No Slide Title
Download
Report
Transcript No Slide Title
Network Analyzer Basics
Author: David Ballo
Network Analyzer Basics
DJB 12/96 na_basic.pre
1
Network analysis is not...
Router
Bridge
Repeater
Hub
Your IEEE 802.3 X.25 ISDN
switched-packet data stream is
running at 147 MBPS with a
-9
BER of 1.523 X 10 . . .
Network Analyzer Basics
DJB 12/96 na_basic.pre
2
Duplexers
Diplexers
Filters
Couplers
Bridges
Splitters, dividers
Combiners
Isolators
Circulators
Attenuators
Adapters
Opens, shorts, loads
Delay lines
Cables
Transmission lines
Waveguide
Resonators
Dielectrics
R, L, C's
Low
Integration
High
What types of devices are tested?
Passive
RFICs
MMICs
T/R modules
Transceivers
Receivers
Tuners
Converters
Antennas
Switches
Multiplexers
Mixers
Samplers
Multipliers
Diodes
Device type
VCAs
Amplifiers
VCOs
VTFs
Oscillators
Modulators
VCAtten's
Transistor
s
Active
Network Analyzer Basics
DJB 12/96 na_basic.pre
3
Device Test Measurement Model
Complex
84000
RFIC test
Ded. Testers
VSA
SA
Harm. Dist.
LO stability
Image Rej.
Gain/Flat. Compr'n
Phase/GD AM-PM
Isolation
Rtn Ls/VSWR
Impedance
S-parameters
Response tool
VNA
TG/SA
SNA
NF
NF Mtr.
Imped. An.
Simple
Param. An.
Power Mtr.
Det/Scope
Intermodulation
Distortion
NF
Full call
sequence
Pulsed S-parm.
Pulse profiling
BER
EVM
ACP
Regrowth
Constell.
Eye
LCR/Z
I-V
Measurement plane
Absol.
Power
Gain/Flatness
DC CW
Simple
Swept
freq
RF
Swept
power
Noise
2-tone
Stimulus type
Multitone
Complex
Pulsedmodulation
Protocol
Complex
Network Analyzer Basics
DJB 12/96 na_basic.pre
4
Agenda
Why do we test components?
What measurements do we make?
Network analyzer hardware
Signal separation devices
Broadband versus narrowband detection
Dynamic range
T/R versus S-parameter test sets
Three versus four samplers
Error models and calibration
Smith chart review
Transmission line basics
Reflection and transmission parameters
S-parameter definition
Types of measurement error
One- and two-port models
Error-correction choices
TRL versus TRL*
Basic uncertainty calculations
Typical measurements
Advanced topics
Network Analyzer Basics
DJB 12/96 na_basic.pre
5
Why do we need to test components?
Components often used as building blocks
Need to verify specifications
Examples:
filters to remove harmonics
amplifiers to boost LO power
mixers to convert reference signals
When used to pass communications signals, need to ensure
distortionless transmission
Linear networks
constant amplitude
linear phase / constant group delay
Nonlinear networks
harmonics, intermodulation
compression
noise figure
When absorbing power (e.g. an antenna),
need to ensure good match
Network Analyzer Basics
DJB 12/96 na_basic.pre
6
Linear Versus Nonlinear Behavior
A * Sin 360° * f ( t - t )
°
A
Linear behavior:
input and output frequencies are the same
(no additional frequencies created)
output frequency only undergoes magnitude
and phase change
Time
to
Sin 360° * f * t
A
Time
f
1
Input
phase shift =
to * 360° * f
Frequency
Output
DUT
Nonlinear behavior:
output frequency may undergo frequency shift
(e.g. with mixers)
additional frequencies created (harmonics,
intermodulation)
f
1
Frequency
Time
f
1
Frequency
Network Analyzer Basics
DJB 12/96 na_basic.pre
7
Criteria for Distortionless Transmission
Linear Networks
Linear phase over bandwidth of
interest
Phase
Magnitude
Constant amplitude over bandwidth of interest
Frequency
Frequency
Network Analyzer Basics
DJB 12/96 na_basic.pre
8
Magnitude Variation with Frequency
F(t) = sin wt + 1 /3 sin 3wt + 1 /5 sin 5wt
Time
Time
Magnitude
Linear
Network
Frequency
Network Analyzer Basics
DJB 12/96 na_basic.pre
Frequency
Frequency
Network Analyzer Basics
DJB 12/96 na_basic.pre
H
9
Phase Variation with Frequency
F(t) = sin wt + 1 /3 sin 3wt + 1 /5 sin 5wt
Linear Network
Time
Magnitude
Time
Frequency
0°
Frequency
-180°
Frequency
-360°
Network Analyzer Basics
DJB 12/96 na_basic.pre
10
Criteria for Distortionless Transmission
Nonlinear Networks
Saturation, crossover, intermodulation, and other nonlinear effects can cause signal
distortion
Time
Frequency
Time
Frequency
Network Analyzer Basics
DJB 12/96 na_basic.pre
11
Example Where Match is Important
KPWR
FM 97
Wire and bad antenna (poor match at
97 MHz) results in 150 W radiated
power
KPWR
FM 97
Proper transmission line and antenna results in
1500 W radiated power - signal is received about
three times further!
Good match between antenna and RF amplifier is extremely
important to radio stations to get maximum radiated power
Network Analyzer Basics
DJB 12/96 na_basic.pre
12
The Need for Both Magnitude and Phase
S
1. Complete characterization of
linear networks
21
S 11
S
S 22
4.
12
Time Domain
Characterization
Mag
2. Complex impedance needed to
design matching circuits
Time
High Frequency
Transistor Model
5. Vector Accuracy Enhancement
Base
3. Complex values needed for
device modeling
Error
Collector
Measured
Emitter
Actual
Network Analyzer Basics
DJB 12/96 na_basic.pre
13
Agenda
Why do we test components?
What measurements do we make?
Network analyzer hardware
Error models and calibration
Typical measurements
Advanced topics
Network Analyzer Basics
DJB 12/96 na_basic.pre
14
High-Frequency Device Characterization
Lightwave Analogy
Incident
Transmitted
Reflected
Network Analyzer Basics
DJB 12/96 na_basic.pre
15
Smith Chart Review
.
