Upload
anabel-fields
View
215
Download
0
Tags:
Embed Size (px)
Citation preview
11
ME1000 RF CIRCUIT DESIGN
This courseware product contains scholarly and technical information and is protected by copyright laws and international treaties. No part of this publication may be reproduced by any means, be it transmitted, transcribed, photocopied, stored in a retrieval system, or translated into any language in any form, without the prior written permission of Acehub Vista Sdn. Bhd. or their respective copyright owners.
The use of the courseware product and all other products developed and/or distributed by Acehub Vista Sdn. Bhd. are subject to the applicable License Agreement.
For further information, see the Courseware Product License Agreement.
http://dreamcatcher.asia/cw
2
3B. RF Microwave Filters
3
1.0 Basic Filter Theory
4
Introduction
• An ideal filter is a linear 2-port network that provides perfect transmission of signal for frequencies in a certain passband region, infinite attenuation for frequencies in the stopband region, and a linear phase response in the passband (to reduce signal distortion).
• The goal of filter design is to approximate the ideal requirements within acceptable tolerance with circuits or systems consisting of real components.
5
Categorization of Filters
• Low pass filter (LPF), high pass filter (HPF), bandpass filter (BPF), bandstop filter (BSF), arbitrary type, etc.
• In each category, the filter can be further divided into active and passive types.
• In an active filter, there can be amplification of the signal power in the passband region; a passive filter do not provide power amplification in the passband.
• Filters used in electronics can be constructed from resistors, inductors, capacitors, transmission line sections, and resonating structures (e.g., piezoelectric crystal, Surface Acoustic Wave (SAW) devices, mechanical resonators, etc.).
• An active filter may contain a transistor, FET, and an op-amp.
Filter
LPF BPFHPF
Active Passive Active Passive
6
Filter Frequency Response
• Frequency response implies the behavior of the filter with respect to steady-state sinusoidal excitation (e.g., energizing the filter with a sine voltage or current source and observing its output).
• There are various approaches to displaying the frequency response:– Transfer function H() (the traditional approach)– Attenuation factor A()– S-parameters, e.g., s21()
– Others, such as ABCD parameters, etc.
7
Filter Frequency Response (cont’d)
• Low pass filter (passive)
Filter H()
V1() V2()ZL
c
A()/dB
0 c
3
10
20
30
40
50
1
21020A
V
VLognAttenuatio (1.1b)
1
2
V
VH (1.1a)
c
|H()|
1Transfer function
Arg(H())
Complex value
Real value
8
Filter Frequency Response (cont’d)
• Low pass filter (passive) continued...
• For the impedance matched system, using s21 to observe the filter response is more convenient, as this can be easily measured using a vector network analyzer (VNA).
Zc
01
221
01
111
22
aa a
bs
a
bs
ZcZc
Transmission lineis optional
c
20log|s21()|
0 dB
Arg(s21())
FilterZcZc
ZcVs
a1 b2
Complex value
9
• Low pass filter (passive) continued...
Filter Frequency Response (cont’d)
A()/dB
0 c
3
10
20
30
40
50
Filter H()
V1() V2() ZL
Passband
Stopband
Transition band
Cut-off frequency (3 dB)
10
• High pass filter (passive)
Filter Frequency Response (cont’d)
A()/dB
0 c
3
10
20
30
40
50
c
|H()|
1
Transfer function
Stopband
Passband
11
Filter Frequency Response (cont’d)
Bandpass filter (passive) Bandstop filter
A()/dB
40
1
3
30
20
10
0 2o
1
|H()|
1 Transfer function
2o
A()/dB
40
1
3
30
20
10
0 2o
1
|H()|
1
Transfer function
2o
12
Basic Filter Synthesis Approaches
• Image Parameter Method.
ZoZo Zo
ZoZo
Filter Zo
H1() H2() Hn() Zo
Zo
• Consider a filter to be a cascade of linear 2-port networks.• Synthesize or realize each 2-port network, so that the combine effect gives the required frequency response.• The ‘image’ impedance seen at the input and output of each network is maintained.
The combinedresponse
Response ofa singlenetwork
13
Basic Filter Synthesis Approaches (cont’d)
• Insertion Loss Method.
Filter Zo
Zo
Use the RCLM circuit synthesis theorem to come up with a resistive terminatedLC network that can produce theapproximate response. Zo
IdealApproximate with rational polynomialfunction
|H()|
obsbnsnbns
oasansnansKsH
11
1
11
1
We can also use Attenuation Factor or |s21| for this.
Approximate ideal filter responsewith polynomial function:
14
Our Scope
• Only concentrate on passive LC and stripline filters.
• Filter synthesis using the Insertion Loss Method (ILM). The Image Parameter Method (IPM) is more efficient and suitable for simple filter designs, but has the disadvantage that arbitrary frequency response cannot be incorporated into the design.
15
2.0 Passive LC Filter Synthesis Using the Insertion Loss Method
16
Insertion Loss Method (ILM)
• The insertion loss method (ILM) enables a systematic way to design and synthesize a filter with various frequency responses.
• The ILM method also enables a filter performance to be improved in a straightforward manner, at the expense of a ‘higher order’ filter.
• A rational polynomial function is used to approximate the ideal |H()|, A(), or |s21()|.
• Phase information is totally ignored.
• Ignoring phase simplifies the actual synthesis method. An LC network is then derived which will produce this approximated response.
• The attenuation A() can be cast into power attenuation ratio, called the Power Loss Ratio, PLR, which is related to A()2.
