
dirt.
Clean
connectors
with
a
cotton
or
lint-‐free
swab
and
an
electronics
contact
cleaner
like
DeoxIT®
D5
if
needed.
Allow
solvents
to
evaporate
before
turning
on
any
electrical
devices
nearby.
4. Connect
the
function
generator
output
to
the
speaker
input
wires.
5. On
the
generator,
set
the
waveform
to
Sine,
the
frequency
to
1kHz,
and
the
amplitude
to
1V
(peak-‐to-‐
peak)
and
listen
for
sounds
from
the
speaker.
6. OPTIONAL:
Adjust
the
frequency
and
voltage
of
the
generator
to
test
the
frequency
response
and
volume
of
the
speaker.
Humans
can
typically
hear
frequencies
from
20Hz
up
to
20kHz.
A
properly
working
speaker
should
have
a
noticeable
change
in
output
sound
when
the
frequency
and
amplitude
of
the
input
signal
are
adjusted.
If
the
speaker
does
not
have
sound
output,
it
may
need
to
be
replaced.
Audio
Amplifier
(Optional)
The
audio
amplifier
circuit
is
designed
to
take
the
low-‐level
audio
signal
output
from
the
audio
decoder
and
amplify
the
power
so
that
the
signal
is
strong
enough
to
power
the
speaker.
If
the
audio
amplifier
is
not
working,
the
speaker
may
have
low
or
inaudible
output.
A
simple
functional
test
for
an
audio
amplifier
is
this:
Does
the
amplifier
take
a
small
input
signal
and
produce
an
output
signal
that
can
drive
the
speaker?
More
specific
tests
that
can
help
characterize
the
performance
of
an
amplifier
are
available
but
are
beyond
the
scope
of
this
document.
Required Hardware:
1. Study
the
schematics
for
the
design.
Clearly
identify
high
voltages
and
ensure
adequate
shielding
from
any
high
voltage
areas.
Disconnect
the
audio
input
and
output
wires
from
the
audio
amplifier.
2. Connect
the
function
generator
to
the
amplifier
audio
inputs
and
set
the
generator
to
an
audio
tone
(1kHz
for
example)
at
a
low
voltage
(10mVp-‐p
or
so)
3.
Connect
the
oscilloscope
to
the
amplifier
outputs.
A
speaker
can
also
be
used
in
place
of
the
oscilloscope,
to
hear
the
results
rather
than
see
them.
4.
Power
on
the
amplifier
and
turn
on
the
generator
and
scope.
5.
Configure
the
generator
to
output
a
sine
wave
at
2kHz
or
so.
The
audio
range
of
human
hearing
lies
from
20Hz
to
20kHz,
so
2kHz
offers
a
nice
starting
point.
6.
Enable
the
output
on
the
generator
and
observe
the
output
on
the
oscilloscope.
Start
with
a
small
voltage
output
on
the
generator
(10mV
or
less)
and
compare
this
to
the
output
of
the
amplifier
as
shown
on
the
oscilloscope.
The
output
frequency
measured
by
the
scope
should
match
the
input
frequency
from
the
generator
and
the
output
amplitude
should
be
higher.
Adjust
the
frequency
and
amplitude
of
the
generator
and
observe
the
output
on
the
scope.
If
the
output
signal
shows
excessive
distortion,
incorrect
frequencies,
or
excessive
noise,
it
may
require
repair
or
replacement.
Detector/Demodulator
The
detector
removes
the
carrier
RF
frequency
(the
channel
to
which
the
transmitter
is
tune)
and
passes
the
data
(or
audio)
through
to
the
next
stage.
The
simplest
test
to
perform
on
the
detector/demodulator
is
to
input
a
signal
at
the
receiver
IF
frequency
(determined
by
the
type
of
receiver
and
the
specific
design)
that
includes
the
modulation
type
for
the
receiver
type.
For
an
AM
receiver,
add
AM
modulated
audio.
For
FM
add
FM
modulated
audio.
For
data,
add
the
specific
data
modulation
type
to
the
RF
carrier.
In any case, the idea is the same. Deliver a known test signal to the detector and monitor the output.
Required Hardware:
• A
signal
source
capable
of
creating
the
modulated
signal
(AM,
FM,
or
data
type
that
matches
the
receiver
type)
at
the
IF
frequency
of
the
receiver.
This
could
be
a
single
instrument
or
it
could
be
combination
of
a
standalone
frequency
or
pattern
generator
coupled
with
an
RF
source
(Figure
6-‐3).
Many
RF
sources
have
external
modulation
inputs
that
can
accept
a
modulated
input
signal
and
will
mix
it
with
an
RF
output.
Figure
6-‐3:
A
Rigol
DG1022
arbitrary
waveform
generator
and
a
Rigol
DSG815
RF
Source.
• Oscilloscope
(optional).
Visually
verify
the
output
of
the
detector
stage
using
an
oscilloscope
if
there
is
no
audio
capability
with
the
receiver.
The
oscilloscope
can
also
be
used
to
check
the
data
output
visually
if
the
receiver
is
designed
for
data
transmission
as
well.
Test Steps:
1. Study
the
schematics
for
the
design.
Clearly
identify
high
voltages
and
ensure
adequate
shielding
from
any
high
voltage
areas.
2. Configure
the
RF
source
to
output
the
proper
IF
signal;
this
includes
the
modulated
and
carrier
portions
of
the
signal.
3. Connect
the
RF
source
to
the
detector
input
and
set
it
to
an
appropriate
power
level
for
the
input
to
the
detector.
4. Enable
the
power
to
the
detector
and
output
for
the
RF
source.
The
output
signal
can
be
monitored
using
the
speaker
(if
applicable)
by
using
an
oscilloscope
to
verify
the
amplitude
and
waveform
output
from
the
detector
visually.
