This action might not be possible to undo. Are you sure you want to continue?
https://www.scribd.com/doc/51104038/CHP4MICROWAVEFILTERSWITHEXAMPLES
03/08/2013
text
original
EKT 441
MICROWAVE COMMUNICATIONS
CHAPTER 4:
MICROWAVE FILTERS
2
INTRODUCTION
What is a Microwave filter ?
linear 2port network
controls the frequency response at a certain point in
a microwave system
provides perfect transmission of signal for
frequencies in a certain passband region
infinite attenuation for frequencies in the stopband
region
a linear phase response in the passband (to reduce
signal distortion).
f2
3
INTRODUCTION
The goal of filter design is to approximate the ideal
requirements within acceptable tolerance with
circuits or systems consisting of real components.
f1
f3
f2
Commonly used block Diagram of a Filter
4
INTRODUCTION
Why Use Filters?
RF signals consist of:
1. Desired signals ± at desired frequencies
2. Unwanted Signals (Noise) ± at unwanted
frequencies
That is why filters have two very important
bands/regions:
1. Pass Band ± frequency range of filter where it
passes all signals
2. Stop Band ± frequency range of filter where it
rejects all signals
5
INTRODUCTION
Categorization of Filters
Lowpass filter (LPF), Highpass 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 active filter, there can be amplification of the of the signal
power in the passband region, passive filter do not provide
power amplification in the passband.
Filter 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, and also mechanical resonators etc.).
Active filter may contain transistor, FET and Opamp.
Filter
LPF BPF HPF
Active Passive Active Passive
6
INTRODUCTION
Types of Filters
1. Lowpass Filter
Passes low freq
Rejects high freq
f1
f2
f1
2. Highpass Filter
Passes high freq
Rejects low freq
f1
f2
f2
7
INTRODUCTION
3. Bandpass Filter
Passes a small range of
freq
Rejects all other freq
4.Bandstop Filter
Rejects a small range of
freq
Passes all other freq
f1
f3
f2
f1
f3
f1
f2 f2
f3
8
INTRODUCTION
Filter Parameters
Pass bandwidth; BW(3dB) = f
u(3dB)
± f
l(3dB)
Stop band attenuation and frequencies,
Ripple difference between max and min of
amplitude response in passband
Input and output impedances
Return loss
Insertion loss
Group Delay, quality factor
9
INTRODUCTION
Lowpass filter (passive).
A Filter
H(æ)
V
1
(æ) V
2
(æ)
Z
L
A(æ)/dB
æ
0
æ
c
3
10
20
30
40
50
´ )
´ )
¦
¦
'
+
'
=
æ
æ
1
2
10
20 A
V
V
Log n Attenuatio
(1.1b)
´ )
´ )
´ ) æ
æ
æ
1
2
V
V
H =
(1.1a)
æ
c
H(æ)
æ
1
Transfer
function
Arg(H(æ))
æ
10
INTRODUCTION
For impedance matched system, using s
21
to observe the filter response
is more convenient, as this can be easily measured using Vector
Network Analyzer (VNA).
Z
c
0
1
2
21
0
1
1
11
2 2
= =
= =
a a
a
b
s
a
b
s
Transmission line
is optional
æ
c
20logs
21
(æ)
æ
0dB
Arg(s
21
(æ))
æ
Filter
Z
c
Z
c
Z
c
V
s
a
1
b
2
Complex value
11
INTRODUCTION
A(æ)/dB
æ
0
æ
c
3
10
20
30
40
50
A Filter
H(æ)
V
1
(æ) V
2
(æ)
Z
L
Passband
Stopband
Transition band
Cutoff frequency (3dB)
Low pass filter response (cont)
12
INTRODUCTION
High Pass filter
A(æ)/dB
æ
0
æ
c
3
10
20
30
40
50
æ
c
H(æ)
æ
1
Transfer
function
Stopband
Passband
13
INTRODUCTION
Bandpass filter (passive). Bandstop filter.
æ
A(æ)/dB
40
æ
1
3
30
20
10
0
æ
2
æ
o
æ
1
H(æ)
æ
1 Transfer
function
æ
2
æ
o
æ
A(æ)/dB
40
æ
1
3
30
20
10
0
æ
2
æ
o
æ
1
H(æ)
æ
1
Transfer
function
æ
2
æ
o
14
INTRODUCTION
6 8 0
r ( )
Filter Response
 50
 0
 0
 0
 0
0
.
 .00 8 7.90
 .0057
I t t r L
I rt i L
Figure 4.1: A 10 GHz Parallel Coupled Filter Response
Pass BW (3dB)
Stop band frequencies and attenuation
Q factor
Insertion Loss
15
FILTER DESIGN METHODS
Filter Design Methods
Two types of commonly used design methods:
 Image Parameter Method
 Insertion Loss Method
Image parameter method yields a usable filter
However, no clearcut way to improve the design i.e to control the
filter response
16
FILTER DESIGN METHODS
Filter Design Methods
The insertion loss method (ILM) allows a systematic way to design
and synthesize a filter with various frequency response.