+jX
90
o
Polar plane
1.0
.8
.6
0
+R
.4
+ 180
-
0
-jX
o
.2
0
o
Rectilinear impedance plane
-90 o
Constant X
Z L = Zo
G=
Smith Chart maps rectilinear impedance
plane onto polar plane
Constant R
0
Z L = 0 (short)
G= 1
±180
Z L=
O
G =1
(open)
0
O
Smith Chart
Network Analyzer Basics
DJB 12/96 na_basic.pre
16
Power Transfer
RS
RL
Load Power (normalized)
For complex impedances, maximum power
transfer occurs when ZL = ZS* (conjugate match)
1.2
1
Zs = R + jX
0.8
0.6
0.4
0.2
0
0
1
2
3
4
5
6
7
8
RL / RS
9
10
ZL = Zs* = R - jX
Maximum power is transferred when RL = RS
Network Analyzer Basics
DJB 12/96 na_basic.pre
17
Transmission Line Review
Low frequencies
I
Wavelength >> wire length
Current (I) travels down wires easily for efficient power transmission
Voltage and current not dependent on position
High frequencies
Wavelength or << wire (transmission line) length
Need transmission-line structures for efficient power transmission
Matching to characteristic impedance (Z0)
important for low reflection
Voltage dependent on position along line
is very
Network Analyzer Basics
DJB 12/96 na_basic.pre
18
Transmission Line Terminated with Zo
Zs = Zo
Zo = characteristic impedance of transmission
line
Zo
V inc
Vrefl = 0! (all the incident power is absorbed in
the load)
For reflection, a transmission line terminated in Zo behaves like an infinitely long transmission
line
Network Analyzer Basics
DJB 12/96 na_basic.pre
19
Transmission Line Terminated with Short, Open
Zs = Zo
V inc
Vrefl
In phase (0 ) for oopen
o
Out of phase (180 ) for short
For reflection, a transmission line terminated in a short or open reflects all power back to
source
Network Analyzer Basics
DJB 12/96 na_basic.pre
20
Transmission Line Terminated with 25 W
Zs = Zo
ZL = 25 W
V inc
Vrefl
Standing wave pattern does not go to zero as
with short or open
Network Analyzer Basics
DJB 12/96 na_basic.pre
21
High-Frequency Device Characterization
Incident
Transmitted
R
B
Reflected
A
TRANSMISSION
REFLECTION
Reflected
Incident
=
SWR
S-Parameters
S11,S22
Reflection
Coefficient
G, r
A
Transmitted
R
Incident
Return
Loss
Impedance,
Admittance
R+jX,
G+jB
=
B
R
Group
Delay
Gain / Loss
S-Parameters
Transmission
S21,S12
Coefficient
T,t
Insertion
Phase
Network Analyzer Basics
DJB 12/96 na_basic.pre
22
Reflection Parameters
G
Coefficient
Reflection
Vreflected
=
=r
Vincident
Return loss = -20 log(r),
r
F
=
0
Z L + ZO
G
Emax
Emin
No reflection
(ZL = Zo)
=
ZL - ZO
r
Voltage Standing Wave Ratio
Emax
VSWR =
Emin
=
1+r
1-r
Full reflection
(ZL = open, short)
1
dB
RL
0 dB
1
VSWR
Network Analyzer Basics
DJB 12/96 na_basic.pre
23
Transmission Parameters
V Incident
V Transmitted
DUT
Transmission Coefficient =
T
=
VTransmitted
V Incident
V
Insertion Loss (dB) = - 20 Log
V
V
Gain (dB) = 20 Log
V
Trans
Trans
=
t
= - 20 log
t
Inc
= 20 log
t
Inc
Network Analyzer Basics
DJB 12/96 na_basic.pre
24
Deviation from Linear Phase
Use electrical delay to remove linear
portion of phase response
RF filter response
Linear electrical length
added
Deviation from linear
phase
Phase 1 /Div
o
o
Phase 45 /Div
(Electrical delay function)
+
Frequency
Low resolution
yields
Frequency
Frequency
High resolution
Network Analyzer Basics
DJB 12/96 na_basic.pre
25
What is group delay?
Frequency
w
tg
Group
Delay
Dw
Phase
Group Delay
to
D
Group Delay (t g)
Average Delay
=
=
w
f
-d
dw
-1
360 o
Frequency
*
d
df
Deviation from constant group delay
indicates distortion
in radians
in radians/sec
Average delay indicates transit time
in degrees
in Hz (
w = 2p f )
Network Analyzer Basics
DJB 12/96 na_basic.pre
26
Phase
Phase
Why measure group delay?
f
f
-d
dw
Group
Delay
Group
Delay
-d
dw
f
f
Same p-p phase ripple can result in different group delay
Network Analyzer Basics
DJB 12/96 na_basic.pre
27
Low-Frequency Network Characterization
H-parameters
V1 = h11I1 + h12V2
V2 = h21I1 + h22V2
Y-parameters
I1 = y11V1 + y12V2
I2 = y21V1 + y22V2
Z-parameters
V1 = z11I1 + z12I2
V2 = z21I1 + z22I2
h11 =
V1
I1
V2=0
(requires short circuit)
h12 =
V1
V2
I1=0
(requires open circuit)
All of these parameters require measuring voltage and current (as a function of
frequency)
Network Analyzer Basics
DJB 12/96 na_basic.pre
28
Limitations of H, Y, Z Parameters
(Why use S-parameters?)
H,Y, Z parameters
Hard to measure total voltage and current at device ports at high
frequencies
Active devices may oscillate or self-destruct with shorts / opens
S-parameters
Relate to familiar measurements
reflection coefficient ...)
Relatively easy to measure
Can cascade S-parameters of multiple
system performance
Analytically convenient
CAD programs
Flow-graph analysis
(gain, loss,
a1
devices
S11
toReflected
predict
Transmitted
b2
DUT
Port 2
Port 1
b1
Transmitted
S 21
Incident
S12
S22
Reflected
a2
Incident
b1 = S11 a 1 + S 12 a 2
b 2 = S21 a 1 + S22 a 2
Can compute H, Y,or Z parameters from S-parameters if desired
Network Analyzer Basics
DJB 12/96 na_basic.pre
29
Measuring S-Parameters
a1
Forward
S
Incident
Z0
S 11
b1
S 21 =
Transmitted
Incident
b1
= a
1
b
a2 = 0
a2 = 0
Reflected
Incident
S 22 =
2
= a
1
a2 = 0
S 12 =
a1 = 0
Z0
Incident
S 12
b2
= a
2
b
a1 = 0
1
= a
2
a1 = 0
b2
Reverse
Reflected
Load
Transmitted
Transmitted
S 22
DUT
b1
Load
DUT
Reflected
Reflected
Incident
S 11 =
b2
Transmitted
21
a2
Incident
Network Analyzer Basics
DJB 12/96 na_basic.pre
30
Measuring Nonlinear Behavior
Most common measurements:
Using a spectrum analyzer + source(s)
harmonics, particularly second and third
intermodulation products resulting from two or more carriers
Using a network analyzer and power sweeps
gain compression
RL 0 dBm
AM to PM conversion
Noise figure
8563A
LPF
LPF
SPECTRUM ANALYZER
ATTEN
10 dB
10 dB / DIV
9 kHz - 26.5 GHz
DUT
CENTER 20.00000 MHz
RB 30 Hz
VB 30 Hz
SPAN 10.00 kHz
ST 20 sec
Network Analyzer Basics
DJB 12/96 na_basic.pre
31
What is the difference between network and spectrum
analyzers?
Hard: getting (accurate) trace
Easy: interpreting results
.
