EMT 359/3 – Analog Electronic II

Outlines
1. Introduction
2. Advantages of Active Filters over Passive Filters
3. Types of filter
4. Filter Response Characteristic
5. Active Low-Pass Filter
6. Active High-Pass Filter
7. Active Band-Pass Filter
8. Active Band-Stop Filter
9. Summary
 Introduction
 Filters are circuits that are capable of passing signals within
a band of frequencies while rejecting or blocking signals of
frequencies outside this band.
 This property of filters is also called “frequency selectivity”.
 Passive filters - built using components such as resistors,
capacitors and inductors.
 Active filters - employ transistors or op-amps in addition to
resistors and capacitors.
 Active filters are mainly used in communication and signal
processing circuits.
 They are also employed in a wide range of applications such
as entertainment, medical electronics, etc.
Advantages of Active Filters over Passive Filters

1. Active filters can be designed to provide required

gain, and hence no attenuation as in the case of
passive filters.

2. No loading problem, because of high input

resistance and low output resistance of op-amp.

3. Active Filters are cost effective as a wide variety of

economical op-amps are available.
 Active Filter
There are 4 basic categories of active filters:
 Low-Pass filters

 High-Pass filters

 Band-Pass filters

 Band-Stop filters

Each of these filters can be built by using op-amp or

transistor as the active element combined with RC,
RL or RLC circuit as the passive elements.
 Active Filter
 The passband is the range of
frequencies that are allowed to
pass through the filter.
 The critical frequency, fc is
specified at the point where
the response drops by 3 dB
from the passband response
(i.e. to 70.7% of the passband
response)
 The stopband is the range of
frequencies that have the most
attenuation
 The transition region is the
area where the fall-off occurs
 Low-Pass Filter
 Allows the frequency from
0 Hz to critical frequency, fH
(also known as cutoff
frequency)
 Ideally, the response drops
abruptly at the critical
frequency fH

Ideal response
Actual response
 1st Order Low-Pass Filter
1
Vo o
H  s   s  CR 
Vi s
1 s  o
CR
1
o 
CR
o
H  s 
s  o
o 1
H  j   
j   o 1  j 
o
 1st Order Low-Pass Filter
1
H  j  
2
 
1   
 o 
 
 
 1 
H  j  dB  20 log  
2
   
 1    
  o  
 1st Order Low-Pass Filter
When 0
1
H  j   1
2
 0 
1   
 o 
And
H  j  dB  20 log 1  0 dB
 1st Order Low-Pass Filter
When 1
  c  o 
CR
1 1
H  j     0.707
 c 
2
11
1   
 o 
And
H  j  dB  20 log 0.707  3 dB

1 is known as 3-dB or cutoff

c 
CR frequency (rad/s)
 1st Order Low-Pass Filter
STRAIGHT-LINE APPROX.
0 dB
-3 dB
SLOPE
= -20 dB/decade
-10 dB

PASSBAND

-20 dB

-30 dB

-40 dB
0.01f c 0.1f c fc 10f c 100f c
 2nd Order Low-Pass Filter
Vi  V1 V1  Vo
 sCV1 
R R R V1 R
Vi  sCR  2  Vo
 V1  …(1) + I1 I2 I3 I4 +
R  R  R C C
Vi Vo
V1  Vo - -
 sCVo
R
V1   sCR  1Vo …(2)
Vi  sCR  2  Vo
(1) & (2)    sCR  1Vo 
R  R  R 1
Vo 1 H  s 
 CR 
2

H  s   3 1
Vi  sCR  2  3sCR  1 s s
2

CR  CR  2
 2nd Order Low-Pass Filter
o2
H  s  2  1 
s  s3o  o2 where o  CR 
 
0 dB

-3 dB SLOPE
= -40 dB/decade
-20 dB

-40 dB PASSBAND

-60 dB

-80 dB
0.01f c 0.1f c fc 10f c 100f c
 High-Pass Filter
 Allows the frequencies
above the critical frequency Gain,
fL. (also known as the cutoff
1
frequency.
 Ideally, the response rises
abruptly at the critical
frequency.
0 fL f
Stop Pass
band band
Ideal response
Actual response
 1st Order High-Pass Filter
Vo R
 s  1
Vi R
sC
s s
 
s
1 s   o
CR
s
H  s 
s  o
1
Where; o 
CR
 1st Order High-Pass Filter
j 1
H  j   
j   o 1  j  o

