Lecture 5

Microwave Applications

Filter Design

Amr M. E. Safwat, Ph.D.

Professor
Ain Shams University
Cairo Egypt

Course Content
 Review and introduction
 Planar transmission lines
 Network theory
 Resonators
 Filter design
 Microwave components
 Matching networks
 Amplifier design
 RF measurement

Microwave applications Filter Design (2)

1
 Outline
Insertion Loss method
Flat Response LPF prototype
Chebyshev filter
Filter transformation, Impedance and frequency scaling
Filter implementation on microstrip

Microwave applications Filter Design (3)

Microwave Filter
A microwave filter is a two-port network used to control the frequency
response at a certain point in a microwave system by providing
transmission at frequencies within the passband of the filter and
attenuation in the stopband of the filter.

Microwave filter theory and practice began in the years preceding

World War II.

The image parameter method of filter design was developed in the late
1930s and was useful for low-frequency filters in radio and telephony.

Today, most microwave filter design is done with computer-aided

design (CAD) packages based on the insertion loss method.

Microwave applications Filter Design (4)

2
 Insertion-Loss Method
The perfect filter would have zero insertion loss in the
passband, infinite attenuation in the stopband and a linear
phase response (to avoid signal distortion) in the passband.

The image parameter method may yield a usable filter

response, but there is no clear-cut way to improve the
design.

The insertion loss method, however, allows a high degree

of control over the passband and stopband amplitude and
phase characteristics with a systematic way to synthesize a
desired response.

Microwave applications Filter Design (5)

Insertion-Loss Method (Cont'd)

In the insertion loss method a filter response is defined by its
insertion loss, or power loss ratio, PLR:
Power available from source Pinc 1
PLR   
Power delivered to load Pload 1 | ( ) |2

Symmetry of Z() and ()

Since v(t) and i(t) has to be real function and since we have
(v(t)=v*(t)):
  
1 1 1
 V ( )e d
jt
v(t )   V * ( )e d   V * ( )e
 jt jt
v * (t )  d
2  2 
2 

V ( ) V * ( )
V ( )  V * () I ( )  I * ( ) Z ( )    Z * ( )
I ( ) I * ( )

Re(V()) is an even function and Im(V()) is an odd

function, similarly are I() and Z()
Microwave applications Filter Design (6)

3
 Insertion-Loss Method (Cont'd)
Z ( )  Z 0 R( )  jX ( )  Z 0
As for the reflection coefficient: ( )  
Z ( )  Z 0 R( )  jX ( )  Z 0

R( )  jX ( )  Z 0 R( )  jX ( )  Z 0

 * ( )    ( )
R( )  jX ( )  Z 0 R( )  jX ( )  Z 0

Re(()) is an even function and Im(()) is an odd function

|(()|2 is an even function and it can be expanded in

terms of 2.

M ( 2 ) M ( 2 )
| ( ) |2  PLR  1 
M ( 2 )  N ( 2 ) N ( 2 )

Microwave applications Filter Design (7)

Practical Filter Responses

Maximally flat: Also called the binomial
or Butterworth response, it provides the
flattest possible passband response for a
given filter complexity, or order. For a low-
pass filter, it is specified by:
 2N
PLR  1  k 2 ( )
c
where N is the order of the filter, and c is the cutoff frequency.
If @ c the insertion loss is 3 dB (which is usually the case), hence k=1
The insertion loss increases at the rate of 20N dB/decade.

Equal ripple. A Chebyshev polynomial is used to specify 

the insertion loss of an N -order low-pass filter as: PLR  1  k 2TN2 ( )
c
Sharper cutoff will result although the passband response will have ripples of
amplitude 1 +k2. the insertion loss also increases at the rate of 20N dB/decade

Microwave applications Filter Design (8)

4
 Practical Filter Reponses (cont'd)
Elleptic Function: Minimum
stopband attenuation, in which case a
better cutoff rate can be obtained.
Such filters have equal-ripple
responses in the passband as well as
the stopband.