Easy: getting trace
Hard: interpreting results
Power
Amplitude Ratio
8563A
Measures known
signal
Frequency
Network analyzers:
measure components, devices, circuits,
sub-assemblies
contain source and receiver
display ratioed amplitude and phase (frequency or power
sweeps)
SPECTRUM ANALYZER
9 kHz - 26.5 GHz
Measures
unknown signals
Frequency
Spectrum analyzers:
measure signal amplitude characteristics (carrier level,
sidebands, harmonics...)
are receivers only (single channel)
can be used for scalar component test (no phase) with tracking
gen. or ext. source(s)
Network Analyzer Basics
DJB 12/96 na_basic.pre
32
Agenda
Why do we test components?
What measurements do we make?
Network analyzer hardware
Error models and calibration
Typical measurements
Advanced topics
Network Analyzer Basics
DJB 12/96 na_basic.pre
33
Generalized Network Analyzer Block Diagram
Incident
Transmitted
DUT
SOURCE
Reflected
SIGNAL
SEPARATION
INCIDENT
(R)
REFLECTED
(A)
TRANSMITTED
(B)
RECEIVER / DETECTOR
PROCESSOR / DISPLAY
Network Analyzer Basics
DJB 12/96 na_basic.pre
H
34
Source
Supplies stimulus for system
Swept frequency or power
Traditionally NAs used separate source
Open-loop VCOs
Synthesized sweepers
Most HP analyzers sold today have integrated, synthesized sources
Integrated, synthesized sources
Network Analyzer Basics
DJB 12/96 na_basic.pre
35
Signal Separation
Measuring incident signals for ratioing
50 W
6 dB
Splitter
usually resistive
non-directional
broadband
50 W
6 dB
Main signal
Coupled signal
Coupler
directional
low loss
good isolation, directivity
hard to get low freq performance
Network Analyzer Basics
DJB 12/96 na_basic.pre
36
Signal Separation
Separating incident and reflected signals
Coupler
directional
low loss
good isolation, directivity
hard to get low freq performance
Bridge
used to measure reflected signals
only
broadband
higher loss
Detector
Test Port
Network Analyzer Basics
DJB 12/96 na_basic.pre
37
Forward Coupling Factor
Coupling, forward
-20 dBm
.01 mW
Source
Z0
-.046 dBm
.99 mW
0 dBm
1 mW
Example of 20 dB Coupler
P coupling forward
Coupling Factor (dB) = -10 log
P incident
Network Analyzer Basics
DJB 12/96 na_basic.pre
38
Directional Coupler Isolation
Coupling, reverse
-50 dBm
.00001 mW
(Reverse Coupling Factor)
this is an error signal during measurements
Source
Z0
0 dBm 1
mW
-.046 dBm
.99 mW
Example of 20 dB Coupler "turned around"
Isolation Factor (dB) = -10 log
Pcoupled reverse
Pincident
Network Analyzer Basics
DJB 12/96 na_basic.pre
39
Directional Coupler Directivity
Directivity (dB) = 10 log
Directivity =
Pcoupled forward
Pcoupled reverse
Coupling Factor
Isolation
Directivity (dB) = Isolation (dB) - Coupling Factor (dB)
Example of 20 dB Coupler with 50 dB isolation:
Directivity = 50 dB - 20 dB = 30 dB
Network Analyzer Basics
DJB 12/96 na_basic.pre
40
Measuring Coupler Directivity the Easy Way
1.0 (0 dB) (reference)
Good approximation for coupling factors 10
dB
Coupler
Directivity
35 dB
short
Source
.018 (35 dB) (normalized)
Directivity = 35 dB - 0 dB =
35 dB
Source
load
Assume perfect load
Network Analyzer Basics
DJB 12/96 na_basic.pre
41
Interaction of Directivity with the DUT (Without Error Correctio
0
Data Max
Directivity
Device
30
Add in Phase
60
Device
Device
Frequency
Directivity
Return Loss
DUT RL = 40 dB
Data Min
Data = Vector Sum
Directivity
Cancel
Data 0
Network Analyzer Basics
DJB 12/96 na_basic.pre
42
Directional Bridge
50 W
50 W
50 ohm load at test port balances
the
bridge - detector reads zero
Extent of bridge imbalance
indicates impedance
Measuring magnitude and phase of imbalance
gives complex impedance
"Directivity" is difference between maximum
and minimum balance
Detector
50 W
Test Port
Network Analyzer Basics
DJB 12/96 na_basic.pre
43
Incident
Transmitted
DUT
Detector Types
Reflected
SOURCE
SIGNAL
SEPARATION
INCIDENT
(R)
REFLECTED
(A)
TRANSMITTED
(B)
RECEIVER / DETECTOR
Diode
PROCESSOR / DISPLAY
Scalar broadband (no phase
information)
DC
RF
AC
Tuned Receiver
IF = F LO F RF
RF
ADC / DSP
IF Filter
Scalar narrowband
(magnitude only)
Vector
(magnitude and phase)
LO
Network Analyzer Basics
DJB 12/96 na_basic.pre
44
Broadband Diode Detection
Easy to make broadband
Inexpensive compared to tuned receiver
Good for measuring frequency-translating devices
Improve dynamic range by increasing power
Medium sensitivity / dynamic range
10 MHz
26.5 GHz
Network Analyzer Basics
DJB 12/96 na_basic.pre
45
Narrowband Detection - Tuned Receiver
ADC / DSP
Best sensitivity / dynamic range
Provides harmonic / spurious signal rejection
Improve dynamic range by increasing power,
averaging
Trade off noise floor and measurement speed
10 MHz
decreasing IF bandwidth, o
26.5 GHz
Network Analyzer Basics
DJB 12/96 na_basic.pre
46
Front Ends: Mixers Versus Samplers
Sampler-based front end
ADC / DSP
S
ADC / DSP
Mixer-based front end
Harmonic
generator
f
frequency "comb"
It is cheaper and easier to make broadband front
ends using samplers instead of mixers
Network Analyzer Basics
DJB 12/96 na_basic.pre
47
Comparison of Receiver Techniques
Narrowband (tuned- receiver)
detection
Broadband (diode) detection
0 dB
0 dB
-50 dB
-50 dB
-100 dB
-100 dB
-60 dBm Sensitivity
higher noise floor
false responses
< -100 dBm Sensitivity
high dynamic range
harmonic immunity
Dynamic range = maximum receiver power - receiver noise floor
Network Analyzer Basics
DJB 12/96 na_basic.pre
48
Dynamic Range and Accuracy
Dynamic range is very important
for measurement accuracy!
Error (dB, deg)
Error Due to I nterfering Signal
100
+ magn (dB)
- magn (dB)
10
phase (± deg)
phase error
1
magn error
0.1
0.01
0.001
H
0
-5
-10
-15
-20
-25 -30 -35 -40 -45
Interfering signal (dB)
-50
-55
-60
-65
-70
Network Analyzer Basics
DJB 12/96 na_basic.pre
Network Analyzer Basics
DJB 12/96 na_basic.pre
49
Traditional Scalar Analyzer
Traditional scalar system consists of processor/display and
source
Example: HP 8757D
requires external detectors, couplers, bridges, splitters
good for low-cost microwave scalar applications
RF
R A
B
RF
Detector
R A
Detector
Detector
Bridge
Reflection
DUT
DUT
Termination
Transmission
Network Analyzer Basics
DJB 12/96 na_basic.pre
50
B
Modern Scalar Analyzer
Everything necessary for transmission and reflection measurements is
internal!