ILM method also allows filter performance to be improved in a
straightforward manner, at the expense of a µhigher order¶ filter.
A rational polynomial function is used to approximate the ideal H(æ),
A(æ) or s21(æ).
Phase information is totally ignored.Ignoring phase simplified the
actual synthesis method. An LC network is then derived that will
produce this approximated response.
Here we will use A(æ) following [2]. The attenuation A(æ) can be cast
into power attenuation ratio, called the Power Loss Ratio, P
LR
, which
is related to A(æ)
2
.
17
FILTER DESIGN METHODS
´ )
´ )
2
1
1
1
2
1
1
Load to delivered Power
network source from available Power
æ
æ
I
¦
¦
¦
I
= = =
=
A
P
A
P
Load
P
inc
P
LR
P
P
LR
large, high attenuation
P
LR
close to 1, low attenuation
For example, a lowpass
filter response is shown
below:
P
LR
large, high attenuation
P
LR
close to 1, low attenuation
For example, a lowpass
filter response is shown
below:
Z
L V
s
Lossless
2port network
I
1
´æ)
Z
s
P
A
P
in
P
L
P
LR
(f)
LowPass filter P
LR
f
1
Low
attenuation
High
attenuation
f
c
(2.1a)
18
P
LR
and s
21
In terms of incident and reflected waves, assuming Z
L
=Z
s
= Z
C
.
Z
c V
s
Lossless
2port network
Z
c
P
A
P
in
P
L
a
1
b
1
b
2
2
21
1
2
2
1
2
2
2
1
2
1
2
1
s
b
a
b
a
!
! ! !
(2.1b)
19
FILTER RESPONSES
Filter Responses
Several types filter responses:
 Maximally flat (Butterworth)
 Equal Ripple (Chebyshev)
 Elliptic Function
 Linear Phase
20
THE INSERTION LOSS METHOD
Practical filter response:
Maximally flat:
 also called the binomial or Butterworth response,
 is optimum in the sense that it provides the flattest possible
passband response for a given filter complexity.
 no ripple is permitted in its attenuation profile
N
c
LR
k P
¦
¦
'
+
'
+ =
æ
æ
2
1
[ .1 ]
æ ± frequency of filter
æ
c
± cutoff frequency of filter
N ± order of filter
21
THE INSERTION LOSS METHOD
Equal ripple
 also known as Chebyshev.
 sharper cutoff
 the passband response will have ripples of amplitude 1 +k
2
¦
¦
'
+
'
!
c
N
T k
[
[
2 2
1 [ .11]
æ ± frequency of filter
æ
c
± cutoff frequency of filter
N ± order of filter
22
THE INSERTION LOSS METHOD
Figure .3: Maximally flat and equalripple low pass filter response.
23
THE INSERTION LOSS METHOD
Elliptic function:
 have equal ripple responses in the passband and
stopband.
 maximum attenuation in the passband.
 minimum attenuation in the stopband.
Linear phase:
 linear phase characteristic in the passband
 to avoid signal distortion
 maximally flat function for the group delay.
24
THE INSERTION LOSS METHOD
Figure .4: Elliptic function lowpass filter response
25
THE INSERTION LOSS METHOD
Filter
Specification
Lowpass
Prototype
Design
Scaling &
Conversion
Filter
Implementation
Optimization
& Tuning
Normally done using
simulators
Figure . : The process of the filter design by the insertion
loss method.
26
THE INSERTION LOSS METHOD
Figure . : Low pass filter prototype, N = 2
Low Pass Filter Prototype
27
THE INSERTION LOSS METHOD
Figure . : Ladder circuit for low pass filter prototypes and their
element definitions. (a) begin with shunt element. (b) begin with
series element.
Low Pass Filter Prototype ± Ladder Circuit
28
THE INSERTION LOSS METHOD
g = generator resistance, generator conductance.
g
k
= inductance for series inductors, capacitance for shunt
capacitors.
(k=1 to N)
g
N+1
= load resistance if g
N
is a shunt capacitor, load
conductance if g
N
is a series inductor.
As a matter of practical design procedure, it will be
necessary to determine the size, or order of the filter. This is
usually dictated by a specification on the insertion loss at
some frequency in the stopband of the filter.
29
THE INSERTION LOSS METHOD
Figure 4.8: Attenuation versus normalized frequency for maximally flat
filter prototypes.
Low Pass Filter Prototype ± Maximally Flat
30
THE INSERTION LOSS METHOD
Figure 4.9: Element values for maximally flat LPF prototypes
31
THE INSERTION LOSS METHOD
´ ) [
2 2
1
N
T k !
For an equal ripple low pass filter with a cutoff frequency Ȧ
c
=
1, The power loss ratio is:
´ )