17
More on ILM
• There is a historical reason why phase information is ignored. Original filter synthesis methods are developed in the 1920s–60s, for voice communication. The human ear is insensitive to phase distortion, thus only the magnitude response (e.g., |H()|, A()) is considered.
• Modern filter synthesis can optimize a circuit to meet both magnitude and phase requirements. This is usually done using computer optimization procedures with ‘goal functions’.
Extra
18
Power Loss Ratio (PLR)
211
12
11
Load todeliveredPower network source from availablePower
AP
AP
LoadPincP
LRP
•PLR large, high attenuation•PLR close to 1, low attenuationFor example, a low passfilter response is shownbelow:
•PLR large, high attenuation•PLR close to 1, low attenuationFor example, a low passfilter response is shownbelow:
ZLVs
Lossless2-port network
1
Zs
PAPin
PL
PLR(f)
Low pass filter PLRf
1
0
Low attenuation
Highattenuation
fc
(2.1a)
19
PLR and s21
• In terms of incident and reflected waves, assuming ZL = Zs = ZC.
ZcVs
Lossless2-port network
Zc
PAPin
PL
a1
b1
b2
221
1
2
21
222
1
212
1
sLR
ba
b
a
LPAP
LR
P
P
(2.1b)
20
PLR for the Low Pass Filter (LPF)
• Since |1()|2 is an even function of , it can be written in terms of 2 as:
• PLR can be expressed as:
• Various types of polynomial function in can be used for P(). The requirement is that P() must either be an odd or even function. Among the classical polynomial functions are:– Maximally flat or Butterworth functions– Equal ripple or Chebyshev functions– Elliptic function– Many, many more
21 PPLR
22
22
NM
M
2
2
22
2
1
12
11
1 1
N
M
NM
MLRP
2
22
N
MP
(2.2)
(2.3a)
(2.3b)
This is also knownas Characteristic Polynomial
The characteristics we need from [P()]2 for LPF: • [P()]2 0 for < c
• [P()]2 >> 1 for >> c
The characteristics we need from [P()]2 for LPF: • [P()]2 0 for < c
• [P()]2 >> 1 for >> c
PP
21
Characteristic Polynomial Functions
• Maximally flat or Butterworth:
• Equal ripple or Chebyshev:
• Bessel or linear phase:
N
cP
2 , 2
1
21
1
0
nCCC
C
C
C
nnn
N
factor ripple , NCP
N = order of theCharacteristicPolynomial P()
N = order of theCharacteristicPolynomial P()
12 jBjBP
2 , 12
1
1
22
1
1
0
nsBssBssB
ssB
sB
sB
nnn
N
(2.4a)
(2.4b)
(2.4c)
22
Examples of PLR for the Low Pass Filter
• PLR of the low pass filter using 4th order polynomial functions (N = 4) – Butterworth, Chebyshev (ripple factor =1), and Bessel. Normalized to c = 1 rad/s, k = 1.
0 0.5 1 1.5 21
10
100
1 103
1 104
PLRbt ( )
PLRcb ( )
PLRbs ( )
Butterworth
Chebyshev
Bessel
222
)( 1481
cckP chebyshevLR
242
)( 1
ckP hButterwortLR
1051054510
11
234
1051
2)(
cs
cs
cs
cs
BesselLR
sB
jBjBkP
PLR
Ideal
If we convert into dB,this ripple is equal to3 dB
k=1
23
Examples of PLR for the Low Pass Filter (cont’d)
• PLR of the low pass filter using the Butterworth characteristic polynomial, normalized to c = 1 rad/s, k = 1. 2
2)( 1
N
chButterwortLR kP
0 0.5 1 1.5 21
10
100
1 103
1 104
1 105
PLR 2( )
PLR 3( )
PLR 4( )
PLR 5( )
PLR 6( )
PLR 7( )
N=2
N=6
N=4
N=5
N=3
N=7 Conclusion:The type of polynomialfunction and the orderdetermine the attenuation rate in the stopband.
Conclusion:The type of polynomialfunction and the orderdetermine the attenuation rate in the stopband.
24
Characteristics of Low Pass Filters Using Various Polynomial Functions
• Butterworth: Moderately linear phase response, slow cutoff, smooth attenuation in the passband.
• Chebyshev: Bad phase response, rapid cutoff for a similar order, contains ripple in the passband. May have impedance mismatch for N even.
• Bessel: Good phase response, linear. Very slow cutoff. Smooth amplitude response in the passband.
25
Low Pass Prototype Design
• A lossless linear, passive, reciprocal network that can produce the insertion loss profile for the low pass filter is the LC ladder network.
• Many researchers have tabulated the values for the L and C for the low pass filter with cut-off frequency c = 1 rad/s, that works with the source and load impedance Zs = ZL = 1 .
• This low pass filter is known as the Low Pass Prototype (LPP).
• As the order N of the polynomial P increases, the required element also increases. The no. of elements = N.
L1=g2 L2=g4
C1=g1 C2=g3 RL= gN+1
1
L1=g1L2=g3
C1=g2 C2=g4RL= gN+1g0= 1
Dual of eachother
26
Low Pass Prototype Design (cont’d)
• The LPP is the ‘building block’ from which real filters may be constructed.
• Various transformations may be used to convert it into a high pass, bandpass, or other filter of arbitrary center frequency and bandwidth.
• The following slides show some sample tables for designing LPP for Butterworth and Chebyshev amplitude response of PLR.