Not
all
receivers
have
this
specific
design.
Some
designs
incorporate
other
hardware
or
integrate
the
individual
components
shown
in
one
package.
Figure
6-‐4:
Receiver
with
Selection
and
IF
sections
outlined
in
orange
IF
Amplification
The
input
signal
amplitude
can
be
decreased
significantly
in
the
IF
section.
To
make
demodulation
more
successful,
there
can
also
be
an
IF
amplification
stage.
Testing
the
IF
amplifier
is
simply
a
matter
of
sourcing
a
known
modulated
signal
and
carrier
into
the
amplifier
and
measuring
the
output.
Required Hardware:
o A
signal
source
capable
of
creating
the
modulated
signal
(AM,
FM,
or
data
type
that
matches
the
receiver
type)
at
the
IF
frequency
of
the
receiver.
This
could
be
a
single
instrument
or
it
could
be
the
combination
of
a
standalone
frequency
or
pattern
generator
coupled
with
an
RF
source.
Many
RF
sources
have
external
modulation
inputs
that
can
accept
a
modulated
input
signal
and
will
mix
it
with
an
RF
output.
o Spectrum
analyzer,
oscilloscope,
or
the
previously
tested
portions
of
the
receiver
(Audio
detector,
Amplifier,
and
Speaker)
to
monitor
the
true
audio
output,
if
the
rest
of
the
system
has
tested
to
be
functioning
properly.
Test Steps:
1. Study
the
schematics
for
the
design.
Clearly
identify
high
voltages
and
ensure
adequate
shielding
from
any
high
voltage
areas.
2.
Configure
the
RF
source
to
output
the
proper
IF
signal;
this
includes
the
modulated
and
carrier
portions
of
the
signal.
3. Connect
the
RF
source
to
the
detector
input
and
set
it
to
an
appropriate
power
level
for
the
input
to
the
amplifier.
NOTE:
Input
IF
signals
to
the
amplifier
are
typically
very
small
(-‐100dBm).
If
the
RF
source
being
used
is
not
capable
of
output
amplitude
values
that
are
low
enough,
add
external
attenuation.
External
attenuation
is
also
useful
to
help
protect
the
spectrum
analyzer
input
if
the
amplifier
output
is
expected
to
be
high
enough
to
damage
the
input
of
the
spectrum
analyzer.
4. Enable
the
power
to
the
amplifier
and
output
for
the
RF
source.
The
output
signal
can
be
monitored
using
a
spectrum
analyzer
or
use
an
oscilloscope
to
verify
the
amplitude
and
waveform
output
from
the
amplifier
visually.
Testing
the
LO
is
as
simple
as
monitoring
the
output
frequency
using
a
spectrum
analyzer,
oscilloscope,
or
frequency
counter.
Required Hardware:
Test Steps:
1. Study
the
schematics
for
the
design.
Clearly
identify
high
voltages
and
ensure
adequate
shielding
from
any
high
voltage
areas.
2. Connect
the
output
of
the
LO
to
the
input
of
the
spectrum
analyzer.
3. Configure
the
analyzer
center
frequency
to
correspond
to
the
expected
output
frequency
of
the
LO.
4. Apply
power
to
the
LO
and
check
the
output
frequency
and
amplitude
on
the
analyzer.
Compare
these
values
to
the
specifications
for
the
LO
being
tested.
If
the
analyzer
is
capable
of
performing
a
Max
Hold
trace,
enable
it
and
monitor
the
LO
over
a
period
of
time.
Max
Hold
traces
provide
a
histogram
of
the
input
over
successive
scans.
This
effectively
holds
maximum
amplitude
encountered
at
each
frequency
step
and
can
allow
identifying
drifting
LO
values.
Figure
6-‐5
shows
the
drift
of
an
LO
over
time
captured
using
a
Max
Hold
(purple)
trace
over
a
period
of
time.
Note
that
the
frequency
of
this
oscillator
has
drifted
across
a
range
of
frequencies.
Figure
6-‐5:
10MHz
LO
output
on
a
spectrum
analyzer.
The
yellow
trace
is
a
Clear
Write
trace
type.
The
purple
trace
is
a
Max
Hold
trace
type
showing
the
drift
of
the
LO
over
time.
If
the
LO
is
not
working
properly,
a
function
generator
set
to
the
LO
frequency
can
be
used
as
a
temporary
replacement
for
the
nonfunctional
LO
stage.
This
will
allow
troubleshooting
the
remaining
portions
of
the
receiver.
Frequency
Conversion/Mixer
The
main
function
of
the
frequency
conversion
stage
is
to
mix
the
input
signal
(RF
and
modulated
signal)
with
the
LO
and
convert
it
to
the
intermediate
frequency
(IF)
and
filter
out
unwanted
mixing
products.
This
is
similar
to
the
Mixer
and
IF
stage
of
the
swept
superheterodyne
spectrum
analyzer
discussed
in
chapter
3.
In
this
test,
an
RF
source
with
modulation
capabilities
(or
a
separate
function
generator
acting
as
the
modulation
input)
is
used
to
deliver
an
RF
signal
with
modulation
to
the
input
of
the
IF
section.
A
spectrum
analyzer
(or
the
previously
tested
receiver
sections,
if
functional)
is
used
to
check
the
performance
of
the
IF
stage.
Required
Hardware:
• A
signal
source
capable
of
creating
the
modulated
signal
(AM,
FM,
or
data
type
that
matches
the
receiver
type)
at
the
IF
frequency
of
the
receiver.
This
could
be
a
single
instrument
or
it
could
be
the
combination
of
a
standalone
frequency
or
pattern
generator
coupled
with
an
RF
source.
