Lecture 7

Lecture 7
Microwave Network
Analysis
A. Nassiri -ANL
Massachusetts Institute of Technology

The RF Cavities
US Particle Accelerator Schooland Components
Boston forUniversity
Accelerators USPAS 2010
June 16th, 2006
S-Parameter Measurement Technique
VVM: The vector voltmeter measures the magnitude of a

reference and test voltage and the difference in phase between the
voltages. Because it can measure phase, it allows us to directly
measure the S-parameters of a circuit
Unfortunately, the use of the directional couplers and test cables

connecting the measuring system to the vector voltmeter
introduces unknown attenuation and phase shift into the
measurements. These can be compensated for by making
additional “calibration” measurements.

The RF Cavities
June 16th, 2006 2
Reflection measurements: S11 or S22
(HP 8657B) (HP 8508A)

Signal Gen VVM
A B
N-BNC N-BNC
N-BNC
BNC cable
N-BNC N-BNC
N-BNC 20 dB 20 dB
 DUT
BNC cable
(HP 778D Dual Directional Coupler) Matched load

The RF Cavities
June 16th, 2006 3
From the setup, it is seen that the voltage at channel A of the VVM (AD)
is proportional to the amplitude of the voltage wave entering the device
under test (DUT) (aD1). Similarly, the voltage at channel B (BD) is
proportional to the amplitude of the voltage wave reflected from DUT
(bD1). Thus we can write
A D = K A a1D
B D = K B b1D
Where KA and KB are constants that depend on the connecting cables.
Since aD2 is zero because of the matched load at port 2, S11 is given by
b1D B D K B
S 11 = D = D
a1 A KA

The RF Cavities
June 16th, 2006 4
To find KA and KB it is necessary to make a second measurement with a

known DUT. This is called a “calibration” measurement. If the DUT is
removed and replaced by a short circuit, the voltage at channel A (As)
and channel B(Bs) are given by
A S = K A a1S
B S = K B b1S
Where as1 is the amplitude of the voltage wave entering the short and bs1
is the amplitude of the voltage wave reflected from the short. However,
for a short circuit the ratio of these amplitudes is –1 (reflection
coefficient of a short). Thus
S S
b1 B K B
S
= S = −1
a1 A K A
The RF Cavities
June 16th, 2006 5
BD 
 
KB BS  AD 
=− S S 11 = −  S 
KA A B 
 
 AS 
 
Note: since VVM displays quantities in terms of magnitude and
phase we can rewrite S11 as
BD 
  = Γ D ∠φD
 AD 
ΓD
S 11 = S ∠ φD − φS − π
Γ
( ) 
BS


  = Γ S ∠φS
 AS 
 
The RF Cavities
June 16th, 2006 6
Transmission measurements: S12 or S21
Generator Channel A Channel B

Matched load Matched load
N-BNC
 DUT 
BNC cable
Matched load
The DUT is connected directly between two directional couplers.

Voltage at A of the VVM is proportional to the voltage wave incident on
the DUT while the voltage at B of the VVM is proportional to voltage
wave transmitted through the DUT.

The RF Cavities
June 16th, 2006 7
A D = K A a1D b 2D B D K B
S 21 = D = D
B D
= K B b2 D ⇒ a1 A KA
To find out the constants a calibration measurement must be made.
Remove the DUT and connect both directional couplers directly
together. The Known DUT in this case is just a zero-length guide with a
transmission coefficient of unity. The measured voltages are:
E E
A E = K A a1E b2 B K B
where E
= E =1
B E
= K B b2 E a1 A KA
E
KB B
∴ = E
KA A
The RF Cavities
June 16th, 2006 8

BD 
 
 AD 
S 21 = −  E 
TD
S 21 = E ∠ θD − θE
T
( )
B 
 
 AE 
 
where
BD  BE 
  = T D ∠θD   = T E ∠θE
 AD   AE 
   

The RF Cavities
June 16th, 2006 9
Scattering Parameters
 Scattering Parameters (S-Parameters) plays a major role is network

analysis
 This importance is derived from the fact that practical system

characterizations can no longer be accomplished through simple open- or
short-circuit measurements, as is customarily in low-frequency
applications.
 In the case of a short circuit with a wire; the wire itself possesses an
inductance that can be of substantial magnitude at high frequency.
 Also open circuit leads to capacitive loading at the terminal.

The RF Cavities
June 16th, 2006 10
 In either case, the open/short-circuit conditions needed to determine

Z-, Y-, h-, and ABCD-parameters can no longer be guaranteed.
 Moreover, when dealing with wave propagation phenomena, it is not

desirable to introduce a reflection coefficient whose magnitude is unity.
 For instance, the terminal discontinuity will cause undesirable voltage

and/or current wave reflections, leading to oscillation that can result in
the destruction of the device.
 With S-parameters, one has proper tool to characterize the two-port

network description of practically all RF devices without harm to DUT.

The RF Cavities
June 16th, 2006 11
Definition of Scattering Parameters
 S-parameters are power wave descriptors that permit us to define the

input-output relations of a network in terms of incident and reflected
power waves.
a1 a2
 

[S] 
b1 b2
an – normalized incident power waves
bn – normalized reflected power waves

The RF Cavities
June 16th, 2006 12
1
an = (V n + Z I n ) (1)
2 Z
1
bn = (V n − Z I n ) (2)
2 Z
 Index n refers either to port number 1 or 2. The impedance Z0 is the
characteristic impedance of the connecting lines on the input and output
side of the network.