´

!
1
0
[
N
T
[ .12]
Where 1 + k
2
is the ripple level in the passband. Since the
Chebyshev polynomials have the property that
[ .12] shows that the filter will have a unity power loss ratio at
Ȧ = for N odd, but the power loss ratio of 1 + k
2
at Ȧ = for N
even.
Low Pass Filter Prototype ± Equal Ripple
32
THE INSERTION LOSS METHOD
Figure 4.10: Attenuation versus normalized frequency for equalripple filter
prototypes. (0.5 dB ripple level)
33
THE INSERTION LOSS METHOD
Figure 4.11: Element values for equal ripple LPF prototypes (0.5 dB ripple
level)
34
THE INSERTION LOSS METHOD
Figure 4.12: Attenuation versus normalized frequency for equalripple filter
prototypes (3.0 dB ripple level)
35
THE INSERTION LOSS METHOD
Figure 4.13: Element values for equal ripple LPF prototypes (3.0 dB ripple
level).
36
FILTER TRANSFORMATIONS
L L
s
R R R
R R
R
C
C
L R L
0
'
0
'
0
'
0
'
=
=
=
=
[ .13a]
[ .13b]
[ .13c]
[ .13d]
Low Pass Filter Prototype ± Impedance Scaling
37
FILTER TRANSFORMATIONS
'
'
k k
c
k
k k
c
k
C j C j jB
L j L j jX
æ
æ
æ
æ
æ
æ
= =
= =
The new element values of the prototype filter:
c
[
[
[ n
Frequency scaling for the low pass filter:
[ .14]
[ .1 a]
[ .1 b]
38
FILTER TRANSFORMATIONS
The new element values are given by:
c
k k
k
c
k k
k
[ [
[ [
0
0
! !
! !
[ .1 a]
[ .1 b]
39
FILTER TRANSFORMATIONS
[
[
[
c
n
Low ass to high ass tra sformation
The frequency substitution:
k c
k
k c
k
[
[
0
0
1
!
!
The new component values are given by:
[ .1 ]
[ .1 a]
[ .1 b]
40
BANDPASS & BANDSTOP
TRANSFORMATIONS
¦
¦
'
+
'
(
!
¦
¦
'
+
'
n
[
[
[
[
[
[
[
[
[ [
[
[
0
0
0
0 1 2
0
1
0
1 2
[
[ [
! (
2 1 0
[ [ [ !
Where,
The center frequency is:
[ .1 ]
[ .2 ]
[ .21]
Low ass to Bandpass transformation
41
BANDPASS & BANDSTOP
TRANSFORMATIONS
k
k
k
k
0
0
[
[
(
!
(
!
The series inductor, L
k
, is transformed to a series LC circuit with
element values:
The shunt capacitor, C
k
, is transformed to a shunt LC circuit with
element values:
0
0
[
[
(
!
(
!
k
k
k
k
[ .22a]
[ .22b]
[ .23a]
[ .23b]
42
BANDPASS & BANDSTOP
TRANSFORMATIONS
1
0
0
¦
¦
'
+
'
A ÷
æ
æ
æ
æ
æ
0
1 2
æ
æ æ
! A
2 1 0
æ æ æ !
Where,
The center frequency is:
[ .24]
Low pass to Bandstop transformation
43
BANDPASS & BANDSTOP
TRANSFORMATIONS
k
k
k
k
(
!
(
!
0
0
1
[
[
The series inductor, L
k
, is transformed to a parallel LC circuit with
element values:
The shunt capacitor, C
k
, is transformed to a series LC circuit with
element values:
0
0
1
[
[
k
k
k
k
(
!
(
!
[ .2 a]
[ .2 b]
[ .2 a]
[ .2 b]
44
BANDPASS & BANDSTOP
TRANSFORMATIONS
45
EXAMPLE .1
Design a maximally flat low pass filter with a cutoff
freq of 2 GHz, impedance of , and at least 1 dB
insertion loss at 3 GHz. Compute and compare with
an equalripple (3. dB ripple) having the same order.
46
EXAMPLE .1 (Cont)
Solution:
First find the order of the maximally flat filter to satisfy the
insertion loss specification at 3 GHz.
618 . 0
618 . 1
0 . 2
618 . 1
618 . 0
5
4
3
2
1
=
=
=
=
=
g
g
g
g
g
We can find the normalized freq by using:
5 . 0 1
2
3
1 = =
c
æ
æ
47
EXAMPLE .1 (Cont)
The ladder diagram of the LPF prototype to be used is as follow:
L L
s
L R L
0
0
0
0
!
!
!
!
c
R
g
C
æ
0
1
1
=
c
g R
L
[
2 0
2
!
c
R
g
C
æ
0
3
3
=
c
g R
L
æ
4 0
4
=
c
R
g
C
æ
0
5
5
=
48
EXAMPLE .1 (Cont)
´ )´ )
984 . 0
10 2 2 50
618 . 0
9
0
1
1
=
 
= =
z æ
c
R
g
C
´ )
438 . 6
10 2 2
618 . 1 50
9
2 0
2
=
 

= =
z æ
c
g R
L
´ )´ )
183 . 3
10 2 2 50
00 . 2
9
0
3
3
=
 
= =
z æ
c
R
g
C
´ )
438 . 6
10 2 2
618 . 1 50
9
4 0
4
!
v v
v
! !
T [
c
g R
L
´ )´ )
984 . 0
10 2 2 50
618 . 0
9
0
5
5
=
 
= =
z æ
c
R
g
C
pF
nH
pF
nH
pF
LPF prototype for maximally flat filter
49
EXAMPLE .1 (Cont)
´ )´ )
541 . 5
10 2 2 50
4817 . 3
9
0
1
1
=
 
= =
z æ
c
R
g
C
´ )
031 . 3
10 2 2
7618 . 0 50
9
2 0
2
=
 

= =
z æ
c
g R
L
´ )´ )
223 . 7
10 2 2 50
5381 . 4
9
0
3
3
!
v v
! !
T [
c
R
g
´ )
031 . 3
10 2 2
7618 . 0 50
9
4 0
4
=
 

= =
z æ
c
g R
L
´ )´ )
541 . 5
10 2 2 50
4817 . 3
9
0
5
5
!
v v
! !
T [
c
R
g
pF
nH
pF
nH
pF
4817 . 3
7618 . 0
5381 . 4
7618 . 0
4817 . 3
5
4
3
2
1
=
=
=
=
=
g
g
g
g
g
LPF prototype for equal ripple filter:
50
THE INSERTION LOSS METHOD
Filter
Specification
Lowpass
Prototype
Design
Scaling &
Conversion
Filter
Implementation
Optimization
& Tuning
Normally done using
simulators
51
SUMMARY OF STEPS IN FILTER
DESIGN
A. Filter Specification
1. Max Flat/Equal Ripple,
2. If equal ripple, how much pass band ripple allowed? If max
flat filter is to be designed, cont to next step
3. Low Pass/High Pass/Band Pass/Band Stop
4. Desired freq of operation
. Pass band & stop band range
. Max allowed attenuation (for Equal Ripple)
52
SUMMARY OF STEPS IN FILTER
DESIGN (cont)
B. Low Pass Prototype Design
1. Min Insertion Loss level, No of Filter
Order/Elements by using IL values
2. Determine whether shunt cap model or series
inductance model to use
3. Draw the lowpass prototype ladder diagram
4. Determine elements¶ values from Prototype Table
53
SUMMARY OF STEPS IN FILTER
DESIGN (cont)
C. Scaling and Conversion
1. Determine whether if any modification to the
prototype table is required (for high pass, band
pass and band stop)
2. Scale elements to obtain the real element values
54
SUMMARY OF STEPS IN FILTER
DESIGN (cont)
D. Filter Implementation
1. Put in the elements and values calculated from
the previous step
2. Implement the lumped element filter onto a
simulator to get the attenuation vs frequency
response
55
EXAMPLE .2
Design a band pass filter having a 0.5 dB
equalripple response, with N = 3. The center
frequency is 1 GHz, the bandwidth is 10%,
and the impedance is 50 .
56
EXAMPLE .2 (Cont)
Solution: The low pass filter (LPF) prototype ladder diagram is
shown as follow:
A = 0.1 N = 3 æ = 1 GHz
R
S
L
1
L
3
C
2
R
L
57
EXAMPLE .2 (Cont)
From the equal ripple filter table (with 0.5 dB ripple), the filter
elements are as follow;
L R g
L g
g
L g
! !
! !
! !
! !
000 . 1
5963 . 1
0967 . 1
5963 . 1
4
3
3
2
1
2
1
58
EXAMPLE .2 (Cont)
Transforming the LPF prototype to the BPF prototype
R
S
R
L
C
1
C
2
C
3
L
1
L
2
L
3
59
EXAMPLE .2 (Cont)
´ )´ )
pF
L Z
C 199 . 0
5963 . 1 10 2 2 50
1 . 0
9
1
0 0
1
!
v v v
!
(
!
T [
´ )
nH
Z L
L 0 . 127
1 . 0 10 1 2
50 5963 . 1
9
0
1
1
0
=
  