27
Table for the Butterworth LPP Design
N g1 g2 g3 g4 g5 g6 g7 g8 g91 2.0000 1.00002 1.4142 1.4142 1.00003 1.0000 2.0000 1.0000 1.00004 0.7654 1.8478 1.8478 0.7654 1.00005 0.6180 1.6180 2.0000 1.6180 0.6180 1.00006 0.5176 1.4142 1.9318 1.9318 1.4142 0.5176 1.0000
7 0.4450 1.2470 1.8019 2.0000 1.8019 1.2470 0.4450 1.0000
8 0.3902 1.1111 1.6629 1.9615 1.9615 1.6629 1.1111 0.3902 1.0000
See Example 2.1 in the following slides on how the constant values g1, g2, g3, … etc., are obtained.
28
Table for the Chebyshev LPP Design
• Ripple factor 20log10 = 0.5 dB
• Ripple factor 20log10 = 3.0 dB
N g1 g2 g3 g4 g5 g6 g7
1 0.6986 1.00002 1.4029 0.7071 1.98413 1.5963 1.0967 1.5963 1.00004 1.6703 1.1926 2.3661 0.8419 1.98415 1.7058 1.2296 2.5408 1.2296 1.7058 1.00006 1.7254 1.2479 2.6064 1.3137 2.4578 0.8696 1.9841
N g1 g2 g3 g4 g5 g6 g7
1 1.9953 1.00002 3.1013 0.5339 5.80953 3.3487 0.7117 3.3487 1.00004 3.4389 0.7483 4.3471 0.5920 5.80955 3.4817 0.7618 4.5381 0.7618 3.4817 1.00006 3.5045 0.7685 4.6061 0.7929 4.4641 0.6033 5.8095
29
Table for the Maximally-Flat Time Delay LPP Design
N g1 g2 g3 g4 g5 g6 g7 g8 g9 1 2.0000 1.0000 2 1.5774 0.4226 1.0000 3 1.2550 0.5528 0.1922 1.0000 4 1.0598 0.5116 0.3181 0.1104 1.0000 5 0.9303 0.4577 0.3312 0.2090 0.0718 1.0000 6 0.8377 0.4116 0.3158 0.2364 0.1480 0.0505 1.0000
7 0.7677 0.3744 0.2944 0.2378 0.1778 0.1104 0.0375 1.0000
8 0.7125 0.3446 0.2735 0.2297 0.1867 0.1387 0.0855 0.0289 1.0000
30
Example 2.1: Finding the Constants for the LPP Design
CRLjRLCR
RVRCjLjRR
RV
LjR
Vss
RCjR
sRCjR
V 221
1
211
222222
2
22
212
1
CRLRLC
RVRL
sVP
281
sRA VP
42
2222
41
222222
81
22
8
2
2
222222
2
2
1
222
LCR
R
CRLRLC
RVR
V
PP
LR
LCCRL
CRLLCRPs
s
L
A
andThus
Therefore we can compute the power loss ratio as:
[P()]2
R
RVs C
L R jL
RVs 1/jC V1
Consider a simple case of a 2nd order low pass filter:
Extra
31
42
21
42
2222
41 11 2 aaLCCRLP LCR
LR
42422)( 10111 hButterwortLRP
21122 LCa LC
Extra
CRLLC
LCCRLa
R
R2
21
22
41
1 00 2
PLR can be written in terms of polynomial of 2:
For Butterworth response with k = 1, c = 1:
(E1.1)
(E1.2)
Comparing equations (E1.1) and (E1.2):
Setting R = 1 for the Low Pass Prototype (LPP):
1R
CL
CL
LCCLCLLC
0
02
2
22221 4142.12
22 2
C
CLC
4142.1CL
(E1.3)(E1.4)
Thus from equation (E1.4):
Using (E1.3)
Compare this result withN=2 in the table for the LPP Butterworth response.This direct ‘brute force’approach can beextended to N=3, 4, 5…
Compare this result withN=2 in the table for the LPP Butterworth response.This direct ‘brute force’approach can beextended to N=3, 4, 5…
Example 2.1: Finding the Constants for the LPP Design (cont’d)
32
Example 2.1: Verification
Vin Vout
ACAC1
Step=0.01 HzStop=2.0 HzStart=0.01 Hz
AC
CC1C=1.4142 F
LL1
R=L=1.4142 H R
R1R=1 Ohm
RR2R=1 OhmV_AC
SRC1
Freq=freqVac=polar(1,0) V
Extra
33
Example 2.1: Verification (cont’d)
Eqn PLR=PA/PL
Eqn PA=1/8 EqnPL=0.5*mag(Vout)*mag(Vout)
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.80.0 2.0
5.0E3
1.0E4
1.5E4
2.0E4
0.0
2.5E4
freq, Hz
PLR
Extra
m1freq=m1=-3.056
160.0mHz
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.80.0 2.0
-40
-30
-20
-10
0
-50
5
freq, Hz
dB(V
out/0
.5)
160.0m-3.056
m1
–3 dB at 160 mHz (milliHertz!!),which is equivalent to 1 rad/s
Power loss ratioversus frequency
34
Impedance Denormalization and Frequency Transformation of LPP
• Once the LPP filter is designed, the cut-off frequency c can be transformed to other frequencies.
• Furthermore the LPP can be mapped to other filter types such as high pass, bandpass, and bandstop.
• This frequency scaling and transformation entails changing the value and configuration of the elements of the LPP.
• Finally the impedance presented by the filter at the operating frequency can also be scaled, from unity to other values; this is called impedance denormalization.