The RF Cavities
June 16th, 2006 13
 Inverting (1) leads to the following voltage and current expressions:
V n = Z  (a n + b n ) (3)
1
In = (a n − b n ) (4 )
Z

The RF Cavities
June 16th, 2006 14
 Recall the equations for power:
1
2
* 1
2
{ }
2 2
Pn = Re Vn I n =  an − bn  (5)

 Isolating forward and backward traveling wave components in (3) and
(4), we see
+
Vn
an = +
= Z I n (6)
Z
V n−
bn = = − Z  I n− (7 )
Z
The RF Cavities
June 16th, 2006 15
 We can now define S-parameters:
b1   S 11 S 12  a1 
 =    (8)
b 2  S 21 S 22  a 2 

The RF Cavities
June 16th, 2006 16
b1 Refkected powe wave at port 1

S11 = = (9)
a1 a = 0 Incident power wave at port 2
2
b2 Transmitte d powe wave at port 2

S 21 = = (10 )
2
b2 Refkected powe wave at port 2

S 22 = = (11)
1
b1 Transmitte d powe wave at port 1

S12 = = (12 )
1

The RF Cavities
June 16th, 2006 17
Observations
 a2=0, and a1=0 ⇒ no power waves are returned to the network at

either port 2 or port 1.
 However, these conditions can only be ensured when the connecting

transmission line are terminated into their characteristic impedances.
 Since the S-parameters are closely related to power relations, we can

express the normalized input and output waves in terms of time
averaged power.
 The average power at port 1 is given by
+ 2 + 2
P1 =
1 V1
2 Z
(1 − Γ ) = 2in
2 1 V1
Z
(1 − S ) 11
2
(13)

The RF Cavities
June 16th, 2006 18
 The reflection coefficient at the input side is expressed in terms of S11

under matched output according:
−
V1 b1
Γin = + = = S 11 (14 )
V1 a1 a
2 =0
 This also allow us to redefine the VSWR at port 1 in terms of S11 as
1 + S 11
VSWR = (15 )
1 − S 11
The RF Cavities
June 16th, 2006 19
 We can identify the incident power in (13) and express it in terms of

a1:
+ 2
1 V1
2
a1
= Pinc = (16 )
2 Z 2
Maximal available power
from the generator
 The total power at port 1 (under matched output condition) expressed
as a combination of incident and reflected powers:
( ) ( )
2
1 2 a1
P1 = Pinc + Prefl
2
= a1 − b1 = 1 − Γin
2
(17 )
2 2
The RF Cavities
June 16th, 2006 20
 If the reflected coefficient, or S11, is zero, all available power from the
source is delivered to port 1 of the network. An identical analysis at port
2 gives
( ) ( )
2
1 a2
2 2
P2 = a 2 − b 2 = 1 − Γout
2
(18 )
2 2

The RF Cavities
June 16th, 2006 21
Meaning of S-Parameters
 S-parameters can only be determined under conditions of perfect

matching on the input or the output side.
a1
Zo a2=0
  
VG1 Zo [S] Zo
   ZL
b2 b1
Measurement of S11 and S21 by matching the line impedance Zo
at port 2 through a corresponding load impedance ZL=Zo

The RF Cavities
June 16th, 2006 22
 This configuration allows us to compute S11 by finding the input

reflection coefficient:
Z in − Z 
S 11 = Γin = (19 )
Z in + Z 
 Taking the logarithm of the magnitude of S11 gives us the return loss
in dB
RL = −20 log S 11 (20)

The RF Cavities
June 16th, 2006 23
 With port 2 properly terminated, we find

−
b2 V2 Z
S 21 = = (21)
a1 a
2 =0
(V1 + Z I 1 ) (2 Z  ) I + =V + =0
2 2
 Since a2=0, we can set to zero the positive traveling voltage and
current waves at port 2.
 Replacing V1 by the generator voltage VG1 minus the voltage drop

over the source impedance Zo, VG1-ZoI1 gives
2V 2− 2V 2
S 21 = = (22 )
VG 1 VG 1
The RF Cavities
June 16th, 2006 24
 The forward power gain is

2
V2
G  = S 21 =
2
(23)
VG 1 2
 If we reverse the measurement procedure and attach a generator
voltage VG2to port 2 and properly terminate port 1, we can determine the
remaining two S-parameters, S22 and S12.
a2
a1=0 Zo
  
Zo Zo [S] Zo VG2
  
b1 b2
The RF Cavities
June 16th, 2006 25
 To compute S22 we need to find the output reflection coefficient Γout in

a similar way for S11:
Z out − Z 
S 22 = Γout = (24 )
Z out + Z 
−
b1 Z V1
S 12 = = (25 )
a 2 a =0 (V 2 + Z I 2 ) (2 Z  )
1 I 1+ =V1+ =0
2V1−
2
2V1 V1
S 12 = = (26 ) Gor = S 12 =
2
(27 )
VG 2 VG 2 VG 2 2
Reverse power gain

The RF Cavities
June 16th, 2006 26
Determination of a T-network elements
Find the S-parameters and resistive elements for the 3-dB

attenuator network. Assume that the network is placed into a
transmission line section with a characteristic line
impedance of Zo=50 Ω
R1 R2
  
Port 1 R3 Port 2
  

The RF Cavities
June 16th, 2006 27
An attenuator should be matched to the line impedance and

must meet the requirement S11= S22= 0.
R1 R2
   R 3 (R 2 + 50Ω )
Z in = R1 + = 50Ω
(R 3 + R 2 + 50Ω )
R3 50Ω
Because of symmetry, it is clear
that R1=R2.
  