1
H  j  
2
 o 
1  
   
 
 1 
H  j  dB  20 log 
2 
 1   o  
    
 1st Order High-Pass Filter
1
H  j  
2
 o 
1  
 

When   0 H  j   

When    H  j   1  0 dB

When   c  o H  j   0.707  3 dB
 1st Order High-Pass Filter
STRAIGHT-LINE APPROX.

0 dB
-3 dB

SLOPE
-10 dB = 20 dB/decade

PASSBAND
-20 dB

-30 dB

-40 dB
0.01f c 0.1f c fc 10f c 100f c
 2nd Order High-Pass Filter
C V1 C

+ +
Vi R R Vo
- -

 s2   1 
H  s    2 
2  where  
 o CR 
 s  s3o  o   
 2nd Order High-Pass Filter

0 dB

-3 dB

-20 dB SLOPE
=4
20 dB/decade

-40 dB

-60 dB

-80 dB
0.001f c 0.01f c 0.1f c fc 10f c 100f c
 1st & 2nd Order Filter
Low-Pass Filter (LPF) High-Pass Filter
 c  s
H  s  H  s 
s  c s  c

1 1
 H  j    H  j  
 o
1 j 1 j
o 
o 2
 2

 H  s  2
s
 H  s    2 
2 
s  s3o  o2  s  s3o  o 
 Band-Pass Filter
Allows frequencies
between a lower cutoff
Gain,
frequency (fL) and an 1
upper cutoff frequency
(fH).
0 fL fH f
Pass
band
Ideal response
Actual response
Band-Pass Filter
 Bandwidth (BW)

BW  f c 2  f c1
 Center frequency

f0  f c1 f c 2
 Quality factor (Q) is the
ratio of center frequency fo
to the BW
f0
Q
BW
 Band-Stop Filter
Frequencies below fc1
Gain,
(fL) and above fc2 (fH)
are passed. 1

0 fL fH f
Stop
band
Ideal response
Actual response
 Filter Response Characteristics
Identified by the shape
of the response curve
• Passband flatness
• Attenuation of
frequency outside the
passband
Three types:

1. Butterworth
2. Bessel
3. Chebyshev
 Filter Response Characteristics
1. Butterworth
 Amplitude response is very flat in passband.
 The roll-off rate -20 dB per decade (per filter order).
 normally used when all frequencies in the passband must have
the same gain.
2. Chebyshev
 overshoot or ripples in the passband.
 The roll-off rate greater than –20 dB.
 can be implemented with fewer poles and less complex
circuitry for a given roll-off rate.
3. Bessel
 Linear phase response.
 Ideal for filtering pulse waveforms.
 Filter Response Characteristics
Damping Factor
determines the type of
response characteristic
either Butterworth,
Chebyshev, or Bessel.
The output signal is fed
back into the filter
circuit with negative
feedback determined
R1
by the combination of DF  2 
R1 and R2 R2
 Filter Response Characteristics
Critical Frequency
 The critical frequency, fc is
determined by the values of
R and C in the frequency-
selective RC circuit.
 For a single-pole (first-
order) filter, the critical
frequency is: 1
fc 
2 RC
 The above formula can be
used for both low-pass and
high-pass filters
 Filter Response Characteristics
Roll-off rate
 Greater roll-off rates can be achieved with more poles.
 Each RC set of filter components represents a pole.
 Cascading of filter circuits also increases the poles which
results in a steeper roll-off.
 Each pole represents a –20 dB/decade increase in roll-off
Filter Response Characteristics

Roll-off depends on
number of poles
 Active High-Pass Filters
At critical frequency,
Resistance = capacitive
reactance

R  Xc
1
R
c C
critical frequency:
1
1 fc 
R 2RC
2f c C
Active High-Pass Filters

Roll-off
depends on
number of
poles.
 Active High-Pass Filters
A Single-Pole Filter