• Let Zo be the new system impedance value. The following slide summarizes the various transformation from the LPP filter.
35
Impedance Denormalization and Frequency Transformation of LPP (cont’d)
212or 21 o
o 12
c
oLZ
coZ
C
ocLZ1
C
Z
c
o
o
oLZ
ooLZ
oo Z
C
C
Z
o
o
o
oZL
oo ZL1
C
Z
o
o
ooZ
C
LPP to Low Pass
LPP to High Pass
LPP toBandpass
LPP toBandstop
Note that the inductor always multiplies with Zo while the capacitor divides with Zo
Note that the inductor always multiplies with Zo while the capacitor divides with Zo
(2.5a) (2.5b)
L
C
Center frequency Fractional bandwidth
36
Summary of Passive LC Filter Design Flow Using the ILM Method
• Step 1: From the requirements, determine the order and type of approximation functions to use.– Insertion loss (dB) in the passband ?– Attenuation (dB) in the stopband ?– Cut-off rate (dB/decade) in the transition band ?– Tolerable ripple?– Linearity of phase?
• Step 2: Design a normalized low pass prototype (LPP) using the L and C elements.
L1=g2 L2=g4
C1=g1 C2=g3 RL= gN+1
1
|H()|
0
1
1
37
Summary of Passive Filter Design Flow Using the ILM Method (cont’d)
• Step 3: Perform frequency scaling, and denormalize the impedance.
• Step 4: Choose suitable lumped components, or transform the lumped circuit design into distributed realization.
|H()|
0
1
1 2
50
Vs15.916pF
0.1414pF79.58nH
0.7072nH 0.7072nH15.916pF50
RL
All uses the microstrip stripline circuitAll uses the microstrip stripline circuit
38
Filter vs. Impedance Transformation Network
• If we ponder carefully, the sharp observer will notice that the filter can be considered as a class of impedance transformation network.
• In the passband, the load is matched to the source network, much like a filter.
• In the stopband, the load impedance is highly mismatched from the source impedance.
• However, the procedure described here only applies to the case when both load and source impedance are equal and real.
Extra
39
Example 2.2A: LPF Design – Butterworth Response
• Design a 4th order Butterworth low pass filter, Rs = RL= 50 , fc = 1.5 GHz.
L1=0.7654H L2=1.8478H
C1=1.8478F C2=0.7654FRL= 1 g0= 1
L1=4.061 nH L2=9.803 nH
C1=3.921 pF C2=1.624 pFRL= 50 g0=1/50
noRZR
c
no
LZL
co
n
Z
CC
50Z
rad/s 104248.95.12
o
9
GHzc
Steps 1 & 2: LPP
Step 3: Frequency scalingand impedance denormalization
40
• Design a 4th order Chebyshev low pass filter, 0.5 dB ripple factor, Rs = 50 , fc = 1.5 GHz.
Example 2.2B: LPF Design – Chebyshev Response
L1=1.6703H L2=2.3661H
C1=1.1926F C2=0.8419FRL= 1.9841
g0= 1
L1=8.861 nH L2=12.55 nH
C1=2.531 pF C2=1.787 pFRL= 99.2
g0=1/50
noRZR
c
no
LZL
co
n
Z
CC
50Z
rad/s 104248.95.12
o
9
GHzc
Steps 1 & 2: LPP
Step 3: Frequency scalingand impedance denormalization
41
Example 2.2 (cont’d)
0.5 1.0 1.5 2.0 2.50.0 3.0
-20
-10
0
-30
5
freq, GHz
dB(S
(2,1
))dB
(LP
F_b
utte
rwor
th..S
(2,1
))
Chebyshev
Butterworth
|s21|
Ripple is roughly 0.5 dB
0.5 1.0 1.5 2.0 2.50.0 3.0
-300
-250
-200
-150
-100
-50
-350
0
freq, GHz
Pha
se_c
heby
shev
Pha
se_b
utte
rwor
thArg(s21)
Chebyshev
Butterworth
Better phaselinearity for ButterworthLPF in the passband
Computer simulation resultusing AC analysis (ADS2003C)
Eqn Phase_chebyshev = if (phase(S(2,1))<0) then phase(S(2,1)) else (phase(S(2,1))-360)
Note: Equation used in Data Display of ADS2003Cto obtain a continuous phase display with the built-infunction phase( ).
42
Example 2.3: BPF Design
• Design a bandpass filter with Butterworth (maximally flat) response.
• N = 3
• Center frequency fo = 1.5 GHz
• 3 dB Bandwidth = 200 MHz or f1 = 1.4 GHz, f2 = 1.6 GHz
• Impedance = 50 Ω
43
Example 2.3 (cont’d)
• From table, design the low pass prototype (LPP) for 3rd order Butterworth response, c = 1.
Zo=1
g1 1.000F
g3 1.000F
g2 2.000H
g4
12<0o
Hz 1592.0
12
21
c
cc
f
f
Simulated resultusing PSpice
Voltage across g4
Steps 1 & 2: LPP
44
Example 2.3 (cont’d)
• LPP to bandpass transformation
• Impedance denormalization
133.0
497.1
6.12
4.12
12
21
2
1
o
GHzfff
GHz
GHz
o
50
Vs15.916 pF
0.1414 pF79.58 nH
0.7072 nH 0.7072 nH15.916 pF50
RL
o
oLZ
ooLZ
oo Z
C
C
Z
o
o
Step 3: Frequency scalingand impedance denormalization
45
Example 2.3 (cont’d)
• Simulated result using PSpice:
Voltage across RL
46
All Pass Filter
• There is also another class of filter known as the All Pass Filter (APF).