One-port (reflection) and response (transmission) calibrations
Narrowband and broadband
detectors
Large display
Synthesized source
Transmission/reflection test set
Network Analyzer Basics
DJB 12/96 na_basic.pre
51
Spectrum Analyzer / Tracking Generator
RF in
IF
8563A
SPECTRUM ANALYZER
9 kHz - 26.5 GHz
LO
DUT
Spectrum analyzer
TG out
DUT
f = IF
Tracking generator
Key differences from network analyzer:
one channel -- no ratioed or phase measurements
More expensive than scalar NA
Only error correction available is normalization
Poorer accuracy
Small incremental cost if SA is already needed
Network Analyzer Basics
DJB 12/96 na_basic.pre
52
Modern Vector Analyzer
Features:
Synthesizer
15 MHz to 60 MHz
MUX
RF
Reference
Test
Set
detector
300 kHz
to
3 GHz
Phase
Lock
integrated source
sampler-based front end
tuned receiver
magnitude and phase
vector-error correction
T/R or S-parameter test sets
996 kHz
A
S
B
S
R
S
4 kHz
4 kHz
4 kHz
DU
T
ADC
CPU
Display
Digital
Control
Source
Test
Set
Receiver
Note: modern scalar analyzers like HP 8711/13C look just like vector analyzers,
but they don't display phase
Network Analyzer Basics
DJB 12/96 na_basic.pre
53
T/R Versus S-Parameter Test Sets
Transmission/Reflection Test Set
S-Parameter Test Set
Source
Source
Transfer switch
R
R
B
A
Port 1
Port 2
Fwd
Port 2
Port 1
Fwd
DUT
RF always comes out port 1
port 2 is always receiver
response, one-port cal available
B
A
DUT
Rev
RF comes out port 1 or port 2
forward and reverse measurements
two-port calibration possible
Network Analyzer Basics
DJB 12/96 na_basic.pre
54
Three Versus Four-Channel Analyzers
Source
Source
Transfer switch
Transfer switch
R1
R
A
A
B
B
R2
Port 1
Port 2
Port 1
Port 2
3 samplers
4 samplers
cheaper
TRL*, LRM* cal only
includes:
HP 8753D
HP 8720D (std.)
more expensive
true TRL, LRM cal
includes
HP 8720D (opt. 400)
HP 8510C
Network Analyzer Basics
DJB 12/96 na_basic.pre
55
Processor / Display
Incident
Transmitted
DUT
SOURCE
Reflected
50 MHz-20GHz
H
SIGNAL
SEPARATION
NETWORK ANALYZER
ACTIVE CHANNEL
CH2 START 775.000 000 M Hz
CH1 START 775.000 000 M Hz
INCIDENT
(R)
REFLECTED
(A)
TRANSMITTED
(B)
ENTRY
STOP 925.000 000 M Hz
STOP 925.000 000 M Hz
Hld
RESPONSE
PASS
2
Cor
PRm
880.435 000 M Hz
1
PASS
STIMULUS
R CHANNEL
INSTRUMENT STATE
Hld
RECEIVER / DETECTOR
1
1
Cor
PRm
T
Duplexer Test - Tx-Ant and Ant-Rx
839.470 000 M Hz
CH2
CH1
PROCESSOR / DISPLAY
S 12
S 21
HP-IB STATUS
log M AG
log M AG
10 dB/
10 dB/
REF 0 dB
REF 0 dB
1_ -1.2468 dB
1_ -1.9248 dB
PORT 1
CH2 START 775.000 000 MHz
CH1 START 775.000 000 MHz
STOP 925.000 000 MHz
STOP 925.000 000 MHz
Hld
markers
limit lines
pass/fail indicators
linear/log formats
grid/polar/Smith charts
PORT 2
PASS
2
Cor
PRm
880.435 000 MHz
1
PASS
Hld
1
1
Cor
PRm
Duplexer Test - Tx-Ant and Ant-Rx
839.470 000 MHz
CH2
CH1
S12
S21
log MAG
log MAG
10 dB/
10 dB/
REF 0 dB
REF 0 dB
1_ -1.2468 dB
1_ -1.9248 dB
Network Analyzer Basics
DJB 12/96 na_basic.pre
56
R L
S
Internal Measurement Automation
Simple: recall states
More powerful:
Test sequencing
available on HP 8753 / 8720 families
keystroke recording
some advanced functions
IBASIC
available on HP 8711 family
sophisticated programs
custom user interfaces
ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789 + - / * = < > ( ) & "" " , . / ? ; : ' [ ]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
ASSIGN @Hp8714 TO 800
OUTPUT @Hp8714;"SYST:PRES; *WAI"
OUTPUT @Hp8714;"ABOR;:INIT1:CONT OFF;*WAI"
OUTPUT @Hp8714;"DISP:ANN:FREQ1:MODE SSTOP"
OUTPUT @Hp8714;"DISP:ANN:FREQ1:MODE CSPAN"
OUTPUT @Hp8714;"SENS1:FREQ:CENT 175000000 HZ;*WAI"
OUTPUT @Hp8714;"ABOR;:INIT1:CONT OFF;:INIT1;*WAI"
OUTPUT @Hp8714;"DISP:WIND1:TRAC:Y:AUTO ONCE"
OUTPUT @Hp8714;"CALC1:MARK1 ON"
OUTPUT @Hp8714;"CALC1:MARK:FUNC BWID"
OUTPUT @Hp8714;"SENS2:STAT ON; *WAI"
OUTPUT @Hp8714;"SENS2:FUNC 'XFR:POW:RAT 1,0';DET NBAN; *WAI"
OUTPUT @Hp8714;"ABOR;:INIT1:CONT OFF;:INIT1;*WAI"
OUTPUT @Hp8714;"DISP:WIND2:TRAC:Y:AUTO ONCE"
OUTPUT @Hp8714;"ABOR;:INIT1:CONT ON;*WAI"
END
Network Analyzer Basics
DJB 12/96 na_basic.pre
57
HP Families of HF Vector Analyzers
Microwave
HP 8720D family
40 GHz
economical
fast, small
test mixers, high- power
amps
S-parameter
RF
HP 8712/14C
3 GHz
low cost, fast
narrowband and broadband
detection
T/R test set only
HP 8510C family
110 GHz in coax
pulse systems
antenna meas.
