Circuit Analisys 1

U4. Introduction to passive

filters
Circuit Analysis, 2012-13
Circuit Analisys 2
Introduction
Definition of a filter: A device (i.e.
quadripole) that selects a interval of
frequencies from an input signal, whose
amplitudes and phase can be modified.
Analogical filters (signal is no discrete)
Passive: made only with R,L and C
Active: have also operational amplifiers,
transistors,
Digital filters (signal is discrete)
Circuit Analisys 3
Introduction
X(s) Y(s)
Q
t (s)
Input signal (X(s))
Low-pass filter
High-pass filter
Band-pass filter
Band-stop filter
Output
signal (Y(s))
Classification according to the selected range of frequencies
Circuit Analisys 4
Transfer Function H(
s
s)
Definition (Laplace Transform domain (L))
[ ]
[ ]
,
signal input
signal output
) (
) (
) (
L
L
= =
s X
s Y
s H
I
1
(s) I
2
(s)
V
2
(s) V
1
(s)
Filter
) (
) (
) ( ,
) (
) (
) (
) (
) ( ,
) (
) (
) (
2
1
1
2
1
2
1
2
s V
s I
s Y
s I
s I
s G
I
s V
s Z
s V
s V
s G
T I
T V
= =
= =
X(s) Y(s)
H(s)
with initial conditions equal zero (v
C
(0)=i
L
(0)=0).
H(s) is independent of the applied signal.
Note that H(s)=Y(s) when x(t)= (t), (h(t)=L
-1
[H] is the impulse response).
For example
Circuit Analisys 5
Transfer Function
Example 1
Example 2
What kind of filter are they?
Circuit Analisys 6
Poles and zeros of H(
s
s)
H(s) can be written as a fraction of two polynomials with
real-valued and positive coefficients:
( )
( )

=
+ + +
+ + +
=

N
n
M
m
N
N
N
M
M
M
p s
c s
K
b s b s
a s a s
K s H
1
1
1
1
1
1
...
...
) (
where c
m
and p
m
are the zeros and the poles of the
transfer function:
{ }
{ }

) ( lim
0 ) ( lim
s H
s H
m
m
p s
c s
zeros and the poles can also be at s
Circuit Analisys 7
Pole-zero plot
The poles and zeros are usually represented in a
complex plane called the pole-zero plot to help to
convey certain properties of the circuit
The poles and ceros are either real or complex
conjugated
Poles are represented with x
Zeros are represented with o
Circuit Analisys 8
Frequency response of the
Transfer Function H()
Frequency response using sinusoidal signals,
then s=j :
( )
( )
( ) ) ( j exp ) (
j
j
) (

H
p
c
K H
N
n
M
m
=

H() becomes high for close to p

n
H() becomes low for close to c
m
,
)] ( Re[
)] ( Im[
arctan ) (
,
j
j
) (

H
H
p
c
K H
N
n
M
m
=

where
is the amplitude response,
is the phase response.
Circuit Analisys 9
Pole-zero plot and |H()|
Example 3
( )( )
, ) (
2 1
1
p s p s
c s
s H

=
Poles at s = p
1
, p
2
Zero at s = c
1
Poles at s =1j5
Zero at s = 5
Poles at s =1j10
Zero at s = 5
Poles at s =5j10
Zero at s = 5
The closer
the pole to
the imaginary
axis, the
more
pronounced
is the
maximum
Circuit Analisys 10
Pole-zero plot and |H()|
Example 1
,
10
10
) (
1
1
+
=
+
=
s s
s H
RC
RC
Pole at s= -10
Zero at s

+
=
+
=
10
exp
100
10
10
10
) (
2

j
j
H
Data: R=1 k, C=100 F
Circuit Analisys 11
Pole-zero plot and |H()|
Example 2
( )
( )( )
,
2 j 1 2 j 1
1
) (
1 2
2
2
+ + +
+
=
+ +
+
=
s s
s s
s s
s s
s H
LC L
R
L
R
Poles at s = 1j2
Zeros at s = 0,-1
( )
2
2
2
2
2
2
4 5
1
) ( ,
j2 5
j
) (