=
A
=
z æ
´ ) ´ )
p
Z
C
C 91 . 34
50 ) 1 . 0 ( 10 1 2
0967 . 1
9
0
0
2
2
=
 
=
A
=
z æ
´ )
n
C
Z
L 726 . 0
0967 . 1 10 1 2
50 1 . 0
9
2
0
0
2
!
v v v
v
!
(
!
T [
60
EXAMPLE .2 (Cont)
´ )´ )
p
L Z
C 199 . 0
5963 . 1 10 2 2 50
1 . 0
9
3
0 0
3
=
  
=
A
=
z æ
´ )
nH
Z L
L 0 . 127
1 . 0 10 1 2
50 5963 . 1
9
0
3 0
3 =
  

=
A
=
z æ
61
EXAMPLE .3
Design a fivesection high pass lumped
element filter with 3 dB equalripple
response, a cutoff frequency of 1 GHz, and
an impedance of 50 . What is the resulting
attenuation at 0.6 GHz?
62
EXAMPLE .3 (Cont)
Solution: The high pass filter (HPF) prototype ladder diagram is
shown as follow:
N = 5 æ = 1 GHz
At æ
c
= 0.6 GHz, ; referring back to Fig 4.12
The attenuation for N = 5, is about 41 dB
R
S
R
L
L
1
L
3
L
5
C
2
C
3
667 . 0 1
6 . 0
1
1 = =
c
æ
æ
63
EXAMPLE .3 (Cont)
From the equal ripple filter table (with 3.0 dB ripple), the filter
elements are as follow;
L R g
L g
C g
L g
C g
L g
! !
! !
! !
! !
! !
! !
000 . 1
4817 . 3
7618 . 0
5381 . 4
7618 . 0
4817 . 3
6
5
5
4
4
3
3
2
1
2
1
64
EXAMPLE .3 (Cont)
´ )´ )
p
C Z
C
c
18 . 4
7618 . 0 10 1 2 50
1 1
'
9
2
0
2 =
  
= =
z æ
´ )
nH
L
Z
L
c
28 . 2
4817 . 3 10 1 2
50
'
9
1
1
0
=
  
= =
z æ
´ )
nH
L
Z
L
c
754 . 1
5381 . 4 10 1 2
50
'
9
3
0
3
=
  
= =
z æ
Impedance and frequency scaling:
65
EXAMPLE .3 (Cont)
´ )´ )
pF
C Z
C
c
18 . 4
7618 . 0 10 1 2 50
1 1
'
9
4
0
4 !
v v v
! !
T [
´ )
nH
L
Z
L
c
754 . 1
5381 . 4 10 1 2
50
'
9
5
0
5
=
  