Tx/Rx module test
highest accuracy
4 S-parameter display
HP 8753D family
6 GHz
52C: T/R test set
53D: S-parameter
highest RF accuracy
Offset and harmonic RF sweeps
Network Analyzer Basics
DJB 12/96 na_basic.pre
58
HP Families of LF Vector Analyzers
LF
HP E5100A/B
HP 8751A
300 MHz
economical
fast, small
test resonators, filters
parameter analysis
500 MHz
fast list sweep
impedance matching
4 trace display
Combination
HP 4195A
500 MHz
network/spectrum/
impedance (option)
DC output
user-defined functions
HP 4396A
1.8 GHz
network/spectrum/ impedance
(option)
fast, highest accuracy
time-gated spectrum (option)
Network Analyzer Basics
DJB 12/96 na_basic.pre
59
Agenda
Why do we test components?
What measurements do we make?
Network analyzer hardware
Error models and calibration
Typical measurements
Advanced topics
Why do we even need error-correction
and calibration?
It is impossible to make perfect hardware
It would be extremely expensive to make hardware good enough to not require any
error correction
Network Analyzer Basics
DJB 12/96 na_basic.pre
60
Measurement Error Modeling
Systematic errors
due to imperfections in the analyzer and test setup
are assumed to be time invariant (predictable)
can be characterized (during calibration process) and mathematically
removed during measurements
Random errors
vary with time in random fashion (unpredictable)
cannot be removed by calibration
main contributors:
instrument noise (source
phase noise, IF noise floor, etc.)
switch repeatability
connector repeatability
Drift errors
are due to instrument or test-system performance
calibration has been done
are primarily caused by temperature variation
can be removed by further calibration(s)
Errors:
SYSTEMATIC
Measured
Data
changing after a
Unknown
Device
RANDOM
DRIFT
Network Analyzer Basics
DJB 12/96 na_basic.pre
61
Systematic Measurement Errors
R
Directivity
A
B
Crosstalk
DUT
Frequency response
reflection tracking (A/R)
transmission tracking (B/R)
Source
Mismatch
Load
Mismatch
Six forward and six reverse error terms yields 12 error terms for
two-port devices
Network Analyzer Basics
DJB 12/96 na_basic.pre
62
Types of Error Correction
Two main types of error correction:
response (normalization)
simple to perform
only corrects for tracking errors
stores reference trace in memory,
divided by memory
vector
requires more standards
requires an analyzer that can measure phase
accounts for all major sources of systematic error
then does data
thru
SHORT
thru
OPEN
S11
LOAD
A
S11
M
Network Analyzer Basics
DJB 12/96 na_basic.pre
63
What is Vector-Error Correction?
Process of characterizing systematic error terms
measure known standards
remove effects from subsequent measurements.
1-port calibration (reflection measurements)
only 3 systematic error terms measured
directivity, source match, and reflection tracking
Full 2-port calibration
(reflection and transmission measurements)
12 systematic error terms measured
usually requires 12 measurements on four known standards (SOLT)
Some standards can be measured multiple times
THRU is usually measured four times)
Standards defined in cal kit definition file
network analyzer contains standard cal kit definitions
CAL KIT DEFINITION MUST MATCH ACTUAL CAL KIT USED!
(e.g.,
Network Analyzer Basics
DJB 12/96 na_basic.pre
64
Reflection: One-Port Model
Ideal
RF in
S11M
S11A
Error Adapter
If you know the systematic error terms,
you can solve
for the actual S-parameter
Assumes good termination at port two
if testing two-port
devices
If port 2 is connected to the network analyzer and DUT reverse
isolation is low (e.g., filter passband):
assumption of good termination is not valid
two-port error correction yields better results
ED = Directivity
1
RF in
ERT = Reflection tracking
S11M
ES
ED
S11A
ES = Source Match
S11M = Measured
ERT
S11M = ED + ERT
S11A
1 - ES S11A
S11A = Actual
To solve for S11A, we have 3 equations
and 3 unknowns
Network Analyzer Basics
DJB 12/96 na_basic.pre
65
Before and After One-Port Calibration
0
2.0
Data Before
Error Correction
Return Loss (dB)
20
40
1.01
Data After
Error Correction
60
6000
VSWR
1.1
1.001
12000
Network Analyzer Basics
DJB 12/96 na_basic.pre
66
Adapter Considerations
reflection from adapter
desired signal
leakage signal
r
Coupler directivity = 40 dB
r
total = adapter +
Adapter
DUT
Termination
r
DUT
DUT has SMA (f) connectors
APC-7 calibration done here
Worst-case
System
Directivity
28 dB
17 dB
14 dB
Adapting from APC-7 to SMA (m)
APC-7 to SMA (m)
SWR:1.06
APC-7 to N (f) + N (m) to SMA (m)
SWR:1.05
SWR:1.25
APC-7 to N (m) + N (f) to SMA (f) + SMA (m) to (m)
SWR:1.05
SWR:1.25
SWR:1.15
Network Analyzer Basics
DJB 12/96 na_basic.pre
67
Two-Port Error Correction
Port 1
EX
Port 2
S21
a1
ED
ES
S11
A
b1
ERT
ED = Directivity
ES = Source Match
ERT = Reflection Tracking
ETT
A
S22
A
b2
Forward model
a2
EL
S12
A
EL = Load Match
ETT = Transmission Tracking
EX = Isolation
Notice that each actual S-parameter is a function of
all four measured S-parameters
Analyzer must make forward and reverse sweep to
update any one S-parameter
Luckily, you don't need to know these equations to
use network analyzers!!!