Many
RF
sources
have
external
modulation
inputs
that
can
accept
a
modulated
input
signal
and
will
mix
it
with
an
RF
output.
• Spectrum
analyzer
1. Study
the
schematics
for
the
design.
Clearly
identify
high
voltages
and
ensure
adequate
shielding
from
any
high
voltage
areas.
2. Configure
the
RF
source
to
output
the
proper
RF
signal;
this
includes
the
modulated
and
carrier
portions
of
the
signal.
3. Connect
the
RF
source
to
the
mixer
input
and
set
it
to
an
appropriate
power
level
for
the
input
to
the
amplifier.
NOTE:
The
input
signals
at
this
stage
typically
very
small
(-‐100dBm).
If
the
RF
source
being
used
is
not
capable
of
output
amplitude
values
that
are
low
enough,
add
external
attenuation.
External
attenuation
can
also
help
protect
the
spectrum
analyzer
input
if
the
amplifier
output
is
expected
to
be
high
enough
to
damage
the
input
of
the
spectrum
analyzer.
4. Enable
the
power
to
the
amplifier
and
output
for
the
RF
source.
Monitor
the
output
signal
using
a
spectrum
analyzer
or
use
an
oscilloscope
to
verify
the
amplitude
and
waveform
output
from
the
amplifier
visually.
If
access
to
the
mixer/frequency
conversion
section
is
available,
a
spectrum
analyzer
with
a
tracking
generator
can
be
used
to
test
the
filter
performance,
as
outlined
in
the
Testing
Filters
section
of
chapter
4.
Selection/RF
Stage
The
selection
portion
of
a
receiver
is
a
tunable
filter
designed
to
accept
a
defined
frequency
and
reject
others
outside
of
the
selected
band.
A
properly
operating
selection
stage
should
allow
the
selected
frequency
through
and
reject
out-‐of-‐band
signals.
Required Hardware:
Test Steps:
1. Study
the
schematics
for
the
design.
Clearly
identify
high
voltages
and
ensure
adequate
shielding
from
any
high
voltage
areas.
2. Configure
the
tracking
generator
of
the
spectrum
analyzer
to
output
an
amplitude
appropriate
for
the
input
to
the
selection
stage.
3. Connect
the
tracking
generator
output
to
the
selection
stage
input.
4. Configure
the
spectrum
analyzer
to
scan
a
frequency
range
that
is
appropriate
to
the
selection
filter
setting
and
bandwidth.
5. Connect
the
output
of
the
selection
stage
to
the
tracking
generator.
6. Enable
the
tracking
generator
output
and
observe
the
output
of
the
selection
stage
on
the
spectrum
analyzer
display.
The
output
of
the
selection
stage
should
effectively
filter
any
out-‐of-‐band
frequencies
and
pass
in-‐band
frequencies.
It’s
recommended
to
test
other
frequency
settings
on
the
selection
stage
by
changing
the
tuner
on
the
receiver
to
different
channels.
Remember
to
change
the
analyzer
frequency
scan
range
as
well
to
coincide
with
these
new
settings.
Antennas,Cabling,
and
Adapters
The
antenna
is
designed
to
collect
RF
signals
and
deliver
them
through
the
cable
to
the
receiver.
Antennas
come
in
many
shapes,
sizes,
and
materials.
They
can
be
general
purpose,
like
the
old
“rabbit
ears”
antennas
used
with
broadcast
television
or
they
can
be
ruggedized
and
designed
to
be
wavelength
or
directionally
specific.
Regardless
of
their
design,
antennas
are
built
to
transmit
or
receive
electromagnetic
radiation.
A
visual
inspection
of
the
antenna,
the
physical
mounting
hardware,
and
all
cables
and
connectors
can
go
a
long
way
towards
identifying
problem
areas
with
the
receiver.
Antennas
and
cables
that
are
outdoors
are
subject
to
wide
temperature
swings
and
constantly
changing
environmental
conditions.
Corrosion
and
weathering
can
lead
to
high
impedances,
open/broken
connections,
and
a
host
of
other
issues.
Additional
information
can
be
found
in
the
section
on
VSWR
testing
as
covered
in
the
components
section
of
chapter
4.
Thanks to Jeff Covelli (WA8SAJ) for his assistance on this chapter.
Many
processors
and
digital
communications
buses
are
rated
by
their
clock
speed.
Faster
clocks
generally
indicate
faster
performance.
Many
processors
feature
clocks
in
the
4–6GHz
range,
and
many
chip-‐to-‐chip
communications
on
the
circuit
board
are
into
the
hundreds
of
megahertz.
Signals
in
these
frequency
ranges
can
interfere
with
intentional
transmissions
in
these
frequency
bands
and
can
also
cause
other
devices
to
behave
erratically.
These
effects
aren’t
localized
to
processors
and
digital
buses.
Any
circuit
element
that
exhibits
sharp
rise
times,
square
pulse
edges,
or
transient
signals
can
emit
radiation
that
can
interfere
with
other
devices.
To
minimize
the
side
effects,
a
few
standard
measurements
can
be
performed
on
a
design.
This
chapter
introduces
test
methodologies
and
some
fundamental
first
steps
that
can
help
one
gain
a
better
understanding
of
some
common
electromagnetic
interference
(EMI)
tests.
Compliance
vs.
Pre-‐Compliance
Almost
any
electronic
design
slated
for
commercial
use
is
subject
to
EMC
(electromagnetic
compatibility)
testing.
Any
company
intending
to
sell
a
product
must
ensure
that
it
is
tested
to
the
specifications
set
forth
by
the
regulatory
body
of
the
country
in
which
it
will
be
sold.