+
+
=
+
+
= H H
Data: R=1 , C=1/5 F, L=1 H
Circuit Analisys 12
H(s) between (E
g
,R
g
) and R
L
) (
) (
) (
2
s E
s V
s H
g
=
Circuit Analisys 13
Examples of RLC passive filters
Low-pass filter
1st order
2nd order
High-pass filter
1rt order
2nd order
Band-pass filter
Band-stop filter
Circuit Analisys 14
Example: Low-pass filer
Circuit Analisys 15
Example: High-pass filer
Circuit Analisys 16
Example: Band-pass filer
Circuit Analisys 17
Example: Band-stop filter
Circuit Analisys 18
Ideal amplitude response
Low-pass filter
High-pass filter
Band-pass filter
Band-stop filter
PB: Pass band, AB: Attenuation band,
C
: Cut frequency
Circuit Analisys 19
Impulse response of ideal filters
Ideal filters can not be implemented in practice
The real filer have to be approached to the ideal filter
response using different methods
The approached filters have not a constant response in
the PB and are not entirely zero in the AB
There exist a transition band between PB and AB
The order of the filter coincides with the number of
poles of H(s)
The higher the order of the filter the close is his behavior
as an ideal filter
To design filters, layout templates with the tolerance
margins are used
Circuit Analisys 20
Layout templates for filter design
Circuit Analisys 21
Examples of real filters
Circuit Analisys 22
1st order Low-pass filter
Transfer function
c
c
s
k s H

+
= ) (
where
c
is the cut frequency, at this frequency the attenuation of
the signal is 3dB:
2
1
) (
) (
3
) (
) (
log 20
max max
10
= =
= =

H
H
H
H
c c
2 2
) (
c
c
k H

+
=
Circuit Analisys 23
Examples 1st order Low-pass filter
Low-pass RC: Low-pass RL
L
R
H
c
L
R
L
R
=
+
=

,
j
) (
RC
H
c
RC
RC
1
,
j
) (
1
1
=
+
=

Circuit Analisys 24
1st order High-pass filter
Transfer function
c
s
s
k s H
+
= ) (
2
1
) (
) (
3
) (
) (
log 20
max max
10
= =
= =

H
H
H
H
c c
2 2
) (
c
k H

+
=
where
c
is the cut frequency, at this frequency the attenuation of
the signal is 3dB:
Circuit Analisys 25
Examples 1st order High-pass filter
High-pass RC: High-pass RL
L
R
j
H
c
L
R
=
+
=

,
j
) (
RC
H
c
RC
1
,
j
j
) (
1
=
+
=

Circuit Analisys 26
2st order passive filters
Transfer function:
N
2
(s): Polynomial of order 2
: Damping coefficient

0
: Characteristic frequency of the filter
Q=1/(2): Quality factor of the circuit
B=
0
/Q: Band width of the circuit
2
0
2
2
2
0 0
2
2
2 1
2
2
0
) (
2
) ( ) (
) (


+ +
=
+ +
=
+ +
=
s s
s N
k
s s
s N
k
a s a s
s N
k s H
Q
Circuit Analisys 27
2nd order passive filters
With 2nd order filter the four type of filters can be
obtained (with 1st order only two)
The poles of H(s) are:
Depending on the value of (or Q) we distinguish the
following two poles:
Real and different (>1 or Q<1/2)
Real and equal (=1 or Q=1/2)
Complex conjugated (<1 or Q>1/2)
Depending on N
2
(s) we have the following special
cases:
( ) ( ) 1 4 j 1
2
1 j
2
0
2
0 2 , 1
= = Q
Q
s
p

Circuit Analisys 28
2nd order filters
Low-pass filter
High-pass filter
Band-pass filter
Band-stop filter
Circuit Analisys 29
2nd order Low-pass filter
Example:
Circuit Analisys 30
2nd order High-pass filter
Example:
Circuit Analisys 31
2nd order Band-pass filter
B=
0
/Q
2
k
2
0
2 2
0
2
0
0
) (

+ +
=
+ +
=
Bs s
Bs
k
s s
s
k s H
Q
Q
Example:
Circuit Analisys 32
2nd order Band-stop filter
Example:
Circuit Analisys 33
Examples 2nd order filters