Network Analyzer Basics
DJB 12/96 na_basic.pre
68
Crosstalk (Isolation)
Crosstalk definition: signal leakage between ports
Can be a problem with:
High-isolation devices (e.g., switch in open position)
High-dynamic range devices (some filter stopbands)
Isolation calibration
Adds noise to error model (measuring noise floor of system)
Only perform if really needed (use averaging)
if crosstalk is independent of DUT match, use two terminations
if dependent on DUT match, use DUT with termination on output
DUT
Isolation cal when crosstalk is dependent on match of
DUT
LOAD
DUT
DUT
LOAD
Network Analyzer Basics
DJB 12/96 na_basic.pre
69
Errors and Calibration Standards
UNCORRECTED
RESPONSE
1-PORT
DU
T
FULL 2-PORT
SHORT
SHORT
SHORT
OPEN
OPEN
OPEN
LOAD
LOAD
LOAD
thru
Convenient
Generally not accurate
No errors removed
DUT
Easy to perform
Use when highest
accuracy is not required
Removes frequency
response error
DUT
Other errors:
Random (Noise, Repeatability)
Drift
For reflection measurements
Need good termination for high
accuracy with two-port devices
Removes these errors:
Directivity
Source match
Reflection tracking
thru
DUT
Highest accuracy
Removes these errors:
Directivity
Source, load match
Reflection tracking
Transmission tracking
Crosstalk
Network Analyzer Basics
DJB 12/96 na_basic.pre
70
ECal: Electronic Calibration
85060 series)
(HP
H
Impedance States
achieved by shunting transmission line with PIN-diode switches in various
combinations
13 reflective states, from low to high reflection
two thru states plus one isolation state
programmable and highly repeatable
characterized by TRL-calibrated network analyzer
Calibration
four known impedance states presented at each frequency (providing
redundant information)
uses least-squares fit to calculate error terms
yields accuracy between SOLT and TRL
HP 85062A
Electronic Calibration Module
0.045 - 2 GHz
12
7
1
5
Example distribution of impedance
states for reflection calibration at
one frequency
Network Analyzer Basics
DJB 12/96 na_basic.pre
71
Calibration Summary
Reflection
T/R
(one-port)
Test Set (cal type)
S-parameter
SHORT
(two-port)
Reflection tracking
Directivity
Source match
Load match
OPEN
LOAD
T/R
Transmission
error can be corrected
(response,
Test Set (cal type)
S-parameter
(two-port)
isolation)
Transmission Tracking
Crosstalk
Source match
Load match
error cannot be corrected
*
HP 8711C enhanced response cal can correct for source
match during transmission measurements
(
*)
Network Analyzer Basics
DJB 12/96 na_basic.pre
72
Reflection Example Using a One-Port Cal
Analyzer port 2 match:
18 dB (.126)
DUT
16 dB RL (.158)
1 dB loss (.891)
Measurement uncertainty:
-20 * log (.158 + .100)
= 11.4 dB (-4.6dB)
-20 * log (.158 - .100)
= 24.7 dB (+8.7 dB)
.158
(.891)(.126)(.891) = .100
Low-loss bidirectional devices generally require 2-port calibration for low
measurement uncertainty
Network Analyzer Basics
DJB 12/96 na_basic.pre
73
Transmission Example Using Response Cal
RL = 18 dB (.126)
RL = 14 dB (.200)
Thru calibration (normalization) builds error into measurement due to source and load match
interaction
Calibration Uncertainty
= (1 ± r r )
S L
= (1 ± (.200)(.126)
= ± 0.22 dB
Network Analyzer Basics
DJB 12/96 na_basic.pre
74
Transmission Example (continued)
Source match = 14
dB (.200)
DUT
1 dB loss (.891)
16 dB RL (.158)
Load match = 18
dB (.126)
1
(.126)(.158) = .020
(.126)(.891)(.200)(.891) = .020
(.158)(.200) = .032
Total measurement uncertainty:
+0.60 + 0.22 = +0 .82 dB
-0.65 - 0.22 = - 0.87 dB
Measurement uncertainty
= 1 ± (.020+.020+.032)
= 1 ± .072
= + 0.60 dB
- 0.65 dB
Network Analyzer Basics
DJB 12/96 na_basic.pre
75
Measuring Amplifiers with a Response Cal
Source match = 14
dB (.200)
DUT
16 dB RL (.158)
Load match = 18
dB (.126)
1
(.126)(.158) = .020
(.158)(.200) = .032
Total measurement uncertainty:
+0.44 + 0.22 = + 0.66 dB
-0.46 - 0.22 = - 0.68 dB
Measurement uncertainty
= 1 ± (.020+.032)
= 1 ± .052
= + 0.44 dB
- 0.46 dB
Network Analyzer Basics
DJB 12/96 na_basic.pre
76
Transmission Measurements using the Enhanced
Response Calibration
Calibration Uncertainty
= (1 ± r r )
L
S
Effective source match = 35 dB!
Source match = 35
dB (.0178)
DUT
1 dB loss (.891)
16 dB RL (.158)
= (1 ± (.0178)(.126)
= ± .02 dB
Load match =
18 dB (.126)
1
Measurement
(.126)(.158) = .020
uncertainty
=1±
(.126)(.891)(.0178)(.891) = .0018
(.020+.0018+.0028)
= 1 ± .0246
(.158)(.0178) = .0028
= + 0.211 dB
- 0.216
Total measurement uncertainty:
0.22 + .02 = ± 0.24 dB
Network Analyzer Basics
DJB 12/96 na_basic.pre
77
Calculating Measurement Uncertainty After a Two-Port
Calibration
DUT
1 dB loss (.891)
16 dB RL (.158)
Corrected error terms:
(8753D 1.3-3 GHz Type-N)
Directivity
= 47 dB
Source match =
36 dB
Load match = 47 dB
Refl. tracking =
.019 dB
Trans. tracking = .026 dB
Isolation
= 100 dB
Reflection uncertainty
2
S11m = S11 a ± (E D+ S11 a E S+ S21 S12
E
)
a
a L+ S11 E
a RT
= .158 ± (.0045+.158 2*.0158 + .891 2*.0045 + .158*.0022)
= .158 ± .0088 = 16 dB +0.53 dB, -0.44 dB
Transmission uncertainty
S21m = S21a ± (E I + S11 aE S+ S22 aE L+ S21 aS12 E
)
a E
a TT
S +
L S21 E
2
= .891 ± (10 -6+ .158*.0158 + .158*.0045 + .891 *.0158*.0045 + .891*.003)
= .891 ± .0059 = 1 dB ± 0.06 dB
Network Analyzer Basics
DJB 12/96 na_basic.pre
78
Response versus Two-Port Calibration
Measuring filter insertion loss
CH1 S21&M log MAG
CH2 MEM log MAG
Cor
1 dB/
1 dB/
REF 0 dB
REF 0 dB
After 2-port calibration
After response calibration
Uncorrected
Cor
x2 1START 2 000.000 MHz
STOP 6 000.000 MHz
2
Network Analyzer Basics
DJB 12/96 na_basic.pre
79
Thru-Reflect-Line (TRL) Calibration
We know about Short-Open-Load-Thru (SOLT) calibration...
What is TRL?
A two-port calibration technique
Good for noncoaxial environments (waveguide, fixtures, wafer probing)
Uses the same 12-term error model as the more common SOLT cal
Uses practical calibration standards that
fabricated and characterized
Two variations: TRL (requires 4 samplers)
three samplers needed)
Other variations: Line-Reflect-Match (LRM),
Thru-Reflect-Match (TRM), plus many others
are easily
and TRL* (only
Network Analyzer Basics
DJB 12/96 na_basic.pre
H
80
Why Are Four Samplers Better Than Three?
TRL
TRL*
HP 8720D Opt. 400 adds fourth sampler,
allowing full TRL calibration
TRL*
assumes the source and load match of a test port are equal
forward and reverse measurements)
this is only a fair assumption for a three-sampler network analyzer
TRL* requires ten measurements to quantify eight unknowns
TRL
(port symmetry between
Four samplers are necessary for all the measurements required for a full TRL cal (fourteen measurements to quantify ten
unknowns)
TRL and TRL* use identical calibration standards
In noncoaxial applications:
TRL achieves better source match and load match correction than TRL*
What about coaxial applications?
TRL* and SOLT calibration have about the same accuracy
Coaxial TRL is usually more accurate than SOLT but not commonly used
Network Analyzer Basics
DJB 12/96 na_basic.pre
81
Calibrating Non-Insertable Devices
When doing a thru cal, normally test ports mate directly
cables can be connected directly without an adapter
result is a zero-length thru
What is an insertable device?
has same type of connector, but different sex on each port
has same type of sexless connector on each port (e.g. APC-7)
What is a non-insertable device?
one that cannot be inserted in place of a zero-length thru
has same connectors on each port (type and sex)
has different type of connector on each port
one port, coaxial on the other)
What calibration choices do I have for non-insertable devices?