= =
z æ
66
EXAMPLE .4
Design a 4th order Butterworth LowPass Filter. R
s
= R
L
= Ohm, f
c
=
1. GHz.
L
1
=0.7654H L
2
=1.8478H
C
1
=1.8478F C
2
=0.7654F
R
L
= 1 g
0
= 1
L
1
=4.061nH L
2
=9.803nH
C
1
=3.921pF
C
2
=1.624pF
R
L
= 50 g
0
=1/50
n o
R Z R =
c
n
o
L
Z L
[
=
c o
n
Z
C
C
[
=
´ )
50 Z
rad/s 10 4248 . 9 5 . 1 2
o
9
!
v ! ! GHz
c
T [
Step 1&2: LPP
Step 3: Frequency scaling
and impedance denormalization
67
EXAMPLE .
Design a 4th order Chebyshev LowPass Filter, . dB ripple factor. R
s
= Ohm, f
c
= 1. GHz.
L
1
=1.6703H L
2
=2.3661H
C
1
=1.1926F C
2
=0.8419F
R
L
=
1.9841
g
0
= 1
L
1
=8.861nH L
2
=12.55nH
C
1
=2.531pF
C
2
=1.787pF
R
L
=
99.2
g
0
=1/50
n o
R Z R =
c
n
o
L
Z L
[
=
c o
n
Z
C
C
[
=
´ )
50 Z
rad/s 10 4248 . 9 5 . 1 2
o
9
!
v ! ! GHz
c
T [
Step 1&2: LPP
Step 3: Frequency scaling
and impedance denormalization
68
EXAMPLE .
Design a bandpass filter with Butterworth (maximally flat)
response.
N = 3.
Center frequency f
o
= 1. GHz.
3dB Bandwidth = 2 MHz or f
1
=1.4GHz, f
2
=1. GHz.
69
EXAMPLE . (cont)
From table, design the LowPass prototype (LPP) for 3rd order
Butterworth response, æ
c
=1.
Z
o
=1
g
1
1.000F
g
3
1.000F
g
2
2.000H
g
4
1
2<0
o
Hz 1592 . 0
1 2
2
1
! !
! !
T
T [
c
c c
f
f
Step 1&2: LPP
70
EXAMPLE . (cont)
LPP to bandpass transformation.
Impedance denormalization.
´ )
´ )
133 . 0
497 . 1
6 . 1 2
4 . 1 2
1 2
2 1
2
1
! ! (
! !
!
!
o
GHz f f f
GHz
GHz
o
[
[ [
T [
T [
50
V
s
15.916pF
0.1414pF 79.58nH
0.7072nH 0.7072nH
15.916pF
50
R
L
(
o
o
LZ
[
o o
LZ [
A
o o
Z
C
A [
C
Z
o
o
[
A
Step 3: Frequency scaling
and impedance denormalization
INTRODUCTION
What is a Microwave filter ?
linear 2port network controls the frequency response at a certain point in a microwave system provides perfect transmission of signal for frequencies in a certain passband region infinite attenuation for frequencies in the stopband region a linear phase response in the passband (to reduce f2 signal distortion).
2
INTRODUCTION
The goal of filter design is to approximate the ideal requirements within acceptable tolerance with circuits or systems consisting of real components.
f1 f2
f3
Commonly used block Diagram of a Filter
3
Pass Band ± frequency range of filter where it passes all signals 2. Stop Band ± frequency range of filter where it rejects all signals 4 . Desired signals ± at desired frequencies 2.INTRODUCTION Why Use Filters? RF signals consist of: 1. Unwanted Signals (Noise) ± at unwanted frequencies That is why filters have two very important bands/regions: 1.
transmission line sections and resonating structures (e. Active filter may contain transistor. passive filter do not provide power amplification in the passband.INTRODUCTION Categorization of Filters Lowpass filter (LPF). Bandpass filter (BPF). Bandstop filter (BSF). Highpass filter (HPF). piezoelectric crystal. inductors. the filter can be further divided into active and passive types. capacitors. Surface Acoustic Wave (SAW) devices. FET and Opamp.). Filter used in electronics can be constructed from resistors. Filter LPF Active Passive HPF Active BPF Passive 5 .g. and also mechanical resonators etc. there can be amplification of the of the signal power in the passband region. In each category. In active filter. arbitrary type etc.
Lowpass Filter f1 f1 2. Highpass Filter f1 f2 f2 f2 Passes low freq Rejects high freq Passes high freq Rejects low freq 6 .INTRODUCTION Types of Filters 1.
INTRODUCTION 3. f1 f2 f3 Bandpass Filter f2 4.Bandstop f1 f2 f3 Filter f1 f3 Passes a small range of freq Rejects all other freq Rejects a small range of freq Passes all other freq 7 .
quality factor 8 . Ripple difference between max and min of amplitude response in passband Input and output impedances Return loss Insertion loss Group Delay.INTRODUCTION Filter Parameters Pass bandwidth. BW(3dB) = fu(3dB) ± fl(3dB) Stop band attenuation and frequencies.
INTRODUCTION Lowpass filter (passive). 1 V1([) A Filter H([) V2([) H([) Transfer function [ V .
[ H .
1a) [ V1 . ! 2 (1.
[ ZL Arg(H([)) [c A([)/dB 50 40 30 20 10 3 0 [ ¨ V2 .
[ ¸ ¹ Attenuation A ! 20 Log10 © © V .
1b) 9 .[ ¹ ª 1 º [c [ (1.
as this can be easily measured using Vector Network Analyzer (VNA).INTRODUCTION For impedance matched system. a1 Vs Zc Zc b2 Zc Zc Transmission line is optional b b s11 ! 1 s21 ! 2 a1 a !0 a1 a !0 2 2 Filter Arg(s21([)) 20logs21([) 0dB [c [ [ Complex value 10 . using s21 to observe the filter response is more convenient.
INTRODUCTION Low pass filter response (cont) A([)/dB Passband 50 40 30 20 10 3 0 Transition band Stopband [c [ Cutoff frequency (3dB) V1([) A Filter H([) V2([) ZL 11 .