• This type of filter does not produce any attenuation in the magnitude response, but provides phase response in the band of interest.
• APF is often used in conjunction with LPF, BPF, HPF, etc., to compensate for phase distortion.
Extra
Zo BPF APF
f0
|H(f)|
1
f
Arg(H(f))
Example of the APF response
f
|H(f)|
1
0 f
Arg(H(f))
f0
|H(f)|
1
f
Arg(H(f)) Linearphase inpassband
Nonlinearphase in passband
47
Example 2.4: Practical RF BPF Design Using SMD Discrete Components
VARVAR1
Ct_value2=2.9Ct_value=3.5Lt_value=4.8
EqnVar
b82496c3229j000L3param=SIMID 0603-C (2.2 nH +-5%)4_7pF_NPO_0603
C3
b82496c3229j000L2param=SIMID 0603-C (2.2 nH +-5%)
4_7pF_NPO_0603C2
CCt45C=Ct_value2 pF
CCt3C=Ct_value2 pF
S_ParamSP1
Step=1.0 MHzStop=3.0 GHzStart=0.1 GHz
S-PARAMETERS
CPWSUBCPWSub1
Rough=0.0 milTanD=0.02T=1.38 milCond=5.8E+7Mur=1Er=4.6H=62.0 mil
CPWSub
INDQL4
Rdc=0.1 OhmMode=proportional to freqF=800.0 MHzQ=90.0L=15.0 nH
CCt2C=Ct_value pF
CCt1C=Ct_value pF
LLt2
R=L=Lt_value nH
TermTerm2
Z=50 OhmNum=2
LLt1
R=L=Lt_value nH
TermTerm1
Z=50 OhmNum=1
CPWGCPW1
L=28.0 mmG=10.0 milW=50.0 milSubst="CPWSub1"
1_0pF_NPO_0603C1 CPWG
CPW2
L=28.0 mmG=10.0 milW=50.0 milSubst="CPWSub1"
48
Example 2.4 (cont’d)
BPF synthesisusing synthesistool E-synof ADS2003C
49
Example 2.4 (cont’d)
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.80.0 3.0
-100
0
100
-200
200
freq, GHz
phas
e(S(
2,1)
)ph
ase(
RF_
BPF_
mea
sure
d..S
(2,1
))
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.80.0 3.0
-100
0
100
-200
200
freq, GHz
phas
e(S(
2,1)
)ph
ase(
RF_
BPF_
mea
sure
d..S
(2,1
))
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.80.0 3.0
-40
-20
-60
0
freq, GHz
dB(S
(2,1
))dB
(RF_
BPF_
mea
sure
d..S
(2,1
))
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.80.0 3.0
-40
-20
-60
0
freq, GHz
dB(S
(2,1
))dB
(RF_
BPF_
mea
sure
d..S
(2,1
))
|s21|/dB
Arg(s21)/degree
MeasuredSimulated
Measurement is performed with theAgilent 8753ES Vector NetworkAnalyzer, using Full OSL calibration
50
3.0 Microwave Filter Realization Using Stripline Structures
51
3.1 Basic Approach
52
Filter Realization Using Distributed Circuit Elements
• Lumped-element filter realization using surface mounted inductors and capacitors generally works well at lower frequencies (at UHF, say < 3 GHz).
• At higher frequencies, the practical inductors and capacitors loses their intrinsic characteristics.
• Also, a limited range of component values is available from the manufacturer.
• Therefore, for microwave frequencies (> 3 GHz), the passive filter is usually realized using distributed circuit elements such as transmission line sections.
• Here we will focus on stripline microwave circuits.
53
Filter Realization Using Distributed Circuit Elements (cont’d)
• Recall in the study of Terminated Transmission Line Circuit that a length of terminated Tline can be used to approximate an inductor and a capacitor.
• This concept forms the basis of transforming the LC passive filter into distributed circuit elements.
Zo
Zo
Zc ,
l
L
Zc ,
l
C
Zc
,
Zc ,
Zc ,
Zo
Zo
54
Filter Realization Using Distributed Circuit Elements (cont’d)
• This approach is only an approximation. There will be deviation between the actual LC filter response and those implemented with terminated Tline.
• Also, the frequency response of the distributed circuit filter is periodic. • Other issues are shown below.
Zc
,
Zc ,
Zc ,
Zo
Zo
How do we implement a series Tlineconnection? (only practical for certain Tline configuration)
Connection of physicallength cannot beignored at themicrowave region,comparable to
Thus some theorems are used to facilitate the transformation of the LCcircuit into stripline microwave circuits.Chief among these are the Kuroda’sIdentities (See Appendix 1)
Thus some theorems are used to facilitate the transformation of the LCcircuit into stripline microwave circuits.Chief among these are the Kuroda’sIdentities (See Appendix 1)
55
More on Approximating L and C with Terminated Tline: Richard’s Transformation
Zc ,
l
L
jLLjljZZ cin tan
Zin
LZ
l
c tan
(3.1.1a)
Zc ,
l
CZin
jCCjljYY cin tan
CY
l
cZc
1
tan (3.1.1b)
For LPP design, a further requirement isthat:
1tan cl (3.1.1c)82 1tan c
cll
Wavelength atcut-off frequency
Here, instead of fixing Zc and tuning l to approach an L or C,we allow Zc to be a variable too.