Use an uncharacterized thru adapter
Use a characterized thru adapter (modify cal-kit definition)
Swap equal adapters
Adapter removal
(e.g., waveguide
on
DUT
Network Analyzer Basics
DJB 12/96 na_basic.pre
82
Swap Equal Adapters Method
Port 1
Port 1
Adapter
A
Port 1
Port 1
Port 2
DUT
DUT
Accuracy depends on how well the adapters are matched loss, electrical length, match and impedance should all be
equal
1. Transmission cal using adapter A.
Port 2
Adapter
B
Port 2
2. Reflection cal using adapter B.
Adapter
B
Port 2
3. Measure DUT using adapter B.
Network Analyzer Basics
DJB 12/96 na_basic.pre
83
Adapter Removal Calibration
In firmware of HP 8510 family
Can be accomplished with E-Cal (HP 85060) and HP 8753/8720 families
Uses adapter with same connectors as DUT
Adapter's electrical length must be specified within 1/4 wavelength
adapters supplied with HP type-N, 3.5mm, and 2.4mm cal kits are already defined
for other adapters, measure electrical length and modify cal-kit definition
Calibration is very accurate and traceable
See Product Note 8510-13 for more details
Port 1
DUT
Port 2
Port 1
Cal
Adapter
Adapter Port 2
B
1. Perform 2-port cal with adapter on port 2.
Save in cal set 1.
Adapter Port 2
B
2. Perform 2-port cal with adapter on port 1.
Save in cal set 2.
Cal Set 1
Port 1
Cal
Adapter
Cal Set 2
[CAL] [MORE] [MODIFY CAL SET]
[ADAPTER REMOVAL]
Port 1
DUT
Adapter Port 2
B
3. Use ADAPTER REMOVAL
to generate new cal set.
4. Measure DUT without cal adapter.
Network Analyzer Basics
DJB 12/96 na_basic.pre
84
Agenda
Why do we test components?
What measurements do we make?
Network analyzer hardware
Error models and calibration
Typical measurements
Advanced topics
Network Analyzer Basics
DJB 12/96 na_basic.pre
85
Frequency Sweep - Filter Test
CH1 S21
log MAG
10 dB/
CH1 S11
REF 0 dB
log MAG
REF 0 dB
5 dB/
Cor
Stopband
rejection
69.1 dB
CENTER 200.000 MHz
START .300 000 MHz
SPAN 50.000 MHz
STOP 400.000 000 MHz
CH1 S 21
log MAG
Cor
1 dB/
REF 0 dB
Return loss
1
m1:
4.000 000 GHz -0.16 dB
m2-ref: 2.145 234 GHz 0.00 dB
ref
Insertion loss
2
Cor
x2 1
START 2 000.000 MHz
STOP 6 000.000 MHz
Network Analyzer Basics
DJB 12/96 na_basic.pre
2
86
Output Power (dBm)
Power Sweep - Compression
Saturated
output power
Compression
region
Linear region
(slope = small-signal gain)
Input Power (dBm)
Network Analyzer Basics
DJB 12/96 na_basic.pre
87
Power Sweep -Gain Compression
CH1 S21
1og MAG
1 dB/ REF 32 dB
30.991 dB
12.3 dBm
C2
1 dB compression: input power
resulting in 1 dB drop in gain
Ratioed measurement
Output power available (non-ratioed
measurement)
0
0
IF BW 3 kHz
START -10 dBm
CW 902.7 MHz
SWP 420 msec
STOP 15 dBm
Network Analyzer Basics
DJB 12/96 na_basic.pre
88
Power Sweep - AM to PM Conversion
1:Transmission
Log Mag
2:Transmission /M Phase
1.0 dB/
5.0 deg/
Ref 21.50 dB
Ref -115.7 deg
Ch1:Mkr1
-4.50 dBm 20.48 dB
Ch2:Mkr2
1.00 dB
0.86 deg
Use transmission setup with a power
sweep
Display phase of S21
AM - PM = 0.86 deg/dB
2
1
2
1
1
Start -10.00 dBm
Start -10.00 dBm
CW 900.000 MHz
CW 900.000 MHz
Stop 0.00 dBm
Stop 0.00 dBm
Network Analyzer Basics
DJB 12/96 na_basic.pre
89
Agenda
Why do we test components?
What measurements do we make?
Network analyzer hardware
Error models and calibration
Typical measurements
Advanced topics
Time domain
Frequency-translating devices
High-power amplifiers
Multiport devices
In-fixture measurements
Crystal Resonators
Balanced-Cables
Network Analyzer Basics
DJB 12/96 na_basic.pre
90
Time-Domain Reflectometry (TDR)
Analyze impedance versus time
Differentiate inductive and capacitive transitions
High-speed oscilloscope:
yields fast update rate
200 mV step typical
Network analyzer:
broadband frequency sweep (often requires microwave VNA)
inverse FFT to compute time-domain
resolution inversely proportional to frequency span
non-Zo termination
inductive transition
Zo
capacitive transition
Network Analyzer Basics
DJB 12/96 na_basic.pre
91
Time-Domain Gating
TDR and gating can remove undesired reflections
(a form of error
correction)
Only useful for broadband devices (a load or thru for example)
Define gate to only include DUT
CH1 S11&M log MAG
5 dB/
REF 0 dB
Use two-port calibration
PRm
Cor
CH1 MEM Re
PRm
Cor
RISE TIME
29.994 ps
8.992 mm
2
20 mU/ REF 0 U
Gate
1: -45.113 dB 0.947 GHz
1: 48.729 mU 638 ps
2: -15.78 dB 6.000 GHz
2: 24.961 mU 668 ps
1
3: -10.891 mU 721 ps
2
3
Thru in time domain
CH1 START 0 s
STOP 1.5 ns
1
Thru in frequency domain, with and without
gating
START .050 000 000 GHz
STOP 20.050 000 000 GHz
Network Analyzer Basics
DJB 12/96 na_basic.pre
92
Time-Domain Transmission
RF Input
RF Output
CH1 S21 log MAG
15 dB/ REF 0 dB
Main Wave
Surface
Wave
Leakage
Triple Travel
Cor
CH1 S21
log MAG
10 dB/ REF 0 dB
RF
Leakage
Triple
Travel
Cor
Gate off
Gate on
START -1 us
STOP 6 us
Network Analyzer Basics
DJB 12/96 na_basic.pre
93
Frequency-Translating Devices
Medium-dynamic range measurements (35 dB)
R IN
2
1
ACTIVE CHANNEL
ENTRY
RESPONSE
HP 8753D
start:
MHz
stop:
MHz
900
650
FIXED LO: 1 GHz
LO POWER: 13 dBm
start:
MHz
stop:
MHz
R CHANNEL
INSTRUMENT STATE
STIMULUS
Ref In
100
Reference Mixer
R
HP-IB STATUS
350
H
30 KHz-3GHz
NETWORK ANALYZER
PORT 1
L
T
S
Ref Out
PROBE POW ER
FUSED
8753D
PORT 2
RF
10 dB
IF
LO
10 dB
CH1 CONV MEAS
log MAG
10 dB/
REF 10 dB
10 dB
LO
DUT
3 dB
Signal Generator
START 640.000 000 MHz
STOP 660.000 000 MHz
High-dynamic range measurements
Network Analyzer Basics
DJB 12/96 na_basic.pre
94
Lowpass
Filter
High-Power Amplifiers
Preamp
ACTIVE CHANNEL
ENTRY
RESPONSE
Ref In
STIMULUS
INSTRUMENT STATE
R CHANNEL
R
HP-IB
STATUS
H
PROBE POW ER
FUSED
30 KHz-3GHz
NETWORK ANALYZER
8753D
PORT 1
PORT 2
HP 8753D
L
T
S
Source
Preamp
AUT
DUT
R
B
A
AUT
+43 dBm max input (20 watts!)