INTRODUCTION High Pass filter H([) 1 Transfer function 50 40 30 20 10 3 0 A([)/dB Passband [ [c [c [ Stopband 12 .
INTRODUCTION
Bandpass filter (passive).
A([)/dB 40 30 20 10 3 0 [ 40 30 20 10 3 0
Bandstop filter.
A([)/dB
[1 [o [2
[1 H([)
[o
[2
[
H([) 1 Transfer function 1
Transfer function
[ [1 [o [2 [1 [o [2
[
13
INTRODUCTION
Pass BW (3dB)
F ilter R esp o n se
0
Insertion Loss
Q factor0  0
7.90  .0057
.  .00 8
 0
I
 0
t rti
t r L
L
I
50 6 8 r 0 ( )
Figure 4.1: A 10 GHz Parallel Coupled Filter Response
Stop band frequencies and attenuation
14
FILTER DESIGN METHODS
Filter Design Methods
Two types of commonly used design methods:  Image Parameter Method  Insertion Loss Method Image parameter method yields a usable filter However, no clearcut way to improve the design i.e to control the filter response
15
.Ignoring phase simplified the actual synthesis method. called the Power Loss Ratio. ILM method also allows filter performance to be improved in a straightforward manner. The attenuation A([) can be cast into power attenuation ratio. An LC network is then derived that will produce this approximated response. which 16 is related to A([)2. PLR. at the expense of a µhigher order¶ filter.FILTER DESIGN METHODS Filter Design Methods The insertion loss method (ILM) allows a systematic way to design and synthesize a filter with various frequency response. A([) or s21([). A rational polynomial function is used to approximate the ideal H([). Here we will use A([) following [2]. Phase information is totally ignored.
FILTER DESIGN METHODS Zs Lossless 2port network PA +1.
low attenuation For example.[ Vs ZL PL PLR large. high attenuation PLR close to 1. a lowpass filter response is shown below: PLR(f) High attenuation Low attenuation Pin Power delivered to Load Pinc PA 1 ! ! ! PLoad 2 PA «1 +1 .
[ 2 » 1 +1 .
[ ¼ ¬ ½ PLR ! Power available from source network (2.1a) 1 LowPass filter PLR fc f 17 .
1b) 18 . b1 a1 Zc Vs PA Lossless 2port network Pin 1a 2 2 1 a1 ! 2 ! 2 b2 1b 2 2 1 s21 2 b2 Zc PL ! ! (2.PLR and s21 In terms of incident and reflected waves. assuming ZL=Zs = ZC.
Elliptic Function .Linear Phase 19 .Maximally flat (Butterworth) .FILTER RESPONSES Filter Responses Several types filter responses: .Equal Ripple (Chebyshev) .
is optimum in the sense that it provides the flattest possible passband response for a given filter complexity. .THE INSERTION LOSS METHOD Practical filter response: Maximally flat: .no ripple is permitted in its attenuation profile ¨[¸ PLR ! 1 k © ¹ ©[ ¹ ª cº 2 N [ .1 ] [ ± frequency of filter [c ± cutoff frequency of filter N ± order of filter 20 . .also called the binomial or Butterworth response.
11] 21 . .also known as Chebyshev.sharper cutoff .THE INSERTION LOSS METHOD Equal ripple .the passband response will have ripples of amplitude 1 +k2 ¨[ ¸ !1 k T © ¹ ©[ ¹ ª cº 2 2 N [ ± frequency of filter [c ± cutoff frequency of filter N ± order of filter [ .
THE INSERTION LOSS METHOD Figure .3: Maximally flat and equalripple low pass filter response. 22 .
.linear phase characteristic in the passband . 23 .maximum attenuation in the passband.minimum attenuation in the stopband. Linear phase: .to avoid signal distortion . .maximally flat function for the group delay.have equal ripple responses in the passband and stopband.THE INSERTION LOSS METHOD Elliptic function: .
THE INSERTION LOSS METHOD Figure .4: Elliptic function lowpass filter response 24 .
THE INSERTION LOSS METHOD Filter Specification Lowpass Prototype Design Scaling & Conversion Normally done using simulators Optimization & Tuning Filter Implementation Figure . : The process of the filter design by the insertion loss method. 25 .
: Low pass filter prototype.THE INSERTION LOSS METHOD Low Pass Filter Prototype Figure . N = 2 26 .
THE INSERTION LOSS METHOD Low Pass Filter Prototype ± Ladder Circuit Figure . 27 . (b) begin with series element. (a) begin with shunt element. : Ladder circuit for low pass filter prototypes and their element definitions.
capacitance for shunt capacitors. it will be necessary to determine the size. load conductance if gN is a series inductor. generator conductance.THE INSERTION LOSS METHOD g = generator resistance. As a matter of practical design procedure. 28 . This is usually dictated by a specification on the insertion loss at some frequency in the stopband of the filter. (k=1 to N) gN+1 = load resistance if gN is a shunt capacitor. gk = inductance for series inductors. or order of the filter.