In
the
United
States,
the
Federal
Communications
Commission
(FCC)
specifies
rules
on
EMC
testing
using
guidelines
and
test
plans
developed
by
the
International
Special
Committee
on
Radio
Interference
(CISPR)
and
the
International
Electrotechnical
Commission
(IEC).
To
be
sold
legally,
a
sample
of
the
electronic
product
must
pass
a
series
of
specific
tests.
In
many
cases,
companies
can
self-‐test
and
certify
that
the
product
meets
the
standard
limits,
but
they
must
have
detailed
reports
of
the
test
conditions
and
data.
Many
companies
choose
to
have
these
tests
performed
by
accredited
compliance
company.
Testing
that
meets
all
of
the
details
outlined
in
the
test
specifications
for
the
device
is
referred
to
as
compliance
testing.
Fully
compliant
tests
follow
the
guidelines
and
setup
exactly
and
the
results
can
be
used
to
certify
the
device
for
compliance
to
the
test
procedures.
However,
this
type
of
testing
can
be
expensive
and
may
require
specialized
test
environments
and
equipment
that
might
be
beyond
the
reach
of
many
companies.
One
method
to
lower
the
additional
costs
associated
with
ensuring
EMC
compliance
is
to
perform
EMC
testing
throughout
the
design
process,
well
before
sending
the
product
off
for
full
compliance
testing.
This
pre-‐
compliance
testing
can
be
very
cost-‐effective
and
can
be
tailored
to
match
the
conditions
used
for
compliance
testing
closely.
This
will
increase
the
company’s
confidence
in
passing
compliance
the
first
time,
lower
test
costs,
and
speed
time
to
market.
The
following
sections
introduce
pre-‐compliance
test
methods
that
can
be
used
to
troubleshoot
a
design’s
potential
EMI-‐related
issues.
Although
the
simple
tools
and
test
methods
described
can
ease
the
reader
into
the
world
of
EMI/EMC
testing,
there
is
much
more
that
can
be
learned
independently
about
full
compliance
testing.
Of
the
many
texts
available
for
independent
study,
we
recommend
Henry
Otts’s
Electromagnetic
Compatibility,
as
well
as
EMI
Troubleshooting
Cookbook
for
Product
Designers
by
André
and
Wyatt.
Measuring
Radiated
EMI/Near-‐Field
An
unintentional
radiator
is
a
device
that
transmits
RF
as
a
byproduct
or
side
effect
of
its
intended
operation.
In
simple
terms,
an
unintentional
radiator
is
much
like
a
radio
transmitting
noise
across
multiple
stations
at
one
time.
These
radiated
transmissions
can
have
an
adverse
effect
on
other
devices
that
receive
this
undesired
signal.
One
common
test
is
to
measure
the
radiated
emission
of
a
product
by
placing
a
spectrum
analyzer
and
an
antenna
a
few
meters
away
from
it.
This
is
called
a
far-‐field
radiated
emissions
test
(Figure
7-‐1).
Figure
7-‐1:
Simple
block
diagram
of
a
radiated
emissions
test.
In
this
test,
an
antenna
is
attached
to
a
spectrum
analyzer
and
measurements
are
performed
over
a
frequency
span
of
interest.
Although
the
idea
sounds
simple,
this
test
method
has
some
major
problems:
• Most
antennas
are
broadband.
That
is,
they
accept
a
wide
range
of
frequencies
and
do
not
differentiate
between
those
outside
of
the
test
area
and
the
actual
test
device.
This
makes
any
measurements
performed
using
an
antenna
very
susceptible
to
other
RF
signals
like
radio
stations,
WiFi,
etc.
• Other
environmental
factors
(metal
shelving,
desks,
people,
etc.)
can
also
adversely
affect
the
measurements.
This
can
be
due
to
reflections
and
absorption
of
the
signals
by
the
environment
and
can
make
repeatable
and
accurate
measurements
nearly
impossible.
Successfully
executing
fully
compliant
radiated
emissions
tests
involve
a
similar
setup,
but
they
also
require
the
use
of
test
areas
with
very
low
external
RF
sources
that
could
interfere
with
the
measurement.
Open
Air
Test
Sites
(OATS)
are
basically
open
test
areas
in
geographic
locations
that
feature
minimal
RF
signals…
such
as
an
open
prairie
in
the
middle
of
North
Dakota,
for
example.
OATS
facilities
were
very
popular
in
the
twentieth
century,
but
their
numbers
have
dwindled
due
to
the
exponential
increase
in
external
RF
sources.
At
that
time,
most
RF
was
produced
by
AM
and
FM
radio,
and
open
air
facilities
could
be
placed
in
areas
without
much
RF
interference.
Now,
with
WiFi,
mobile
phones,
and
Bluetooth
devices,
RF
signals
are
everywhere.
A
more
useful
solution
comes
in
the
form
of
special
test
chambers
that
attenuate
external
RF
signals
and
minimize
internal
RF
reflections.
These
anechoic
and
semi-‐anechoic
chambers
provide
this
function
but
can
be
very
expensive.
A
small
semi-‐anechoic
chamber
a
few
feet
wide
by
a
few
feet
tall
can
cost
$50,000
or
more,
which
in
most
cases
is
too
costly
for
infrequent
test
needs.
Near-‐field
probing
is
a
test
technique
that
uses
special
probes
that
minimize
the
effects
of
stray
environmental
RF.
Near-‐field
probes
typically
have
two
designs.
Magnetic-‐field,
or
H-‐field,
probes
feature
loops
that
help
couple
the
magnetic
fields
produced
by
time-‐varying
currents.
Both
designs
are
very
sensitive
to
distance
from
the
source
of
the
radiation.
They
need
to
be
within
a
few
inches
of
most
sources
to
measure,
even
with
a
very
sensitive
analyzer
or
preamp
attached.