1 R1
fc  Acl  1 
2RC R2
 Active High-Pass Filters
The Sallen-Key
 second-order (two-pole) filter
 roll-off -40dB per decade

1
fc 
2 RA RB C AC B

LetsRA = RB = R
and CA = CB = C;

1
fc 
2RC
 Active High-Pass Filters
V2  V f R

 R2  I2
V f   Vo C C V2 0
 R1  R2  Vi V1
+
I1 I3 I4
Vo Vo R1  R2 R1 R -
Vo
Ao     1
V2 V f R2 R2 Vf
R1

Vo R2
V2 
Ao
At node V1: I1  I 2  I 3
V1  Vo
sC Vi  V1    sC V1  V2 
R
 Active High-Pass Filters

V1  Vo  Vo 
sC Vi  V1    sC V1   ……(1)
R  Ao 
At node V2: I 3  I 4
V2
sC V1  V2  
R
 Vo  Vo
sC V1   
 Ao  Ao R
Vo Vo
V1  
Ao Ao sCR ……(2)
1  1 1   1 sC 
(2) in (1): sCVi    s 2C   Vo    Vo
R  Ao Ao sCR   R Ao 
 Active High-Pass Filters
  sCR  2   3  Ao  sCR  1 
sCVi   2
Vo

 A o sCR 
Ao  sCR 
2
Vo Ao s 2
H  s   
Vi  sCR    3  Ao  sCR  1 s 2   3  Ao  s  1
2

CR  CR  2

Ao s 2 1
 2 where o 
s   3  Ao o s  o
2
CR

  3  Ao = Damping Factor
Ao s 2
H  s  2 2
s  o s  o
 Active High-Pass Filters

  3  Ao I2
C C V2 0
+
 R1  Vi V1
 3  1   I1 I3 I4
Vo
 R2 
R -

R1
Vf
R1
  2
R2 R2
 Active High-Pass Filters
Cascaded HPF – Six pole
 cascade 3 Sallen-Key two-pole stages
 roll-off -120 dB per decade
 Active Low-Pass Filters
At critical frequency,
Resistance = capacitive
reactance
R  Xc
1
R
c C critical frequency:

1 1
R fc 
2f c C 2RC
 Active Low-Pass Filters
A Single-Pole Filter

1 R1
fc  Acl  1 
2RC R2
 Active Low-Pass Filters
The Sallen-Key
 second-order (two-pole) filter
 roll-off -40dB per decade
1
fc 
2 RA RB C AC B

For RA = RB = R
and CA = CB = C
1
fc 
2RC
 Active Low-Pass Filters
C

R R
+
Vi
C Vo
-

R1
 
2

H  s   Ao  2 o 
2 
 s  o s  o  R2

1 R1
o  &   2
CR R2
 Active Low-Pass Filters
The 3-dB frequency c is related to o by a factor
known as the FREQUENCY CORRECTION FACTOR
(kLP), thus;   k   k LP
c LP o
RC
Parameter table for Sellen-Key 2nd order LPF
Type of response  kLP
Bessel 1.732 0.785
Butterworth 1.414 1
Chebyshev (1 dB) 1.054 1.238
Chebyshev (2 dB) 0.886 1.333
Chebyshev (3 dB) 0.766 1.390
 Example
Design a Bessel 2nd order low-pass filter with a 3-dB
frequency of 5 kHz. Use C = 22 nF.
C

R R
+
Vi
C Vo
-

R1

R2
 Solution
k LP k LP
c  k LPo  R
RC c C

From the table: k  0.785

LP

Hence:
0.785
R  1.136 k
2  5 10  22 10
3 9
 Solution
In order to minimize the offset error, we have to
meet the following condition:
R1 // R2  2 R

Or; R1 R2
 2 R  2.27 k
R1  R2
R1
And:   2 
R2
From the parameter table;   1.732
 Solution
R1
 2    0.268
R2
R1 R2
R1  0.268R2  2.27 k
R1  R2
2
0.268R2
 2.27 k
0.268R2  R2
1.268  2.27
R2   10.74 k
0.268
R1  0.268R2  0.268 10.74  2.88 k
 Solution
C