56
Example 3.1: LPF Design Using Stripline
• Design a 3rd order Butterworth low pass filter, Rs = RL= 50 , fc = 1.5 GHz.
Steps 1 & 2: LPP
Step 3: Convert to Tlines Zc =
0.500
Zc=
1.00
0
1
Zc=
1.00
0
1
Zo=1 g1 1.000H
g3 1.000H
g2 2.000F
g4
1
Length = c/8for all Tlinesat = 1 rad/s
500.0000.21
57
Example 3.1 (cont’d)
Length = c/8for all Tlinesat = 1 rad/s
Step 4: Add an extra Tline on the series connection and apply Kuroda’s 2nd Identity.
Zc =
0.500Z
c=1.
000
1
Zc=
1.00
01
Zc=1.0
Zc=1.0
Extra TlineExtra Tline
5.02
21 Zn
n2Z1=2
l
Z2=10.11 Z
Similar operation isperformed here
21
1
11
122
ZZ
n
Yc
58
Example 3.1 (cont’d)
Zc =
0.500
1
1
Zc=2.0 Zc=2.0
Zc =
2.000
Zc =
2.000
After applying Kuroda’s 2nd Identity
Length = c/8for all Tlinesat = 1 rad/s Since all Tlines have similar physical
length, this approach to stripline filterimplementation is also known as Commensurate Line Approach.
Since all Tlines have similar physicallength, this approach to stripline filterimplementation is also known as Commensurate Line Approach.
59
Example 3.1 (cont’d)
Zc =
25
50
50
Zc=100 Zc=100
Zc =
100
Zc =
100
Length = c/8for all Tlines atf = fc = 1.5 GHz
Zc/Ω /8 @ 1.5 GHz/mm W/mm 50 13.45 2.8525 12.77 8.00100 14.23 0.61
Microstrip line using double-sided FR4 PCB (r = 4.6, H=1.57 mm)
Step 5: Impedance and frequency denormalization Here we multiply allimpedance with Zo = 50
We can work out the correct width W given theimpedance, dielectric constant, and thickness.From W/H ratio, the effective dielectric constanteff can be determined. Use this together withfrequency at 1.5 GHz to find the wavelength.
We can work out the correct width W given theimpedance, dielectric constant, and thickness.From W/H ratio, the effective dielectric constanteff can be determined. Use this together withfrequency at 1.5 GHz to find the wavelength.
60
Example 3.1 (cont’d)
Step 6: The layout (top view)
61
Example 3.1 (cont’d)
m1freq=m1=-6.092
1.500GHz
0.5 1.0 1.5 2.0 2.5 3.0 3.50.0 4.0
-30
-20
-10
-40
0
freq, GHz
dB(S
(2,1
))
1.500G-6.092
m1
dB(B
utte
r_LP
F_L
C..S
(2,1
))
CC1C=4.244 pF
LL2
R=L=5.305 nH
LL1
R=L=5.305 nH
TermTerm2
Z=50 OhmNum=2
TermTerm1
Z=50 OhmNum=1
Simulated results
MSUBMSub1
Rough=0 milTanD=0.02T=0.036 mmHu=3.9e+034 milCond=1.0E+50Mur=1Er=4.6H=1.57 mm
MSubS_ParamSP1
Step=5 MHzStop=4.0 GHzStart=0.2 GHz
S-PARAMETERS
MTEETee3
W3=8.00 mmW2=0.61 mmW1=0.61 mmSubst="MSub1"
MLOCTL5
L=12.77 mmW=8.0 mmSubst="MSub1"
MTEETee2
W3=0.61 mmW2=2.85 mmW1=0.61 mmSubst="MSub1"
MLOCTL7
L=14.23 mmW=0.61 mmSubst="MSub1"
MTEETee1
W3=0.61 mmW2=0.61 mmW1=2.85 mmSubst="MSub1"
MLINTL2
L=25.0 mmW=2.85 mmSubst="MSub1"
TermTerm2
Z=50 OhmNum=2
MLOCTL6
L=14.23 mmW=0.61 mmSubst="MSub1"
TermTerm1
Z=50 OhmNum=1
MLINTL1
L=25.0 mmW=2.85 mmSubst="MSub1"
MLINTL3
L=14.23 mmW=0.61 mmSubst="MSub1"
MLINTL4
L=14.23 mmW=0.61 mmSubst="MSub1"
62
Conclusions for Section 3.1
• Further tuning is needed to optimize the frequency response.
• The method illustrated is good for the low pass and bandstop filter implementation.
• For high pass and bandpass, other approaches are needed.
63
3.2 Further Implementations
64
Realization of LPF Using the Step-Impedance Approach
• A relatively easy way to implement LPF using stripline components.
• Using alternating sections of high and low characteristic impedance Tlines to approximate the alternating L and C elements in an LPF.
• Performance of this approach is marginal as it is an approximation, where a sharp cutoff is not required.
• As usual, beware of parasitic passbands!!!
65
Equivalent Circuit of a Transmission Line Section
Z11 – Z12 Z11 – Z12
Z12
l
Zc
ljZZZ c cot2211
ljZZZ c cosec2112
oeoeo k
(3.2.1a)
(3.2.1b)
(3.2.1c)
Ideal lossless TlineT-network equivalent circuit
2
2cos2sin2
2sin2
2sin
2cos1
sincos
sin1
1211
tan
22
22
2
2
lc
cc
ll
lc
jZ
jZjZ
jZZZ
ll
l
l
l
Positive reactance
Positivesusceptance
66
Approximation for High and Low ZC
• When l < /2, the series element can be thought of as an inductor and the shunt element can be considered a capacitor.