Figure
7-‐2:
Rigol
NFP
Near
Field
EMC
probes.
The
simplest
test
is
to
configure
the
DSA
to
use
the
peak
detector
and
set
the
RBW
and
Span
for
the
area
of
interest
per
the
regulatory
requirements
for
the
device
being
tested.
Then
select
the
proper
E
or
H
probe
for
the
design
and
scan
over
the
surface
of
the
design.
Probe
orientation
(rotation,
distance)
is
also
important
to
consider.
The
probes
act
as
an
antenna,
picking
up
radiated
emissions
from
seams,
openings,
traces,
and
other
elements
that
could
be
emitting
RF.
A
thorough
scan
of
all
of
the
circuit
elements,
connectors,
knobs,
openings
in
the
case,
and
seams
is
crucial.
Figure
7-‐4
shows
using
an
H-‐field
probe
to
test
emitted
radiation.
For
the
first
pass,
configure
the
spectrum
analyzer
to
use
the
peak
detector.
This
will
provide
a
“worst-‐case”
reading
on
the
radiated
RF
and
it
is
the
quickest
path
to
determining
the
problem
areas.
Larger
probes
will
allow
a
faster
scanning
rate,
albeit
with
less
spatial
resolution.
Once
problem
areas
are
identified,
a
few
common
techniques
allow
getting
more
details
on
them.
Whenever
possible,
select
a
spectrum
analyzer
that
has
the
standard
configuration
used
in
full
compliance
testing.
This
includes
a
quasi-‐peak
detector
mode,
EMI
filter,
and
resolution
bandwidth
(RBW)
settings
that
match
the
full
test
requirements
specified
for
the
product.
This
type
of
setup
will
increase
testing
time
but
should
be
used
on
the
problem
areas.
A
full
compliance
test
utilizes
these
settings,
so
pre-‐compliance
testing
with
this
configuration
will
provide
a
greater
degree
of
visibility
into
the
EMI
profile
of
a
design.
Many
instruments
also
support
storing
cable
and
antenna
correction
factors
that
will
make
it
possible
to
see
the
true
signal,
without
the
added
errors
from
the
setup.
The
next
step
in
radiated
testing
includes
using
antennas
in
place
of
the
near-‐field
probes
and
a
rotating
platform
for
the
equipment
under
test
(EUT).
It
can
also
include
a
special
room
that
minimizes
environmental
factors
(semi-‐anechoic).
These
setups
are
beyond
the
scope
of
this
document,
but
references
at
the
end
of
this
chapter
provide
good
references
for
the
details
of
that
setup.
Near-‐Field
Probing
Setup
Board-‐level
emission
testing
can
be
performed
using
a
spectrum
analyzer,
like
the
Rigol
DSA-‐815
(9kHz
to
1.5GHz),
near-‐field
electric
(E)
and
magnetic
(H)
probes,
and
the
appropriate
connecting
cable.
Figure
7-‐3:
The
Rigol
DSA815-‐TG
Spectrum
Analyzer.
Using
the
near-‐field
probes
shown
in
Figure
7-‐3
with
a
Rigol
spectrum
analyzer
requires
a
50Ω
cable
that
terminates
in
an
N-‐type
connector
(spectrum
analyzer
end)
and
SMB
(probe
end).
It’s
also
possible
to
build
probes
by
removing
a
few
centimeters
of
outer
shield
and
insulator
from
a
semi-‐rigid
RF
cable,
bending
it
into
a
loop,
and
dipping
in
plastic
tool
dip
or
other
insulating
material.
Larger
diameter
loops
will
pick
up
smaller
signals
but
do
not
have
as
much
spatial
resolution
as
smaller
diameter
loops.
For
the
first
pass,
configure
the
spectrum
analyzer
to
use
the
peak
detector.
This
setting
ensures
that
the
instrument
is
capturing
the
“worst-‐case”
peak
RF.
It
also
provides
a
fast
scan
rate
to
minimize
the
time
spent
at
one
position
when
scanning
over
the
DUT.
Larger
probes
will
allow
for
a
faster
scanning
rate,
albeit
with
less
spatial
resolution.
Smaller
probes,
like
the
E-‐field
probe,
provide
fine
spatial
resolution
and
can
be
used
to
detect
RF
on
single
pins
of
circuit
elements.
Probe
orientation
(rotation,
distance)
is
also
important
to
consider.
The
probes
act
as
an
antenna,
picking
up
radiated
emissions.
Exposing
the
loop
to
the
largest
perpendicular
field
possible
will
maximize
the
signal
strength.
Figure
7-‐4:
An
example
of
using
an
H-‐field
probe
and
spectrum
analyzer
to
find
trouble
spots
on
a
board.
Note
the
orientation
of
the
H-‐field
probe
Take
care
to
test
enclosure
seams,
openings,
traces,
and
other
elements
that
could
be
emitting
RF
signals.
A
thorough
scan
of
all
of
the
circuit
elements,
connectors,
knobs,
openings
in
the
case,
and
seams
is
crucial
to
identifying
potential
areas
where
RF
can
“leak”
out
of
an
enclosure.
Figure
7-‐5:
Measuring
a
display
ribbon
cable
for
emissions
using
an
H-‐field
probe.
Use
tinfoil
or
conductive
tape
to
cover
suspected
problem
areas
like
vents,
covers,
doors,
seams,
and
cables
coming
through
an
enclosure.
Simply
test
the
area
without
the
foil
or
tape,
then
cover
the
suspected
area,
and
rescan
with
the
probe.
Once
problem
areas
have
been
identified,
implementing
a
few
common
techniques
can
provide
additional
detail.