8: Attenuation versus normalized frequency for maximally flat filter prototypes. 29 .THE INSERTION LOSS METHOD Low Pass Filter Prototype ± Maximally Flat Figure 4.
THE INSERTION LOSS METHOD Figure 4.9: Element values for maximally flat LPF prototypes 30 .
THE INSERTION LOSS METHOD Low Pass Filter Prototype ± Equal Ripple For an equal ripple low pass filter with a cutoff frequency 1. The power loss ratio is: c = ! 1 k T .
Since the Chebyshev polynomials have the property that 0 ® TN .12] Where 1 + k2 is the ripple level in the passband.[ 2 2 N [ .
[ ! ¯ 1 ° [ . but the power loss ratio of 1 + k2 at = for N even.12] shows that the filter will have a unity power loss ratio at = for N odd. 31 .
THE INSERTION LOSS METHOD Figure 4.10: Attenuation versus normalized frequency for equalripple filter prototypes.5 dB ripple level) 32 . (0.
11: Element values for equal ripple LPF prototypes (0.5 dB ripple level) 33 .THE INSERTION LOSS METHOD Figure 4.
0 dB ripple level) 34 .12: Attenuation versus normalized frequency for equalripple filter prototypes (3.THE INSERTION LOSS METHOD Figure 4.
35 .THE INSERTION LOSS METHOD Figure 4.0 dB ripple level).13: Element values for equal ripple LPF prototypes (3.
FILTER TRANSFORMATIONS Low Pass Filter Prototype ± Impedance Scaling L ! R0L ' [ .13d] R ' L ! R0RL 36 .13b] R ! R0 ' s [ .13a] C C ! R0 ' [ .13c] [ .
1 a] [ .FILTER TRANSFORMATIONS Frequency scaling for the low pass filter: [ [n [c The new element values of the prototype filter: [ .14] [ ' jX k ! j Lk ! j[Lk [c [ jBk ! j Ck ! j[Ck' [c [ .1 b] 37 .
FILTER TRANSFORMATIONS The new element values are given by: k ! ! k ! ! 0 k [ .1 b] k [ [c 38 .1 a] [ k [c k 0 [ .
FILTER TRANSFORMATIONS Low ass to high ass tra sformation The frequency substitution: [c [n [ [ .1 ] The new component values are given by: k ! ! 1 0[ c 0 [ .1 b] k [c 39 .1 a] k k [ .
2 ] The center frequency is: [0 ! [1[2 [ .21] 40 . [ .1 ] [2 [1 (! [0 [ .BANDPASS & BANDSTOP TRANSFORMATIONS Low ass to Bandpass transformation [ 0 ¨ [ [0 ¸ 1 ¨ [ [ 0 ¸ © ¹! © ¹ [n ¹ [2 [1 © [0 [ º ( © [0 [ ¹ ª ª º Where.
Lk. is transformed to a shunt LC circuit with element values: ( k ! [ ! [ .BANDPASS & BANDSTOP TRANSFORMATIONS The series inductor.23a] 0 k k k ([ [ . Ck.23b] 0 41 .22a] k [ .22b] k The shunt capacitor. is transformed to a series LC circuit with element values: k k ! ! ([0 ( [0 [ .
[ .24] [2 [1 (! [0 The center frequency is: [0 ! [1[2 42 .BANDPASS & BANDSTOP TRANSFORMATIONS Low pass to Bandstop transformation 1 ¨ [ [0 ¸ [ n (© ¹ ©[ [¹ º ª 0 Where.
Lk.2 a] k [ .BANDPASS & BANDSTOP TRANSFORMATIONS The series inductor. Ck. is transformed to a series LC circuit with element values: 1 k ! [0 ( ( k ! [0 [ . is transformed to a parallel LC circuit with element values: ( k k ! [0 1 ! [0 ( [ .2 b] 43 .2 a] k k [ .2 b] k The shunt capacitor.
BANDPASS & BANDSTOP TRANSFORMATIONS 44 .
45 .EXAMPLE .1 Design a maximally flat low pass filter with a cutoff . Compute and compare with an equalripple (3. impedance of insertion loss at 3 GHz. and at least 1 dB freq of 2 GHz. dB ripple) having the same order.
0 g 2 ! 1.5 2 [c g 1 ! 0.618 46 .EXAMPLE .618 g 3 ! 2.618 g 5 ! 0.1 (Cont) Solution: First find the order of the maximally flat filter to satisfy the insertion loss specification at 3 GHz.618 g 4 ! 1. 3 [ We can find the normalized freq by using: 1 ! 1 ! 0.
1 (Cont) The ladder diagram of the LPF prototype to be used is as follow: L ! R0 L ! 0 s L g1 C1 ! R0[c g3 C3 ! R0[c g5 C5 ! R0[c R0 g 2 L2 ! [c L4 ! R0 g 4 [c ! ! 0 0 L 47 .EXAMPLE .
984 pF 9 .618 ! ! 0.1 (Cont) LPF prototype for maximally flat filter g1 C1 ! R0[c R0 g 2 L2 ! [c g3 C3 ! R0 [ c R0 g 4 L4 ! [c g5 C5 ! R0[c 0.EXAMPLE .
50 .
T v 2 v10 2 50 v1.618 ! ! 6.438 nH 9 .
183pF 9 2 .00 ! ! 3.2T v 2 v10 2.
50 .
618 ! ! 6. T v 2 v 10 50 v1.438 nH 9 .
2T v 2 v10 0.618 ! ! 0.984 pF 9 .
50 .
T v 2 v10 2 48 .
4817 g 2 ! 0.EXAMPLE .4817 ! ! 5.7618 g 3 ! 4.4817 g1 C1 ! R0[c R0 g 2 L2 ! [c g3 3 ! R0[c R0 g 4 L4 ! [c g5 5 ! R0[c 3.541 pF 9 .5381 g 4 ! 0.7618 g 5 ! 3.1 (Cont) LPF prototype for equal ripple filter: g1 ! 3.
50 .
031 nH 9 .2T v 2 v10 50 v 0.7618 ! ! 3.
2T v 2 v10 4.5381 ! ! 7.223 pF 9 .
50 .
7618 ! ! 3.031 nH 9 .2T v 2 v10 50 v 0.
541 pF 9 .2T v 2 v10 3.4817 ! ! 5.
50 .
T v 2 v10 2 49 .
THE INSERTION LOSS METHOD Filter Specification Lowpass Prototype Design Scaling & Conversion Normally done using simulators Optimization & Tuning Filter Implementation 50 .