• For l < /4 and Zc = ZH >> 1:
• For l < /4 and Zc = ZL 1:
2
tan
21211l
ZX
ZZ c l
ZB
Z c sin
11
12
lZX H 0B
0X lZ
BL
1
jX/2
jB
jX/2
X ZH l
B YLlWhen Zc 1l < /4
When Zc >> 1l < /4
Z11 - Z12 Z11 - Z12
Z12
67
Approximation for High and Low ZC (cont’d)
• Note that l < /2 implies a physically short Tline. Thus a short Tline with high Zc (e.g., ZH) approximates an inductor.
• A short Tline with low Zc (e.g., ZL) approximates a capacitor.
• The ratio of ZH/ZL should be as high as possible. Typical values: ZH = 100 to 150 , ZL = 10 to 15 .
H
cL Z
Ll
Lc
CCZ
l
(3.2.2a)
(3.2.2b)
68
Example 3.2: Mapping an LPF Circuit into a Step Impedance Tline Network
• For instance, consider the LPF Design Example 2.2A (Butterworth).
• Let us use the microstrip line. Since a microstrip Tline with low Zc is wide and a Tline with high Zc is narrow, the transformation from circuit to physical layout would be as follows:
L1=4.061 nH L2=9.803 nH
C1=3.921 pF C2=1.624 pFRL= 50 g0=1/50
69
Example 3.2: Physical Realization of LPF
• Using the microstrip line, with r = 4.2, d = 1.5 mm:
• L1 = 4.061 nH, L2 = 9.083 nH, C1 = 3.921 pF, C2 = 1.624 pF
W/d d/mm W/mm e
Zc = 15 10.0 1.5 15.0 3.68Zc = 50 2.0 1.5 3.0 3.21Zc = 110 0.36 1.5 0.6 2.83
19 307.60103356.32 sfk ceLoeLL
19 258.53103356.32 sfk ceHoeHH
70
Example 3.2: Physical Realization of LPF (cont’d)
l2l1
50 line 50 line
l4l3
0.6 mm15.0 mm
3.0 mm
To 50Load
mmZ
Ll
HH
c 5.611
mmZC
lL
Lc 2.912
mml 0.153
mml 8.34
Verification:
7854.0490.042 lL
7854.0202.044 lL
7854.0905.043 lH
7854.0392.041 lH
Nevertheless we stillproceed with the imple-mentation. It will be seenthat this will affect the accuracy of the –3 dB cut-offpoint of the filter.
71
Example 3.2: Step Impedance LPF Simulation with ADS Software
• Transferring the microstrip line design to ADS:
Microstrip line model
Microstrip step junctionmodel
Microstrip line substrate model
72
Example 3.2: Step Impedance LPF Simulation with ADS Software (cont’d)
m1freq=1.410GHzdB(S(2,1))=-3.051
0.5 1.0 1.5 2.0 2.5 3.0 3.50.0 4.0
-20
-15
-10
-5
-25
0
freq, GHz
dB(S
(2,1
))
m1
73
Example 3.2: Step Impedance LPF Simulation with ADS Software (cont’d)
• However if we extent the stop frequency for the S-parameter simulation to 9 GHz...
m1freq=1.410GHzdB(S(2,1))=-3.051
1 2 3 4 5 6 7 80 9
-15
-5
-25
0
freq, GHz
dB(S
(2,1
))
m1
Parasitic passbands,artifacts due to usingtransmission lines
74
Example 3.2: Verification with Measurement
The –3 dB point is around 1.417 GHz!
The actual LPF constructed in year 2000. The Agilent 8720D Vector Network Analyzer is used to perform the S-parameter measurement.
75
Example 3.3: Realization of BPF Using a Coupled Microstrip Line
• Based on the BPF design of Example 2.3:
50
Vs15.916 pF
0.1414 pF79.58 nH
0.7072 nH 0.7072 nH15.916 pF50 RL
J1
–90o
J2
–90o
J3
–90o
J4
–90o
4o
TlineAdmittanceinverter
To RL
To sourcenetwork
See appendix (using Richard’s transformationand Kuroda’s identities)
An array of coupledmicrostrip line
4o
o = wavelength at oo = wavelength at o
Section 1 Section 2 Section 3Section 4
An equivalent circuit model for coupled Tlineswith open circuit attwo ends.
Extra
76
Example 3.3: Realization of BPF Using the Coupled Microstrip Line (cont’d)
• Each section of the coupled stripline contains three parameters: S, W, d. These parameters can be determined from the values of the odd and even mode impedances (Zoo & Zoe) of each coupled line.
• Zoo and Zoe are in turn depend on the “gain” of the corresponding admittance inverter J.