Port 1 Port 2
Circuit for S11 and S21

The RF Cavities
June 16th, 2006 28
We now investigate the voltage V2 = V-2 at port 2 in

terms of V1=V+1.
R1 R2
 R 3 (R1 + 50Ω ) 
 
 ( )
R 3 + R1 + 50Ω  50Ω 
50Ω R3 V2 =  V1
 R 3 (R1 + 50Ω )  50Ω + R1 
 (R + R + 50Ω ) + R1 
 3 1 
Port 2
Port 1

The RF Cavities
June 16th, 2006 29
For a 3 dB attenuation, we require
2V 2 V2 1
S 21 = = = = 0.707 = S 12
VG 1 V1 2
Setting the ratio of V2/V1 to 0.707 and using the
input impedance expression, we can determine R1
and R3
2 −1
R1 = R 2 = Z  = 8.58Ω
2 +1
R 3 = 2 2Z  = 141 .4Ω

The RF Cavities
June 16th, 2006 30
Note: the choice of the resistor network ensures that at the

input and output ports an impedance of 50 Ω is maintained.
This implies that this network can be inserted into a 50 Ω
transmission line section without causing undesired
reflections, resulting in an insertion loss.

The RF Cavities
June 16th, 2006 31
Chain Scattering Matrix
 To extend the concept of the S-parameter presentation to

cascaded network, it is more efficient to rewrite the power wave
expressions arranged in terms of input and output ports. This results
in the chain scattering matrix notation. That is,
a1  T11 T12  b 2 

 =    (28 )
b1  T 21 T 22  a 2 
 It is immediately seen that cascading of two dual-port networks

becomes a simple multiplication.

The RF Cavities
June 16th, 2006 32
A A B B
a1 a2 b1 a2
 
Port 1 [T ]A [T ]B Port 2
 
A A B B
b1 b2 a1 b2
Cascading of two networks A and B

The RF Cavities
June 16th, 2006 33
If network A is described by
a1A  T11A T12A  b 2A 

 A=  A A  A 
(29 )
b1  T 21 T 22  a 2 
And network B by
a1B  T11B T12B  b 2B 

 B= B B  B 
(30 )
b1  T 21 T 22  a 2 

The RF Cavities
June 16th, 2006 34
b 2A  a1B 
 A= B (31)
a 2  b1 
Thus, for the combined system, we conclude
a1  T11
A A A  B
T12 T11 B  B 
T12 b 2
 A=  A A  B B  B 
(31)
b1  T 21 T 22  T 21 T 22  a 2 

The RF Cavities
June 16th, 2006 35
The conversion from S-matrix to the chain matrix notation is similar

as described before.
a1 a1 1
T11 = = = (32 )
b2 a S 21a1 S 21
2 =0
S 22
T12 = − (33)
S 21
S 11
T 21 = (34 )
S 21
− (S 11S 22 − S 12S 21 ) - ∆S
T 22 = = (35 )
S 21 S 21
The RF Cavities
June 16th, 2006 36
Conversely, when the chain scattering parameters are given and we

need to convert to S-parameters, we find the following relations:
b1 T 21b 2 T 21
S 11 = = = (36 )
a2 a T11b 2 T11
2 =0
(T11T 22 −T12T 21 ) ∆T
S 12 = = (37 )
T11 T11
1
S 21 = (38 )
T11
T12
S 22 = − (39 )
T11
The RF Cavities
June 16th, 2006 37
Conversion between Z- and S-Parameters
To find the conversion between the S-parameters and the Z-parameters,

let us begin with defining S-parameters relation in matrix notation
{b } = [S ]{a} (40)
Multiplying by Z gives
{ }
Z  {b} = V − = Z  [S ]{ a} = [S ] V + { } (41)
Adding {V + }= Z  {a} to both sides results in
{V } = [S ]{V + }+ {V + }= ([S ] + [E ]){V + } (42 )

The RF Cavities
June 16th, 2006 38
To compare this form with the impedance expression
{ V } = [Z ]{I }
We have to express {V+} in term of {I}. Subtract [S}{V+} from both sides of
{V + }= Z  {a}
{V + }− [S ]{V + }= Z  ({a} − {b}) = Z  {I } (43)
{V }= Z ([E ] − [S ])
+ −1
{ I } (44 )
The RF Cavities
June 16th, 2006 39
Substituting (44) into (42) yields
{ V } = ([S ] + [E ]){V + }= Z  ([S ] + [E ])([E ] − [S ])−1{ I } (45)

or
[Z ] = Z  ([S ] + [E ])([E ] − [S ])−1 (46 )
Explicitly
−1
 Z 11 Z 12  1 + S 11 S 12  1 − S 11 − S 12 
Z  = Z  − S 
 21 Z 22  S
 21 1 + S 22   21 1 − S 22 
1 + S 11 S 12 
Z  1 − S
 S 1 + S 22  S 12 
= 21 22
(47 )
(1 − S 11 )(1 − S 22 ) − S 21S 12  S 21 1 − S 11 
The RF Cavities
June 16th, 2006 40
Practical Network Analysis