. Low Pass/High Pass/Band Pass/Band Stop Desired freq of operation Pass band & stop band range Max allowed attenuation (for Equal Ripple) 51 . Max Flat/Equal Ripple.SUMMARY OF STEPS IN FILTER DESIGN A. how much pass band ripple allowed? If max flat filter is to be designed. If equal ripple. 2. 4. cont to next step 3. . Filter Specification 1.
Low Pass Prototype Design 1. Determine whether shunt cap model or series inductance model to use 3.SUMMARY OF STEPS IN FILTER DESIGN (cont) B. Draw the lowpass prototype ladder diagram 4. No of Filter Order/Elements by using IL values 2. Determine elements¶ values from Prototype Table 52 . Min Insertion Loss level.
SUMMARY OF STEPS IN FILTER DESIGN (cont) C. band pass and band stop) 2. Scale elements to obtain the real element values 53 . Scaling and Conversion 1. Determine whether if any modification to the prototype table is required (for high pass.
Implement the lumped element filter onto a simulator to get the attenuation vs frequency response 54 . Put in the elements and values calculated from the previous step 2. Filter Implementation 1.SUMMARY OF STEPS IN FILTER DESIGN (cont) D.
The center frequency is 1 GHz.2 Design a band pass filter having a 0.EXAMPLE . the bandwidth is 10%. with N = 3. and the impedance is 50 .5 dB equalripple response. 55 .
1 N=3 [ = 1 GHz RS L1 L3 C2 RL 56 .EXAMPLE .2 (Cont) Solution: The low pass filter (LPF) prototype ladder diagram is shown as follow: ( = 0.
the filter elements are as follow.000 ! RL 2 g 3 ! 1.2 (Cont) From the equal ripple filter table (with 0. g1 ! 1.0967 ! g 4 ! 1.5 dB ripple).5963 ! L1 g 2 ! 1.5963 ! L3 57 .EXAMPLE .
EXAMPLE .2 (Cont) Transforming the LPF prototype to the BPF prototype RS L1 C1 L3 C3 L2 C2 RL 58 .
5963 v 50 ! ! 127.1 ! ! 0.EXAMPLE .199 pF C1 ! 9 Z 0[0 L1 .2 (Cont) L1Z 0 1.1 ( 0.0nH L1 ! 9 [0 ( 2T v1v 10 v 0.
50 2T v 2 v10 v1.5963 .
.
726n 9 2 [0C 2 .1v 50 L2 ! ! ! 0. (Z 0 0.
1).0967 ! ! 34. T v1v10 v1.91 p C2 ! 9 [0 (Z 0 2T v1v10 (0.0967 C2 1.
50 .
59 .
5963 v 50 ! ! 127.1 .0nH L3 ! 9 [0 ( 2T v1v10 v 0.EXAMPLE .2 (Cont) L3Z 0 1.
199 p 9 Z 0[0 L3 .1 C3 ! ! ! 0. ( 0.
50 .
T v 2 v10 v 1.5963 2 60 .
EXAMPLE .3 Design a fivesection high pass lumped element filter with 3 dB equalripple response. What is the resulting attenuation at 0. and an impedance of 50 . a cutoff frequency of 1 GHz.6 GHz? 61 .
6 GHz. is about 41 dB RS L1 C2 L3 C3 L5 RL 62 .3 (Cont) Solution: The high pass filter (HPF) prototype ladder diagram is shown as follow: N=5 [ = 1 GHz 1 [ At [c = 0.12 0.6 [c The attenuation for N = 5.EXAMPLE .667 . referring back to Fig 4. 1 ! 1 ! 0.
g1 ! 3.7618 ! C 4 g 5 ! 3.4817 ! L1 g 2 ! 0. the filter elements are as follow.0 dB ripple).EXAMPLE .3 (Cont) From the equal ripple filter table (with 3.4817 ! L 5 g 6 ! 1.000 ! RL 63 .5381 ! L 3 g 4 ! 0.7618 ! C 2 g 3 ! 4.
EXAMPLE .4817 .28nH 9 [c L1 2T v 1v10 v 3.3 (Cont) Impedance and frequency scaling: Z0 50 L'1 ! ! ! 2.
18 p 9 Z 0[c C 2 . 1 1 C '2 ! ! ! 4.
7618 .50 2T v1v10 v 0.
754nH 9 2 [c L3 . Z0 50 L'3 ! ! ! 1.
T v1v10 v 4.5381 64 .
18 pF C '4 ! 9 Z 0[c C 4 .3 (Cont) 1 1 ! ! 4.EXAMPLE .
7618 .50 2T v 1v 10 v 0.
Z0 50 L'5 ! ! ! 1.754nH 9 [c L 5 2T v 1v 10 v 4.5381 .
65 .
GHz. fc = [ c ! 2T . Rs = RL= 1.8478H Ohm.EXAMPLE .7654H L2=1. L1=0.4 Design a 4th order Butterworth LowPass Filter.
4248 v 109 rad/s 1 Zo ! 50 Step 1&2: LPP g0= 1 C1=1.5GHz ! 9.803nH g0=1/50 C1=3.8478F C2=0.624pF RL= 50 66 .7654F RL= 1 R ! Z o Rn L L ! Zo n [c C! Cn Z o[ c Step 3: Frequency scaling and impedance denormalization L =4.921pF C2=1. .061nH 1 L2=9.
fc = 1. .3661H [ c ! 2T . dB ripple factor. Rs = Ohm. GHz. L1=1.6703H L2=2. Design a 4th order Chebyshev LowPass Filter.EXAMPLE .
5GHz ! 9.1926F C2=0.2 67 .55nH g0=1/50 C1=2. .9841 R ! Z o Rn L L ! Zo n [c C! Cn Z o[ c Step 3: Frequency scaling and impedance denormalization L =8.531pF C2=1.4248 v 109 rad/s 1 Zo ! 50 Step 1&2: LPP g0= 1 C1=1.861nH 1 L2=12.787pF RL= 99.8419F RL= 1.
N = 3. Center frequency fo = 1. GHz. f2=1. Design a bandpass filter with Butterworth (maximally flat) response.4GHz. 3dB Bandwidth = 2 MHz or f1=1.EXAMPLE . GHz. 68 .
Zo=1 g2 2.EXAMPLE . [c=1.000F g3 1.000F g4 1 [ c ! 2Tf c ! 1 f c ! 21 ! 0.1592 Hz T 69 . (cont) From table.000H Step 1&2: LPP 2<0o g1 1. design the LowPass prototype (LPP) for 3rd order Butterworth response.
[1 ! 2T .EXAMPLE . (cont) Step 3: Frequency scaling and impedance denormalization LZ o [o( LPP to bandpass transformation. Impedance denormalization.
4GHz 1 fo ! [ 2 ! 2T . .
497GHz ! 0. .133 [ [1 (! 2 [o C [ o (Z o (Z o [ oC ( [ o LZ o 50 79.7072nH 70 .6GHz 1 f1 f 2 ! 1.1414pF Vs RL 15.916pF 50 0.58nH 0.7072nH 15.916pF 0.
This action might not be possible to undo. Are you sure you want to continue?