HP 8720D Option 085
HP 85118A High-Power Amplifier Test System
Network Analyzer Basics
DJB 12/96 na_basic.pre
95
Multiport Device Test
Port 1
Port 1
Port 2
Port 3
Port 2
Note: unused ports are
terminated
CH1 S21
CH2 S12
PR
m
Co
r
log
log
MAG
MAG
10
10
dB/
dB/
REF 0
REF
dB 0
dB
1_ -1.9248
1_ -1.2468
dB
839.470 000dB
MHz
1
Hld
Directional
Coupler
Multiport test sets:
improve throughput by reducing the number of connections to
DUTs with more than 2 ports
allow simultaneous viewing of two paths
(good for tuning
duplexers)
include mechanical or solid-state switches,
50 or 75 ohms
degrade raw performance so calibration is a must (use two-port cals
whenever possible)
Duplexer Test - Tx-Ant and Ant-Rx
1
Test Set
PAS
S
1
880.435 000
MHz
PR
m
Co
r
2
Hld
PAS
S
CH1 START 775.000 000
CH2
MHz START 775.000 000
MHz
STOP
STOP 925.000
925.000 000
000 MHz
MHz
Network Analyzer Basics
DJB 12/96 na_basic.pre
96
In-Fixture Measurements
Measurement problem: coaxial calibration plane is not the same as the in-fixture
measurement plane
Measurement
Plane
Calibration
Plane
Fixture
ED
ES
ET
Error correction with coaxial calibration
DUT
Loss
Phase shift
Mismatch
Network Analyzer Basics
DJB 12/96 na_basic.pre
97
Characterizing Crystal Resonators/Filters
Ch1
Z: R
phase
40 / REF 0
1: 15.621 U
31.998 984 925 MHz
Min
Cor
1
START 31.995 MHz
SEG START
DUT
STOP
POINTS
STOP 32.058 MHz
POWER
IFBW
1 31.995 MHz
32.008 MHz
200
0 dBm
200Hz
> 2 32.052 MHz
32.058 MHz
200
0 dBm
200Hz
END
Example of crystal resonator measurement
HP E5100A Network Analyzer
Network Analyzer Basics
DJB 12/96 na_basic.pre
98
RF Balanced-Cable Measurements
280.00 ohm
260.00 ohm
240.00 ohm
220.00 ohm
200.00 ohm
180.00 ohm
160.00 ohm
140.00 ohm
120.00 ohm
100.00 ohm
80.00 ohm
0.01 MHz
0.10 MHz
1.00 MHz
10.00 MHz
100.00 MHz
1000.00 MHz
Example of characteristic impedance (Zc) measurement from 10 kHz
to 500 MHz
0.00 dB
-20.00 dB
-40.00 dB
-60.00 dB
-80.00 dB
-100.00 dB
-120.00 dB
HP 4380S RF Balanced-Cable Test System
-140.00 dB
0.01 MHz
0.10 MHz
1.00 MHz
10.00 MHz
100.00 MHz
1000.00 MHz
Example of near-end crosstalk (NEXT) measurement
Network Analyzer Basics
DJB 12/96 na_basic.pre
99
Challenge Quiz
1. Can filters cause distortion in communications systems?
A. Yes, due to impairment of phase and magnitude response
B. Yes, due to nonlinear components such as ferrite inductors
C. No, only active devices can cause distortion
D. No, filters only cause linear phase shifts
E. Both A and B above
2. Which statement about transmission lines is false?
A. Useful for efficient transmission of RF power
B. Requires termination in characteristic impedance for low VSWR
C. Voltage is independent of position along line
D. Used when wavelength of signal is small compared to length of line
E. Can be realized in a variety of forms such as coaxial, waveguide, microstrip
3. Which statement about narrowband detection is false?
A. Is only available in vector network analyzers
B. Provides much greater dynamic range than diode detection
C. Uses variable-bandwidth IF filters to set analyzer noise floor
D. Provides rejection of harmonic and spurious signals
E. Uses mixers or samplers as downconverters
Network Analyzer Basics
DJB 12/96 na_basic.pre
100
Challenge Quiz (continued)
4. Maximum dynamic range with narrowband detection is defined as:
A. Maximum receiver input power minus the stopband of the device under test
B. Maximum receiver input power minus the receiver's noise floor
C. Detector 1-dB-compression point minus the harmonic level of the source
D. Receiver damage level plus the maximum source output power
E. Maximum source output power minus the receiver's noise floor
5. With a T/R analyzer, the following error terms can be corrected:
A. Source match, load match, transmission tracking
B. Load match, reflection tracking, transmission tracking
C. Source match, reflection tracking, transmission tracking
D. Directivity, source match, load match
E. Directivity, reflection tracking, load match
6. Calibration can remove which of the following types of measurement error?
A. Systematic and drift
B. Systematic and random
C. Random and drift
D. Repeatability and systematic
E. Repeatability and drift
Network Analyzer Basics
DJB 12/96 na_basic.pre
101
Challenge Quiz (continued)
7. Which statement about TRL calibration is false?
A. Is a type of two-port error correction
B. Uses easily fabricated and characterized standards
C. Most commonly used in noncoaxial environments
D. Is not available on the HP 8720D family of microwave network analyzers
E. Has a special version for three-sampler network analyzers
8. For which component is it hardest to get accurate transmission and
reflection measurements when using an
8711B scalar network analyzer?
A. Amplifiers because output power causes receiver compression
B. Cables because load match cannot be corrected
C. Filter stopbands because of lack of dynamic range
D. Mixers because of lack of broadband detectors
E. Attenuators because source match cannot be corrected
9. Power sweeps are good for which measurements?
A. Gain compression
B. AM to PM conversion
C. Saturated output power
D. Power linearity
E. All of the above
Network Analyzer Basics
DJB 12/96 na_basic.pre
102