The RF Cavities
June 16th, 2006 41
Linear Versus Nonlinear Behavior
A * Sin 360° * f ( t - t )
°
A
Linear behavior:
input and output frequencies are the
same (no additional frequencies created)
Time
to
Sin 360° * f * t
A
output frequency only undergoes
phase shift =
to * 360° * f
magnitude and phase change
Time
f Frequency
1
Input DUT Output
Nonlinear behavior:
output frequency may undergo frequency
shift (e.g. with mixers)
f Time
1 Frequency
additional frequencies created

(harmonics, inter-modulation)
f Frequency
1

The RF Cavities
June 16th, 2006 42
Criteria for Distortionless Transmission
Linear Networks
Constant amplitude over Linear phase over

bandwidth of interest bandwidth of interest
Magnitude
Phase
Frequency Frequency

The RF Cavities
June 16th, 2006 43
Magnitude Variation with Frequency
1 1
f ( t ) = sin ωt + sin 3ωt + sin 5ωt
3 5
Time
Time
Linear
Network
Magnitude
Frequency Frequency Frequency

The RF Cavities
June 16th, 2006 44
Phase Variation with Frequency
1 1
f ( t ) = sin ωt + sin 3ωt + sin 5ωt
3 5
Linear Network
Time Time
Magnitude
Frequency
0°
Frequency Frequency
-180°
-360 °

The RF Cavities
June 16th, 2006 45
Criteria for Distortionless Transmission
Nonlinear Networks
Saturation, crossover, inter-modulation, and other

nonlinear effects can cause signal distortion
Time Time
Frequency Frequency

The RF Cavities
June 16th, 2006 46
The Need for Both Magnitude and Phase
S 21
1. Complete characterization S 11 S 22
of linear networks 4. Time Domain
S 12 Characterization
2. Complex impedance needed to

Mag
design matching circuits
Time
High Frequency
Transistor Model
5. Vector Accuracy Enhancement

Error
Base Collector
3. Complex values needed for Measured

device modeling
Emitter Actual

The RF Cavities
June 16th, 2006 47
High-Frequency Device Characterization
Lightwave Analogy
Incident
Transmitted
Reflected

The RF Cavities
June 16th, 2006 48
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)
is very important for low reflection
Voltage dependent on position along line

The RF Cavities
June 16th, 2006 49
Transmission Line Terminated with Zo
Zo = characteristic impedance of
Zs = Z o 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

The RF Cavities
June 16th, 2006 50
Transmission Line Terminated with Short, Open
Zs = Z o
V inc
o
Vrefl In phase (0 ) for open
o
Out of phase (180 ) for short
For reflection, a transmission line terminated in a short

or open reflects all power back to source
The RF Cavities
June 16th, 2006 51
Transmission Line Terminated with 25Ω
Zs = Z o
ZL = 25 Ω
V inc
Vrefl
Standing wave pattern does not go to zero as

with short or open

The RF Cavities
June 16th, 2006 52
High-Frequency Device Characterization
Incident
Transmitted
R
B
Reflected
A
REFLECTION TRANSMISSION
Reflected A Transmitted B
= =
Incident R Incident R
Return Group
SWR Gain / Loss Delay
Loss
S-Parameters Impedance, Insertion
S11,S22 Reflection Admittance S-Parameters Phase
Coefficient S21,S12 Transmission
R+jX,
Γ, ρ G+jB Coefficient
Τ,τ

The RF Cavities
June 16th, 2006 53
Reflection Parameters
ZL − ZO
Γ
Vreflected
Reflection
Coefficient
=
Vincident
= ρ Φ =
Z L + ZO
Return loss = -20 log(ρ), ρ = Γ
Voltage Standing Wave Ratio
Emax
Emin Emax 1+ρ
VSWR = =
Emin 1-ρ
No reflection Full reflection

(ZL = Zo) (ZL = open, short)
0 ρ 1
∞ dB RL 0 dB
1 VSWR ∞
The RF Cavities
June 16th, 2006 54
Transmission Parameters
V Incident V Transmitted
DUT
VTransmitted
Transmission Coefficient =Τ = = τ∠φ
VIncident
V
Insertion Loss (dB) = - 20 Log
Trans
= - 20 log τ
V Inc
V
Gain (dB) = 20 Log
Trans
= 20 log τ
V Inc

The RF Cavities
June 16th, 2006 55
Deviation from Linear Phase
Use electrical delay to remove
linear portion of phase response
Linear electrical
Deviation from
RF filter response length added
linear phase
(Electrical delay function)
Phase 45 /Div
Phase 1 /Div
o
o
+ yields
Frequency Frequency Frequency
Low resolution High resolution

The RF Cavities
June 16th, 2006 56
Low-Frequency Network Characterization
H-parameters Y-parameters Z-parameters

V1 = h11I1 + h12V2 I1 = y11V1 + y12V2 V1 = z11I1 + z12I2
V2 = h21I1 + h22V2 I2 = y21V1 + y22V2 V2 = z21I1 + z22I2
h11 = V1
I1 V2=0
(requires short circuit)
h12 = V1 (requires open circuit)
V2 I1=0
All of these parameters require measuring voltage

and current (as a function of frequency)

The RF Cavities
June 16th, 2006 57
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
(gain, loss, reflection coefficient ...)
Relatively easy to measure
S 21
Can cascade S-parameters of multiple a1
Incident Transmitted
devices to predict system performance S11

b2
Reflected DUT
Analytically convenient S 22
Port 1 Port 2 Reflected
CAD programs b1 a2
Flow-graph analysis Transmitted S 12 Incident
Can compute H, Y,or Z parameters from S- b 1 = S11 a 1 + S 12 a 2
parameters if desired b 2 = S21 a 1 + S 22 a 2

The RF Cavities
June 16th, 2006 58
Measuring S-Parameters
S b2
Incident 21 Transmitted
a1
Z0
Forward S 11
DUT Load
Reflected
b1 a2 = 0
Reflected b1
S 11 = = a
Incident 1 a2 = 0 b2
S 22 = Reflected
b = a a1 = 0
Transmitted 2 Incident 2
S 21 = = a
Incident 1 a2 = 0 b
Transmitted 1
S 12 = = a
Incident 2 a1 = 0
a1 = 0 b2
S 22
Z0 DUT
Reflected Reverse
Load
a2
Transmitted S 12 Incident
b1
The RF Cavities
June 16th, 2006 59
What is the difference between .
network and spectrum analyzers?