• And each Jn is given by:
SW W
d
2
2
1
1
ooooo
ooooe
JZJZZZ
JZJZZZ
4,3,2for
1
1
1
21
1
21
21
1
NNo
nno
o
ggZN
ggZn
gZ
J
NnJ
J
Extra
133.0
497.1
6.12
4.12
12
21
2
1
o
GHzfff
GHz
GHz
o
From Example 2.3
77
Example 3.3: Realization of BPF Using the Coupled Microstrip Line (cont’d)
009163.012
11
gZoJ
002969.021
21
2 ggZo
J
002969.032
21
3 ggZo
J
009163.0432
14
ggZoJ
588.371
403.831
2111
2111
ooooo
ooooe
ZJZJZZ
ZJZJZZSection 1:
Section 2:
Section 3:
Section 4:
680.431
523.581
2222
2222
ooooo
ooooe
ZJZJZZ
ZJZJZZ
680.43
523.58
4
4
oo
oe
Z
Z
588.37
403.83
3
3
oo
oe
Z
Z
Note:g1=1.0000g2=2.0000g3=1.0000g4=1.0000
Note:g1=1.0000g2=2.0000g3=1.0000g4=1.0000
Extra
78
Example 3.3: Realization of BPF Using the Coupled Microstrip Line (cont’d)
• In this example, an edge-coupled microstrip line is used to implement the coupled transmission line structures needed in the BPF. Stripline does not suffer from dispersion and its propagation mode is pure TEM mode, however it is more difficult to implement physically due to the fact that the trace is buried within the dielectric.
• Design equations for coupled microstrip line implemented are widely tabulated.
• Here we will use FR4 (r = 4.6, r = 1.0) substrate with 1.0mm dielectric thickness, and 1 ounce copper (about 36m thick) copper laminate. The conductivity of copper is around 5.8107 Siemens/meter.
• Furthermore we will use the LineCal tool in Advanced Design System to work out the dimensions needed for the coupled microstrip line.
Extra
79
Example 3.3: Realization of BPF Using the Coupled Microstrip Line (cont’d)
• Using the LineCal tool to work out the dimensions for sections 1 and 3.
Electrical length (l ), 90o for quarterwavelength.
Zoe
Zoo
Zo
Voltage couplingfactor in dB
Fix the frequency at1.5GHz, the centerof passband
Strategy:1) We ‘tune’ the W andS for the specifiedZoo and Zoe. 2) Then we ‘tune’ the length Lto meet the electrical length of/2 (quarter wavelength) at 1.5GHz.
Extra
80
Example 3.3: Realization of BPF Using the Coupled Microstrip Line (cont’d)
• Using the LineCal tool to work out the dimensions for sections 2 and 4.
Extra
81
Example 3.3: Realization of BPF Using the Coupled Microstrip Line (cont’d)
• Alternatively we can implement our own design tool, as shown below implemented on Microsoft Excel.
Based on design equations from Garg R., Bahl I. J.,”Characteristics of coupled microstriplines”, IEEETransaction on Microwave Theory and Techniques, MTT-27, No.7, pp. 700-705,July 1979.
Strategy:1) We ‘tune’ the W andS for the specified Zoo and Zoe. 2) Based on the width W of a single trace, we work out the effective permittivity, and use this to calculate the phase velocity. 3) From this we find the wavelength at 1.5GHz and work out the required quarter wavelength.
Extra
82
Example 3.3: Coupled-Line BPF Simulation with ADS Software
• Using ideal transmission line elements:
Ideal open circuit
Ideal coupled tline
Extra
83
Example 3.3: Coupled-Line BPF Simulation with ADS Software (cont’d)
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.51.0 10.0
0.2
0.4
0.6
0.8
0.0
1.0
freq, GHz
mag(S
(2,1
))
Parasitic passbands. Artifacts due to using distributed elements, these are not present if lumped components are used.
2fo
Extra
84
Example 3.3: Coupled-Line BPF Simulation With ADS Software (cont’d)
• Using a practical stripline model:
Coupled stripline model
Open circuitmodel
Stripline substrate model
Extra
85
Example 3.3: Coupled-Line BPF Simulation with ADS Software (cont’d)
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.51.0 10.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0.0
1.0
freq, GHz
mag(S
(2,1
))
Attenuation due to losses in the conductor and dielectric
Extra
86
Items for Self-Study
• Network analysis and realizability theory
• Synthesis of terminated RLCM 1-port circuits
• Ideal impedance and admittance inverters and practical implementation
• Periodic structures theory
• Filter design by the Image Parameter Method (IPM).
87
Other Types of Stripline Filter• LPF
• HPF:
SMD capacitor
BPF:
88
Other Types of Stripline Filter (cont’d)
• More BPFs:
• BSF:
89
Appendix 1 – Kuroda’s Identities
90
Kuroda’s Identities
• As extracted from Ref. [2]
122 1
ZZ
n
Z1
l
21
Z Z2/n2
l
nZ1
Z2
l
221
Znn2Z1
l
1Z
Z2
l
21
n
ZZ2/n2
l
1Z
1: n2
Z1
l
221
Znn2Z1
l
21
Z
n2: 1
Note: The inductor represents the shorted Tline while the capacitorrepresents the open-circuit Tline.
Note: Thelength of alltransmissionlines isl = /8
Note: Thelength of alltransmissionlines isl = /8
91
References
[1] R. E. Collin, “Foundations for Microwave Engineering”, 2nd Edition 1992, McGraw-Hill.
[2] D. M. Pozar, “Microwave Engineering”, 2nd Edition 1998, John-Wiley & Sons.* (3rd Edition 2005, John-Wiley & Sons is now available)
Other more advanced references:
[3] W. Chen (Editor), “The Circuits and Filters Handbook”, 1995, CRC Press.*
[4] I. Hunter, “Theory and Design of Microwave Filters”, 2001, The Institution of Electrical Engineers.*
[5] G. Matthaei, L. Young, E.M.T. Jones, “Microwave Filters, Impedance-Matching Networks, and Coupling Structures”, 1980, Artech House.*
[6] F. F. Kuo, “Network Analysis and Synthesis”, 2nd Edition 1966, John-Wiley & Sons.
* Recommended* Recommended