Linear phase In some applications (such as multiplexing filters for

communication systems) it is important to have a linear phase response
in the passband to avoid signal distortion. A linear phase characteristic
can be achieved with the following phase response:
 2N
 ( )  A (1  p( ) )
c
d 
where A is defined as: d   A[1  p(2 N  1)( ) 2 N ]
d c

Microwave applications Filter Design (9)

Process of Filter design by the ILM

Low-pass filter prototypes are normalized in terms of

impedance and frequency. This normalization simplifies
the design of filters for arbitrary frequency, impedance,
and type.
The low-pass prototypes are then scaled to the desired
frequency and impedance; and the lumped-element
components replaced with distributed circuit element" for
implementation at microwave frequencies

Microwave applications Filter Design (10)

5
Flat Response Low-pass Filter prototype
Problem: Derive the normalized
element values L and C, for a
maximally flat response. Assume
a source impedance of 1 and a
cutoff frequency c = 1. Find R?

Solution
 2N
Since we have two elements, N=2 PLR  1  k 2 ( ) PLR  1   4
c
R(1  jRC )
The input impedance of this filter is: Z in  jL 
1   2 R 2C 2
1 | Z in  1 |2
PLR  
1 | ( ) |2 2( Z in  Z in* )
1
1 [(1  R) 2  ( R 2C 2  L2  2 LCR 2 ) 2  L2C 2 R 2 4 ]
4R
R=1 L=C L=C=2

Microwave applications Filter Design (11)

Flat Response Low-pass Filter prototype (Cont'd)

Microwave applications Filter Design (12)

6
 Equal Ripple Low Pass Filter
Chebyshev Polynomials

(1) m cosh( m cosh 1 z ) z  -1


Tm ( z )   cos( m cos 1 z ) -1  z  1
 cosh( m cosh 1 z ) z  1


Microwave applications Filter Design (13)

Equal Ripple Low Pass Filter (Cont'd)


PLR  1  k 2TN2 ( )
c
For a two elements filter N=2, we have:
 
T2 ( )  2( ) 2  1 PLR  1  k 2 (2 2  1) 2  1  k 2  4 2 k 2  4 4 k 2
c c
1 | Z in  1 |2
PLR  
1 | ( ) |2 2( Z in  Z in* )
1
1 [(1  R) 2  ( R 2C 2  L2  2 LCR 2 ) 2  L2C 2 R 2 4 ]
4R
1
[(1  R) 2  k 2 ( R 2C 2  L2  2LCR2 )  4k 2 L2C 2 R 2  4k 2
4R

Problem: R is function of k. Solution: Use odd number of elements or a

matching circuit

Microwave applications Filter Design (14)

7
 Equal Ripple Low Pass Filter (Cont'd)

0.5 dB ripple level 3.0 dB ripple level

Microwave applications Filter Design (15)

Filter Transformation
Impedance Scaling

If the source impedance is R0, then all values derived above will be
normalized as follows:

R's  R0
R'L  R0 RL
L' R0 L

C '  C / R0

Where R's , R'L, C' and L' are the parameters' value after normalization.

Microwave applications Filter Design (16)

8
 Filter Transformation (Cont'd)
Frequency Scaling

Low pass prototype Low pass with High pass with

frequency scaling frequency scaling
Low pass filter

In this case, all  will be replaced as follows: 
c
 L
Hence, we will have: jX L  j L  jL' L' 
c c

jBC  j C  jC ' C' 
C
c c
Microwave applications Filter Design (17)

Filter Transformation (cont'd)

Frequency Scaling

High pass filter

c
In this case, all  will be replaced as follows: 

c 1 1
Hence, we will have: jX L   j L C '
 jC ' c L
 1
jBC   j c C  L'
1
 jL' c C

To include the impedance scaling thus:

1 R0
C'  L'
c LR0 c C

Microwave applications Filter Design (18)

9
 Filter Transformation (cont'd)
Frequency Scaling
Band pass filter
In this case, all 
will be replaced
as follows:
0   1   2  1
 (  0) (  0) where   and 0  21
2  1 0   0  0

Hence, we will have:

1  0 L L  L
jX L  j (  )L  j j 0 C'  series L' 
 0  0  0 L 0 

1  0 C C  C
jBC  j (  )C  j j 0 L' '  Shunt C' ' 
 0  0  0 C 0 

Microwave applications Filter Design (19)

Summary of Filter Transformation

Microwave applications Filter Design (20)

10
 Example 8.3.
Design a maximally flat low-pass
filler with a cutoff frequency of 2
GHz, impedance of 50 , and at
least 15 dB insertion loss at 3 GHz.