Hard: getting (accurate) trace Easy: getting trace
Easy: interpreting results Hard: interpreting results
SPECTRUM ANALYZER 9 kHz - 26.5

8563A GHz
Amplitude Ratio
Power
Measures Measures
known signal unknown
signals
Frequency Frequency
Network analyzers: Spectrum analyzers:

 measure components, devices, circuits,  measure signal amplitude characteristics
sub-assemblies (carrier level, sidebands, harmonics...)
 contain source and receiver  are receivers only (single channel)
 display ratioed amplitude and phase  can be used for scalar component test (no
(frequency or power sweeps) phase) with tracking gen. or ext. source(s)

The RF Cavities
June 16th, 2006 60
Signal Separation
Measuring incident signals for ratioing
50 Ω
6 dB Splitter
usually resistive
non-directional
50 Ω
6 dB
broadband
Coupler
Main directional
low loss
signal
good isolation, directivity

Coupled hard to get low freq performance
signal

The RF Cavities
June 16th, 2006 61
Forward Coupling Factor
Coupling, forward
-20 dBm
.01 mW
Source
Z0
0 dBm -.046 dBm

1 mW .99 mW
Example of 20 dB Coupler
P coupling forward
Coupling Factor (dB) = -10 log
P incident

The RF Cavities
June 16th, 2006 62
Directional Coupler Isolation
(Reverse Coupling Factor)
Coupling, reverse
this is an error signal during
-50 dBm measurements
.00001 mW
Source
Z0
0 dBm 1 −.046 dBm

mW .99 mW
Example of 20 dB Coupler "turned around"
Pcoupled reverse
Isolation Factor (dB) = -10 log
Pincident

The RF Cavities
June 16th, 2006 63
Directional Coupler Directivity
Pcoupled forward
Directivity (dB) = 10 log
Pcoupled reverse
Coupling Factor
Directivity =
Isolation
Directivity (dB) = Isolation (dB) - Coupling Factor (dB)
Example of 20 dB Coupler with 50 dB isolation:

Directivity = 50 dB - 20 dB = 30 dB

The RF Cavities
June 16th, 2006 64
Measuring Coupler Directivity the Easy Way
1.0 (0 dB) (reference)
Coupler Good approximation for

Directivity coupling factors ≥ 10 dB
35 dB
short
Source .018 (35 dB) (normalized)
Directivity = 35 dB - 0 dB = 35
dB
Source
load Assume perfect load

The RF Cavities
June 16th, 2006 65
Narrowband Detection - Tuned Receiver
ADC / DSP
Best sensitivity / dynamic range

Provides harmonic / spurious signal rejection
Improve dynamic range by increasing power,
decreasing IF bandwidth, or averaging
Trade off noise floor and measurement speed
10 MHz 26.5 GHz

The RF Cavities
June 16th, 2006 66
Comparison of Receiver Techniques
Broadband (diode) Narrowband (tuned-

detection receiver) detection
0 dB 0 dB
-50 dB -50 dB
-100 dB -100 dB
-60 dBm Sensitivity < -100 dBm Sensitivity
higher noise floor high dynamic range

false responses harmonic immunity
Dynamic range = maximum receiver power - receiver noise floor

The RF Cavities
June 16th, 2006 67
Dynamic Range and Accuracy
Dynamic range is very important
for measurement accuracy !
Error (dB, deg) Error Due to Interfering Signal
100
+ magn (dB)
10 - magn (dB)
phase (± deg)
phase error
1
magn error
0.1
0.01
0.001
0 -5 -10 -15 -20 -25 -30 -35 -40 -45 -50 -55 -60 -65 -70
Interfering signal (dB)
The US Particle Accelerator School Boston University June 16th, 2006

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:
Errors:
instrument noise (source SYSTEMATIC
Measured Unknown
phase noise, IF noise floor, etc.)
switch repeatability
Data
RANDOM Device
connector repeatability
Drift errors DRIFT
are due to instrument or test-system performance
changing after a calibration has been done
are primarily caused by temperature variation
can be removed by further calibration(s)

The RF Cavities
June 16th, 2006 69
Systematic Measurement Errors
R A B
Directivity Crosstalk
DUT
Frequency response
 reflection tracking (A/R) Source Load
 transmission tracking (B/R) Mismatch Mismatch
Six forward and six reverse error terms yields 12

error terms for two-port devices

The RF Cavities
June 16th, 2006 70
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,
thru
then does data divided by memory
vector
• requires more standards
• requires an analyzer that can measure phase
• accounts for all major sources of systematic error
SHORT
OPEN
thru
S11A LOAD
S11M