If
possible,
select
a
spectrum
analyzer
that
has
the
standard
configuration
used
in
full
compliance
testing.
This
includes
a
quasi-‐peak
detector
mode,
EMI
filter,
and
resolution
bandwidth
(RBW)
settings
that
match
the
full
test
requirements
specified
for
the
product.
This
type
of
setup
will
increase
testing
time
but
should
be
used
on
the
problem
areas.
A
full
compliance
test
utilizes
these
settings,
so
pre-‐compliance
testing
with
this
configuration
will
provide
a
greater
degree
of
visibility
into
the
EMI
profile
of
a
design.
Tips
Near-‐field
probes
and
a
spectrum
analyzer
can
be
useful
tools
in
troubleshooting
EMI
issues.
• With H-‐field probes, try different probe orientations to help isolate problem areas.
• Remember
to
probe
all
of
the
seams,
openings,
and
cutouts
around
the
enclosure
of
the
device.
Surface
contact
between
mating
parts,
as
well
as
the
finish
of
the
materials,
can
adversely
affect
the
grounding
and
shielding
properties
of
an
enclosure.
• Openings
radiate
just
like
solid
structures.
They
act
like
antennas.
• Ribbon
cables
and
cables/inputs
with
bad
shielding
and
grounds
are
common
causes
of
radiated
emissions.
Measuring
Conducted
EMI
Conducted
EMI
testing
requires
analyzing
the
RF
energy
that
is
coupled
from
the
instrument
or
test
circuit
to
the
main
power
line
to
which
it
is
connected.
RF
signals
traveling
down
the
power
line
can
cause
interference
in
the
AM
radio
transmission
bands.
In
order
to
minimize
this
interference,
it
is
important
to
quantify
the
RF
power
and
frequencies
that
a
product
produces
when
plugged
into
the
power
grid.
Like
radiated
EMI,
conducted
EMI
is
measured
using
a
spectrum
analyzer,
but
it
also
requires
a
transient
limiter
and
a
Linear
Impedance
Stabilization
Network
(LISN).
An
LISN
isolates
the
power
mains
from
the
equipment
under
test,
isolates
any
noise
generated
by
the
EUT,
and
couples
the
signals
generated
by
the
EUT
to
the
spectrum
analyzer.
As
with
emissions
testing,
the
best
start
is
a
scan
over
the
frequency
range
of
interest
using
the
peak
detector
on
the
spectrum
analyzer.
If
any
peaks
are
within
3dB
or
so
of
the
limit
line,
make
note
of
their
frequencies.
Then,
perform
a
quasi-‐peak
scan
over
a
1MHz
span
centered
around
each
peak.
This
will
save
you
measurement
time
and
provide
a
more
accurate
representation
of
the
true
amplitude
of
the
RF
power
of
the
EMI.
Conducted
Emission
Measurement
Setup
The
more
closely
this
setup
matches
a
full
compliance
setup,
the
more
closely
the
data
acquired
will
match
with
the
lab.
However,
this
isn't
always
practical.
Figures 7-‐6 and 7-‐7 show the standard suggested electrical and physical setups for testing conducted
emissions:
Figure
7-‐6:
Electrical
connections
for
conducted
EMC
testing.
Figure
7-‐7:
Physical
connections
for
conducted
EMC
testing.
6. The
horizontal
and
vertical
ground
planes
are
typically
sheets
of
metal
with
surface
areas
twice
the
dimensions
of
the
equipment-‐under-‐test.
7. The
horizontal
and
vertical
ground
planes
should
be
electrically
bonded
to
each
other.
8. Equipment
should
be
placed
on
an
insulated
table
over
the
horizontal
ground
plane.
No
equipment
or
cabling
should
run
below
the
equipment.
9. The
LISN
should
be
electrically
bonded
to
the
horizontal
ground
plane.
The
LISN’s
job
is
to
separate
the
AC
Mains
noise
from
the
conducted
noise
being
generated
by
the
equipment-‐under-‐test.
Select
a
LISN
that
has
the
proper
voltage,
current,
and
frequency
ranges
for
the
equipment-‐under-‐test.
10. Do
not
coil
cables;
minimize
inductive
loops
by
laying
cabling
out
smoothly.
11. The
spectrum
analyzer
should
be
placed
some
distance
away
from
the
horizontal
ground
plane,
typically
a
few
feet
away.
Test
Procedure
Once
the
EUT
is
set
up
and
bonded
the
LISN
and
ground
planes,
power
on
the
spectrum
analyzer
for
at
least
30
minutes
to
ensure
stability
and
accuracy.
Configure
Spectrum
Analyzer
• Option:
If
the
analyzer
has
settings
for
enabling
quasi-‐peak
detection
as
well
as
the
FCC
resolution
bandwidths
of
200Hz,
9kHz,
and
120kHz,
they
can
be
used
to
obtain
data
that
could
more
accurately
represent
data
collected
during
true
compliance
tests.
Be
aware
that
quasi-‐peak
detectors
can
take
a
much
longer
time
for
scan
completion.
• Set
resolution
bandwidth.
NOTE:
The
resolution
bandwidth
is
determined
by
the
standard
and
specific
device
type
under
test.
As
an
example,
FCC
subpart-‐15
specifies
an
RBW
of
9kHz
when
testing
from
150kHz
to
30MHz.
• Consult
the
standards
to
which
the
product
or
circuit
is
being
tested
to
for
more
information
on
the
specifications
governing
the
testing
process.
NOTE: Many specifications give limits and values in dBµV or V.
• Optional: Set scale for volts if the analyzer has that capability.
NOTE:
Some
analyzers
have
pass/fail
features
that
allow
configuring
an
upper
limit
line.