Solution:

| | 1  0.5 N=5
c
From table

Microwave applications Filter Design (21)

Example 8.4
Design a bandpass filler having a 0.5 dB equal-ripple response.
with N = 3. The center frequency is 1 GHz, the bandwidth is 10%,
and the impedance is 50.

Microwave applications Filter Design (22)

11
 Filter Implementation
The Lumped-element filter design discussed in the previous
sections generally works well at low frequencies, but two problems
arise at microwave frequencies.
First, lumped elements such as inductors and capacitors are
generally available only for a limited range of values and are
difficult to implement at microwave frequencies, but must be
approximated with distributed components.
In addition, at microwave frequencies the distances between
filter components is not negligible.

Microwave applications Filter Design (23)

Richard’s transformation
Let's make the transformation     tan l When  is normalized, we have

jX L  jL tan l
@c=
1

jBC  jC tan l

@c=
1

Major disadvantage: In some TL configuration, it is very difficult to realize a

short circuited stub to get the inductor.

Microwave applications Filter Design (24)

12
 Kuroda identities
The four Kuroda identities use redundant transmission line sections to
achieve a more practical microwave filter implementation by
performing any of the following operations:

Physically separate transmission line stubs

Transform series stubs into shunt stubs or vice versa
Change impractical characteristic impedances into more realizable
ones.

The additional transmission line sections are called unit elements UE

and are /8 long at c. The unit elements are thus commensurate with
the stubs used to implement the inductors and capacitors of the
prototype design.

Microwave applications Filter Design (25)

Kuroda identities (Cont'd)

Microwave applications Filter Design (26)

13
 Kuroda identity (Cont'd)
Let's have:     tan l
For a transmission line:
 cos l jZ c sin l 
A B  
C D    j sin l cos l 
 
 Zc 
 1 j Z c
1  j 
 For the left circuit
1  2  Z 1 
 c   1 jZ 2 
A B 1  
C D   1 1 2 Z1
 A B   j
0
1   2  j( Z  Z ) 1   Z 
1
 
For a OC C D    
1  1 2 2
   Z c  For the right circuit
 ( Z1  Z 2 ) 
 1 j 
A B 1 n2
 A B  1 jZ c  C D    
1   2  jn
2
For a SC   2 Z1 
C D  0 1 
1 
    Z 2 Z 2 
L=R if n2=1+Z2/Z1

Microwave applications Filter Design (27)

Example 8.5
Design a low-pass
filter for fabrication Initial design
using micros trip lines.
The specifications are:
cutoff frequency of 4
GHz, third order,
impedance of 50 , Richard
and a 3 dB equal- Transformation
ripple characteristic.

Solution

Two matched lines

Microwave applications Filter Design (28)

14
 Example 8.5 (Cont'd)

Kuroda Transformation

All characteristic impedance

are multiplied by 50 

Implemented circuit
on microstrip

Microwave applications Filter Design (29)

Example 8.5. (Cont'd)

Microwave applications Filter Design (30)

15
STEPPED-IMPEDANCE LOW-PASS FILTERS
Hi-Z, low-Z filters are popular:
They are easier lo design.
They take up less space than a
similar low-pass filter using stubs.
TL distributed Lumped model
For a lossless transmission line, we have:
L 1 C 
Z and vp   
C LC  eff 

L  Z0 / v p High Z distributed Lumped model

and C  Z 0v p

X  Ll  Z 0l / v p  Z 0 l


B  Cl  l  Y0 l Low Z distributed Lumped model
v p Z0
Microwave applications Filter Design (31)

Example 8.7
Example: Design a stepped-
impedance low-pass filter
having a maximally flat
response and a cutoff
frequency of 2.5 GHz. It is
necessary to have more than
20 dB insertion loss at 4.0
GHz. The filler impedance is
50, the highest practical line
impedance is 150 and the
lowest is 10.
solution

| | 1  0.6 N=6
c

Microwave applications Filter Design (32)

16
 Example 8.7 (cont'd)

Where Zl=low impedance=10 and Zh=high impedance=150

Microwave applications Filter Design (33)

Next Time

Microwave components

Microwave applications Filter Design (34)

17