The RF Cavities
June 16th, 2006 71
ABCD parameter representation
Very useful when cascading networks
i`1 i``2
+  +
i`2 +
i``1
v`1  v``2
v`2 v``1
-    -
-
i`2 + i``1 i``2 +

A′ B ′
  A ′′ B ′′  
Port 1  
v`2 v``1 C ′′ D ′′ v``2 Port 2
 C ′ D ′    
- -
 ν1   A ′ B ′   A ′′ B ′′   ν′2′ 
 =     
 i 1  C ′ D ′ C ′′ D ′′  − i 2′′ 
The RF Cavities
June 16th, 2006 72
ABCD parameter representation
ABCD very useful for T.L.
i1 l i2
   cos(βl ) jZ  sin(βl )
A B  
C D  =  j sin(βl )
Z0, β 
cos(βl )
v1 v2
  Z 
 
 
Feedback loop
Output matching
Input matching R Network
R Network
L L
L L  
  
    
  C
Z0, βl C Z0, βl
 
 

The RF Cavities
June 16th, 2006 73
Signal Flow Computations
Complicated networks can be efficiently analyzed in a manner identical
to signals and systems and control.
a a

Z0 ZL  ΓL

b
b
in general
ai Γij bj
 

The RF Cavities
June 16th, 2006 74
Signal Flow Graphs
Basic Rules:
We’ll follow certain rules when we build up a network flow graph.
1. Each variable, a1, a2, b1, and b2 will be designated as a node.
2. Each of the S-parameters will be a branch.
3. Branches enter dependent variable nodes, and emanate from the independent
variable nodes.
4. In our S-parameter equations, the reflected waves b1 and b2 are the dependent
variables and the incident waves a1 and a2 are the independent variables.
5. Each node is equal to the sum of the branches entering it.

The RF Cavities
June 16th, 2006 75
Signal Flow Graphs
Let’s apply these rules to the two S-parameters equations
b1 = S 11a1 + S 12a 2
b 2 = S 21a1 + S 22a 2
First equation has three nodes: b1, a1, and a2. b1 is a dependent node and is
connected to a1 through the branch S11 and to node a2 through the branch S12.
The second equation is similar.
a1  S21
a1   b2
S11
S22
b1  
S22 a2  a2

The RF Cavities
June 16th, 2006 76
Signal Flow Graphs
Complete Flow Graph for 2-Port
a1 b2
S21



S11  S11
 
b1 S12 a2
The relationship between the traveling waves is now easily seen. We
have a1 incident on the network. Part of it transmits through the network
to become part of b2. Part of it is reflected to become part of b1.
Meanwhile, the a2 wave entering port two is transmitted through the
network to become part of b1 as well as being reflected from port two as
part of b2. By merely following the arrows, we can tell what’s going on in
the network. This technique will be all the more useful as we cascade
networks or add feedback paths.
The RF Cavities
June 16th, 2006 77
Arrangement for Signal Flow Analysis
ZG b’ a
  
IG
VG Z0 ZL
 
a’ b
bs b’ a bs a
 

ΓS ΓL
  
ΓS ΓL
a’ b 
Z b
bs = VG
ZG + Z 
The RF Cavities
June 16th, 2006 78
Analysis of Most Common Circuit
bs a1 a2
ZS ΓS ΓL
   
VS Z0 [S] Z0 ZL
   
b1 b2
bs  a1 S21 b2
  
bs a1
ΓS S11 S22 ΓL  
    
b1 S12 a2 a1 1
→
bs  S 12S 21 
1 −  S 11 + ΓL ΓS
 1 − S 22 ΓL 
The RF Cavities
June 16th, 2006 79
bs  a1
 
S 12S 21
ΓS S11
1 − S 22 ΓL
ΓL
  
b1
bs  a1

S 12S 21
ΓS S 11 + ΓL
1 − S 22 ΓL

b1 S 12S 21 
Γin = = S 11 + ΓL b1
a1 1 − S 22 ΓL
Note: Only ΓL = 0 ensures that S11 can be measured.

The RF Cavities
June 16th, 2006 80
Scattering Matrix
The scattered-wave amplitudes are linearly related to the incident wave

amplitudes. Consider the N port junction
V+4 V-5
V+5
If the only incident wave is V+1 then V-4
Port 4 Port 5 V-N

V1− = S 11V1+ V+3 Port N
Port 3
V+N
S 11 is the reflection coefficient V-3 Port 2
Port 1
The total voltage is port 1 is V1 = V1+ +V1− V-1

V+2 V+1
Waves will also be scattered out of other V-2
ports. We will have
V n− = S n1V n+ n = 2 ,3,4 ,...N

The RF Cavities
June 16th, 2006 81
Scattering Matrix
If all ports have incident wave then
V1−   S 11 S 12 S 13 ... S 1N  V1+ 

 −   + 
V 2  =  S 21 S 22 S 23 ... S 2N
 V 2 
 ...   ... ... ... ... ...   ... 
 −   + 
V N  S N 1 S N 2 S N 3 ... S NN  V N 
or
[V ] = [S ][V ]
− +
v i−
[S ] is called the scattering matrix S ij = + for V k+ = 0 (k ≠ j )
vj

The RF Cavities
June 16th, 2006 82
Scattering Matrix
If we choose the equivalent Z0 equal to 1 then the incident power is

given by
1 +2
Vn
2
and the scattering will be symmetrical. With this choice
V = V + +V − ,I = I + + I −
and
1
V +
= (V + I )
2
1
V −
= (V − I )
2
The RF Cavities
June 16th, 2006 83
Scattering Matrix
V+ and V- are the variables in the scattering matrix formulation; but

they are linear combination of V and I.
Other normalization are
V I
ν= i=
Z Z
Just as in the impedance matrix there are several properties of the
scattering matrix we want to consider.
1. A shift of the reference planes
2. S matrix for reciprocal devices
3. S matrix for the lossless devices

The RF Cavities
June 16th, 2006 84
Scattering Matrix
Shift in the reference planes

Consider the following network, where tn is the original location of the
reference plane, and t’n in the new location of the reference plane.
The electrical length between t n and t n′ is

θ = βn l n .
S mn m ≠ n must be multiplied by e − jθn

ln
− j 2θ n
S nn must be multiplied by e
t n′ tn
Why is this a
factor of 2?