This
can
be
useful
when
evaluating
the
frequency
scan
with
respect
to
the
limits
set
forth
by
the
EMC
standard
to
which
testing
is
being
performed.
• Any
limit
lines
can
be
saved
to
the
internal
storage.
• Set
detector
type
to
Positive
Peak.
The
positive
peak
detector
will
show
the
highest
value
and
provide
the
“worst-‐case”
value
for
pre-‐compliance
scans.
• Adding
an
external
transient
limiter
and
10dB
to
20dB
of
external
attenuation
is
recommended
to
minimize
the
likelihood
of
damage
to
the
sensitive
front
end
of
the
spectrum
analyzer.
The
attenuator
protects
the
input
circuit
from
any
unknown
signals
that
could
damage
the
input.
It
also
serves
as
a
convenient
check
on
overloading
after
checking
the
background
readings.
• The
DSA
has
protection
circuitry,
but
there
are
transients
that
are
too
fast
to
protect
against.
• Set
frequency
start,
stop
values
set
forth
in
the
EMC
Specifications
that
apply
to
the
product.
• Set
the
resolution
bandwidth
(RBW)
to
the
value
set
forth
in
the
EMC
Specifications
that
apply
If
the
scan
changes
value
with
different
attenuation
settings,
then
it
is
likely
that
the
input
is
being
overloaded
with
broadband
power
and
additional
attenuation
is
recommended.
Try
comparing
scans
of
20dB
and
30dB,
etc.
until
a
range
without
variation
is
found.
The
objective
is
to
select
the
smallest
attenuation
value
that
does
not
show
errors
due
to
the
overloading
effects
of
the
input
signal.
In
the
worst
case,
the
EUT
may
not
be
able
to
be
tested
successfully
with
a
spectrum
analyzer,
so
a
true
EMI
receiver
with
pre-‐selection
filters
might
be
needed.
• Observe
conducted
emissions
and
look
for
frequency
lines
that
are
above
the
limit
line
previously
set.
Make
note
of
the
frequencies
failing
the
limit
lines.
Quasi-‐Peak
Scans
• Using
the
failed
frequencies
noted
previously,
adjust
the
spectrum
analyzer
to
center
the
failed
peak.
• Note
the
RBW
setting
for
the
scan,
and
make
the
frequency
span
twice
the
RBW
setting
used
for
the
peak
scan.
For
example,
if
there
is
an
over
limit
peak
at
10MHz,
and
an
RBW
of
120kHz,
center
the
quasi-‐peak
scan
at
10MHz,
and
scan
from
9.88MHz
to
10.12MHz.
Optional:
Change
the
detector
type
to
quasi-‐peak.
NOTE:
The
quasi-‐peak
detector
is
based
on
charge
and
discharge
times
of
a
standardized
resonant
circuit.
This
detector
type
can
take
more
than
three
times
the
scan
time
of
a
peak
measurement.
That
is
why
it
is
best
to
use
quasi-‐peak
only
over
short
spans.
• Compare
the
quasi-‐peak
data
to
the
pass/fail
limit
line
for
that
frequency.
• It
is
advisable
to
keep
the
conducted
emissions
at
least
10dB
lower
than
the
specified
limit
line.
This
margin
of
error
will
increase
the
likelihood
of
passing
a
full
compliance
test.
• It
is
also
advisable
to
compare
the
pre-‐compliance
data
and
setup
to
that
of
the
full
compliance
lab
that
will
perform
the
EMC
certification
testing.
This
will
make
it
easier
to
identify
any
problems
with
pre-‐
compliance
testing.
More
comparisons
make
it
possible
to
hone
the
pre-‐compliance
error
budget
and
allow
more
confidence
in
the
results
obtained.
Immunity
Testing
Products
that
contain
electronics
can
be
sensitive
to
radio
frequency
(RF)
interference.
Devices
that
experience
RF
interference
can
be
prone
to
improper
or
failed
operations.
Products
that
suffer
problems
when
exposed
to
RF
interference
are
said
to
be
susceptible
to
interference;
products
that
do
not
exhibit
issues
when
exposed
to
RF
interference
are
said
to
be
immune
to
interference.
Design
analysis,
including
part
selection,
shielding,
and
cable
selection
is
the
first
step
in
creating
a
product
that
is
capable
of
operating
“as
expected”
under
a
certain
degree
of
RF
interference;
however,
testing
early
under
real-‐world
conditions
is
one
sure
way
to
determine
if
a
design
is
susceptible
to
any
issues
with
RF.
The
following
section
illustrates
how
an
RF
generator
and
some
simple
tools
can
be
useful
in
identifying
weaknesses
in
a
product
design.
Radiated
Susceptibility
Radiated
susceptibility
tests
involve
observing
the
operation
of
a
device-‐under-‐test
(DUT)
while
it
is
being
subjected
to
a
known
RF
source.
The
signal
is
delivered
to
the
DUT
using
antennas
for
far-‐field
testing
or
near-‐
field
probes
for
board-‐level
tests.
Most
radiated
susceptibility
regulations
are
based
on
IEC
61000-‐4-‐3,
which
defines
the
test
signals
range
from
80–1,000MHz.
This
signal
can
be
modulated
by
a
1kHz
AM
sine
wave
with
80%
modulation
depth.
The
modulated
signal
helps
to
identify
any
rectification
issues
within
sensitive
circuit
elements
quickly.
In
far-‐field
tests,
an
RF
signal
source,
like
the
Rigol
DSG3000
or
DSG800
series,
is
connected
to
an
antenna
that
is
set
up
a
meter
or
two
from
the
DUT.
The
RF
source
is
then
configured
to
source
an
output
with
80%
AM
modulation
at
1kHz.
The
amplitude
should
be
set
as
high
as
possible,
and
the
carrier
frequency
can
be
set
at
9kHz.