The RF Cavities
June 16th, 2006 85
Scattering Matrix
+ + jθ n
′
V n =V n e
V n′− = V n−e − jθn
e − jθ1  e − jθ1 
 − jθ 2
  S 11 S 12 ... S 1N   − jθ 2

 e    e 
[S ] =  .   S 21 S 22 ... S 2N 
 . 
  ... ... ... ...   
 .    . 
 − jθ N  S N 1 S N 2 ... S nn  
− jθ N 
 e   e 

The RF Cavities
June 16th, 2006 86
Scattering Matrix
Consider a 2×2 scattering matrix where two reference planes

are shifted.
e − jθ1 0   S 11 S 12  e − jθ1 0 
[S ′] =  − jθ 2  S  − jθ 2 
 0 e   21 S 22   0 e 
 S 11e − jθ1 S 12e − jθ1e − jθ2 
[S ′] =  − jθ1 − jθ2 − jθ 2 
S 21e e S 22e 
 S 11e − j 2θ1 S 21e − jθ1e − jθ2 
[S ′] =  − jθ1 − jθ2 − j 2θ2 
S 12e e S 22e 

The RF Cavities
June 16th, 2006 87
Scattering Matrix
Proof of Symmetry of Scattering Matrix

For a reciprocal junction Smn = Snm m≠n provided that Z0=1
1 +2
P = Vn for all modes.
2
V n = V n+ + V n− and I n = I n+ − I n−
∴ I n = V n+ −V n− (Z  = 1)
[V ] = [V +
]+ [V ] −
= [Z ][I ]
= [Z ]V [ ]− [Z ][V ]
+ −

The RF Cavities
June 16th, 2006 88
Scattering Matrix
Defining the unit matrix
1 0
 1 
[u ] =  
 . 
0 
 1
[V ]+ [V ] = [Z ][V ]− [Z ][V ]
+ − + −
Rearrange and factor
([Z ] + [U ])[V − ] = ([Z ] − [U ])[V + ]

or
[V ] = ([Z ] + [U ])
− −1
([Z ] − [U ])[V + ]
but [V ] = [S ][V ]
− +
∴ [S ] = ([Z ] + [U ])−1 ([Z ] − [U ])
The RF Cavities
June 16th, 2006 89
Scattering Matrix
Loss-less Junction
The total power leaving a junction must be equal to the total power entering
the junction power.
N 2 N 2
− +
∑ n
V = ∑ n
V
n =1 n =1
but
N
V n = ∑ S niV i +
−
i =1
so that
N N 2 N 2
+ +
∑ ∑ ni i
S V = ∑ n
V
n =1 i =1 n =1

The RF Cavities
June 16th, 2006 90
Scattering Matrix
Loss-less Junction
V+n are all independent incident voltages, so we choose V+n =0 except for
n=i N N 2 N 2
+
∑ ∑ S niV i = ∑ V n+
n =1 i =1 n =1
N 2
+ + 2
∑ S niV i = Vn
n =1
N 2 N
∑ S ni = ∑ S ni S ni
*
= 1 ∀i .
n =1 n =1
 S 1*i 
 * 
S
[S 1i S 2 i ... S Ni ]  2 i  = 1
 ... 
 * 
S Ni 
The RF Cavities
June 16th, 2006 91
Scattering Matrix
Loss-less 2 Port Junction

For this case [S] is unitary
N
∴ ∑ S nm S np
*
= δ mp
n =1
or
S 11S 11
*
+ S 12 S 12
*
= 1 
 ⇒ S 11 = S 22
S 22 S 22 + S 12 S 12 = 1
* *
S 11S 12
*
+ S 12 S 22
*
=0
The magnitude of the input and output ports are equal in magnitude. Also
2
S 12 = 1 − S 11

The RF Cavities
June 16th, 2006 92
Scattering Matrix
If we know S 11 then we can obtain S 12 and S 22 .
Note: The fraction of power reflected at terminal t, is
P refl 2
= S 11
Pinc
So that the insertion loss due to reflection is
IL = 10 log 1 − S 11 ( 2
)
IL = 20 log S 12

The RF Cavities
June 16th, 2006 93
Scattering Matrix
Example: two-port network Equivalent Circuit
Z1 Z2
  
t2
V1+ I1 Z3 I2 V 2+
l2
l1   
t1 Port 1 Port 2
Assume TE10 modes at t1 and t2

Apply KVL:
V1 = Z 1I 1 + Z 3I 1 + Z 3I 2
V 2 = Z 2 I 2 + Z 3I 2 + Z 3I 1

The RF Cavities
June 16th, 2006 94
Scattering Matrix
If
V1
Z 3 = Z 12 =
I2 I
1 =0
Z 1 = Z 11 − Z 12
Z 2 = Z 22 − Z 12
Then we have
V1 = Z 11I 1 + Z 12 I 2 Z 11 − Z 12 Z 22 − Z 12
V 2 = Z 22 I 2 + Z 12 I 2
and Z 12
[V ] = [Z ][I ]