NOTE:
The
DUT
should
be
configured
in
its
most
commonly
used
state.
All
cabling
(power,
I/O,
etc.)
should
be
connected
and
in
place.
Cables
can
act
like
antennas
and
can
directly
influence
performance.
Observe
the
DUT
for
any
functional
changes
or
issues
such
as
a
glitchy
or
noisy
display.
Now,
increase
the
carrier
frequency
and
check
the
DUT.
Step
the
center
frequency
of
the
generator
and
continue
to
observe
the
DUT,
making
note
of
the
carrier
frequencies
that
cause
issues
and
the
type
of
problems
observed.
After
the
stepping
up
to
1GHz
is
complete,
rotate
the
DUT
with
respect
to
the
antenna
and
re-‐test
to
provide
a
more
thorough
test.
NOTE:
Antennas
should
only
be
used
in
shielded
anechoic
or
semi-‐anechoic
chambers
to
prevent
interference
with
communications
and
emergency
broadcast
bands.
It
is
illegal
to
broadcast
over
many
frequency
bands
without
proper
licensing.
The
use
of
an
RF
source
like
the
Rigol
DSG3000
or
DSG800
series
offers
the
flexibility
to
adjust
the
wavelength,
power,
and
modulation
of
the
output
to
help
identify
problem
areas
quickly.
Figure
7-‐8:
The
Rigol
DSG800
and
DSG3000
RF
Source.
Near-‐field
testing
is
helpful
because
it
does
not
require
specialized
chambers
for
testing.
The
E-‐
and
H-‐field
probes
only
produce
strong
fields
at
distances
less
than
one
inch
from
the
tip
of
the
probe
and
do
not
radiate
efficiently
enough
to
cause
problems
with
broadcast
and
emergency
systems.
Their
small
size
also
allows
pinpointing
the
RF
at
specific
circuit
elements.
A commercial example of near-‐field probes are the Rigol probes shown in Figure 7-‐9.
Figure
7-‐9:
Rigol
NFP
EMC
probes.
The
loops
are
H-‐field
type.
The
RF
source
is
configured
exactly
as
in
the
far-‐field
test,
but
this
time,
the
probe
tip
is
placed
very
close
to
the
circuit
or
elements
of
the
board
at
each
carrier
frequency.
While
scanning
across
the
circuit,
observe
the
DUT
and
be
sure
to
check
for
any
issues,
especially
near
sensitive
analog
circuitry.
Figures
7-‐10
through
7-‐12
show
what
happens
when
the
shielding
is
removed
from
an
oscilloscope
board
and
an
E-‐field
probe
and
a
DSG3000
are
used
to
deliver
RF
signals
into
the
sensitive
analog
front
end.
Note
how,
with
the
shielding
removed,
the
RF
causes
data
corruption
and
changes
the
waveform
significantly.
Figure
7-‐10:
Using
a
near-‐field
probe
on
an
unshielded
oscilloscope
analog
input
circuit.
Figure
7-‐11:
Oscilloscope
data
with
shielding
in
place.
Figure
7-‐12:
Oscilloscope
data
without
shielding.
Additional
Testing
Another
useful
test
technique
is
to
use
a
current
probe
and
RF
source
to
deliver
RF
signals
to
cables
connected
to
the
DUT.
Cables
can
act
like
antennas
and
couple
undesired
signals
to
the
DUT.
This
setup
can
be
used
to
step
through
different
frequencies
and
check
the
susceptibility
of
a
design.
Commercial
current
probes
can
be
used,
but
an
acceptable
current
probe
can
be
built
using
a
snap-‐on
ferrite
choke,
a
few
winds
of
insulated
wire,
some
epoxy,
and
a
BNC
connector
as
shown
in
Figure
7-‐13.
Figure
7-‐13:
Hand-‐made
current
probe.
The
first
step
is
to
set
up
the
DUT
and
connect
all
of
the
cables
that
are
common
to
its
usage.
Configure
the
source
to
output
maximum
power
with
the
same
80%
1kHz
AM
modulation
that
was
suggested
previously
for
far-‐
and
near-‐field
tests
and
observe
the
DUT
for
problems.
Step
the
carrier
frequency
and
observe.
Perform
this
test
to
the
maximum
desired
frequency
and
repeat
the
process
on
each
cable
used
with
the
DUT.
NOTE:
Clamp
probes
should
only
be
used
in
shielded
anechoic
or
semi-‐anechoic
chambers
to
prevent
interference
with
communications
and
emergency
broadcast
bands.
It
is
illegal
to
broadcast
over
many
frequency
bands
without
proper
licensing.
To
demonstrate
the
use
of
a
current
probe,
an
experiment
was
performed
on
a
USB-‐powered
demonstration
board,
using
a
DSG3000
and
a
hand-‐made
probe
clamped
to
a
non-‐filtered
USB
cable
connected
to
the
board.
The
output
signal
was
monitored.
Figure
7-‐14:
Injection
of
RF
to
an
unfiltered
USB
cable.
Figure
7-‐15:
Sine
wave
from
board
without
RF
interference.
However,
when
an
RF
signal
was
applied,
the
output
began
to
show
signs
of
interference.
The
worst
interference
occurred
at
an
RF
carrier
frequency
of
110MHz,
as
shown
in
Figure
7-‐16.
Figure
7-‐16:
Noisy
data
showing
RF
interference
at
111.1MHz
due
to
injected
noise.
Figure
7-‐17:
Zoomed
data
to
show
details.
In
conclusion,
an
RF
source
like
the
Rigol
DSG3000
or
DSG800
and
some
simple
tools
allow
testing
designs
for
immunity
issues
early
in
the
development
process,
saving
time
and
money.