The RF Cavities
June 16th, 2006 95
Scattering Matrix
This can be transformed into an admittance matrix

Y12
 I 1  Y11 Y12  V1 

I  = Y Y  V  Y11 −Y12 Y 22 −Y12
 2   12 22   2 

The RF Cavities
June 16th, 2006 96
Scattering Matrix
Traveling Wave:
+ − δx − δx
V = Ae ,V = Ae
V (x ) = V + ( x ) + V − ( x )
Similarly for current:
V +
(x ) V −
( x)
I (x ) = I (x ) − I (x ) =
+ −
−
Z Z
Reflection Coefficient:
V − (x )
Γ (x ) = +
V (x )
The RF Cavities
June 16th, 2006 97
Scattering Matrix
Introduce “normalized” variables:

ν (x ) = V ( x ) Z  , i (x ) = Z  I (x )
So that
ν(x ) = a (x ) + b (x ) i (x ) = a (x ) − b (x ) and b (x ) = Γ(x )a (x )
This defines a single port network. What about 2-port?
2-port
b1 = S 11a1 + S 12a 2
b 2 = S 21a1 + S 22a 2

The RF Cavities
June 16th, 2006 98
Scattering Matrix
Each reflected wave (b1,b2) has two contributions: one from the incident wave at
the same port and another from the incident wave at the other port.
How to calculate S-parameters?
b1
S 11 = Input reflected coefficient with output matched.
a1 a
2 =0
b1
S 12 = Reverse transmission coefficient with input matched.
a 2 a1=0
b2
S 21 = Transmission coefficient with output matched.
a1 a
2 =0
b2
S 22 = Output reflected coefficient with input matched.
a2 a
1 =0

The RF Cavities
June 16th, 2006 99
Scattering Matrix
Generalized Scattering Matrix:

We define it via [b]=[S][a]. S-matrix depends on the choice of normalized
impedance. Usually 50 Ω, but can be anything and can even be complex!
Calculating Sij:
bi V i − Z *,i I iZ i − Z *,i
S ij = = =
ai a k = 0 ,k ≠ i ,k =1,...n V i + Z  ,i I i Z i + Z  ,i
*
Which is input reflected coefficient with all other ports matched.
bk
S ki =
ai a
k = 0 ,k ≠ i ,k =1,...n
is equal to transducer power gain from i to k with ports other than i matched.

The RF Cavities
June 16th, 2006

Lecture 7

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lecture 7

Uploaded by

Copyright:

Available Formats

Lecture 7

Massachusetts Institute of Technology

VVM: The vector voltmeter measures the magnitude of a

Unfortunately, the use of the directional couplers and test cables

Massachusetts Institute of Technology

(HP 8657B) (HP 8508A)

Massachusetts Institute of Technology

Massachusetts Institute of Technology

To find KA and KB it is necessary to make a second measurement with a

Generator Channel A Channel B

The DUT is connected directly between two directional couplers.

Massachusetts Institute of Technology

Transmission measurements: S12 or S21

Massachusetts Institute of Technology

 Scattering Parameters (S-Parameters) plays a major role is network

 This importance is derived from the fact that practical system

 Also open circuit leads to capacitive loading at the terminal.

Massachusetts Institute of Technology

 In either case, the open/short-circuit conditions needed to determine

 Moreover, when dealing with wave propagation phenomena, it is not

 For instance, the terminal discontinuity will cause undesirable voltage

 With S-parameters, one has proper tool to characterize the two-port

Massachusetts Institute of Technology

 S-parameters are power wave descriptors that permit us to define the

bn – normalized reflected power waves

Massachusetts Institute of Technology

Massachusetts Institute of Technology

 Inverting (1) leads to the following voltage and current expressions:

Massachusetts Institute of Technology

 Recall the equations for power:

 We can now define S-parameters:

Massachusetts Institute of Technology

b1 Refkected powe wave at port 1

b2 Transmitte d powe wave at port 2

b2 Refkected powe wave at port 2

b1 Transmitte d powe wave at port 1

Massachusetts Institute of Technology

 a2=0, and a1=0 ⇒ no power waves are returned to the network at

 However, these conditions can only be ensured when the connecting

 Since the S-parameters are closely related to power relations, we can

 The average power at port 1 is given by

Massachusetts Institute of Technology

 The reflection coefficient at the input side is expressed in terms of S11

 We can identify the incident power in (13) and express it in terms of

Massachusetts Institute of Technology

 S-parameters can only be determined under conditions of perfect

Massachusetts Institute of Technology

 This configuration allows us to compute S11 by finding the input

RL = −20 log S 11 (20)

Massachusetts Institute of Technology

 With port 2 properly terminated, we find

 Replacing V1 by the generator voltage VG1 minus the voltage drop

 The forward power gain is

 To compute S22 we need to find the output reflection coefficient Γout in

Massachusetts Institute of Technology

Find the S-parameters and resistive elements for the 3-dB

Massachusetts Institute of Technology

An attenuator should be matched to the line impedance and

Massachusetts Institute of Technology

We now investigate the voltage V2 = V-2 at port 2 in

Massachusetts Institute of Technology

For a 3 dB attenuation, we require