22 nF

R R
+
Vi 1.136 kW 1.136 kW
C 22 nF Vo
-

R1

2.88 kW
R2 10.74 kW

A Bessel 2nd order LPF

 Solution
10

-25

-50
100Hz 1.0KHz fc 10KHz 100KHz
VDB(C5:2)
Frequency

Frequency response curve of a 2nd order Bessel LPF

 Example
1. Determine critical frequency
2. Set the value of R1 for Butterworth response
 Solution
1. Critical frequency

1
fc   7.23 kHz
2RC
2. Butterworth response, R1/R2 = 0.586

R1  0.586 R2

R1  586 kΩ
 Active Low-Pass Filters
Cascaded LPF – Three-pole
 cascade two-pole and single-pole
 roll-off -60dB per decade
 Active Low-Pass Filters
Cascaded LPF – Four pole
 cascade two-pole and two-pole
 roll-off -80dB per decade
 Example
Determine the capacitance values required to
produce a critical frequency of 2680 Hz if all resistors
in RC low pass circuit is 1.8 k
 Solution

1
fc 
2RC
1
C  0.033 μF
2f c R
C A1  C B1  C A 2  C B 2  0.033 μF
 Active Low-Pass Filters
SELLEN-KEY LPF
 The slope (roll-off) of a filter is associated with the its
order – the higher the order of a filter, the steeper will be
its slope.
 The slope increases by 20 dB for each order, thus;

ORDER SLOPE (dB/decade)

1 20
2 40
3 60
4 80
 Active Low-Pass Filters
Active filters can be cascaded to increase its order,
e.g. cascading a 2nd order and a 1st order filter will
produce a 3rd order filter

Input 2nd order 1st order Output

signal LPF LPF signal

Cascading a 2nd order and a 1st order filter produces a 3rd

order filter with a roll-off of 60dB/decade
 Active Band-Pass Filters
A cascade of a low-pass and high-pass filter

Band-pass filter formed by cascading a two-pole

high-pass and a two-pole low-pass filters
Active Band-Pass Filters

1
f c1 
2 RA1 RB1C A1C B1
1
fc2 
2 RA2 RB 2C A 2C B 2

f0  f c1 f c 2
 Active Band-Pass Filters
Multiple-Feedback
BPF
 The low-pass circuit
consists of R1 and C1.
 The high-pass circuit
consists of R2 and C2.
 The feedback paths are
through C1 and R2.
 Center frequency:
1
f0 
2  R1 // R3  R2C1C2
 Active Band-Pass Filters
Multiple-Feedback BPF
 For C = C = C, the resistor values can be
1 2
obtained using the following formulas:
Q Q Q
R1  R2  R3 
2f oCAo f oC 2f oC (2Q 2  Ao )

 The maximum gain, Ao occurs at the center

frequency.
 Active Band-Pass Filters
State-Variable BPF
 It consists of a summing amplifier and two integrators.
 It has outputs for low-pass, high-pass, and band-pass.
 The center frequency is set by the integrator RC circuits.
1  R5 
R5 and R6 set the Q (bandwidth). Q    1


3  R6 

 Active Band-Pass Filters
The band-pass output peaks sharply the center
frequency giving it a high Q.
 Active Band-Pass Filters
Biquad Filter
 contains an integrator, followed by an inverting amplifier,
and then an integrator.
 In a biquad filter, the bandwidth is independent and the
Q is dependent on the critical frequency.
 Active Band-Stop Filters
The BSF is opposite of BPF in that it blocks a specific
band of frequencies.
The multiple-feedback design is similar to a BPF
with exception of the placement of R3 and the
addition of R4
 Active Band-Stop Filters
State Variable Band-Stop Filter
 Summing the low-pass and the high-pass responses of the
state-variable filter with a summing amplifier creates a
state variable band-stop filter
 Summary
The bandwidth of a low-pass filter is the same as the
upper critical frequency.
The bandwidth of a high-pass filter extends from the
lower critical frequency up to the inherent limits of
the circuit.
The band-pass passes frequencies between the
lower critical frequency and the upper critical
frequency.
A band-stop filter rejects frequencies within the
upper critical frequency and upper critical frequency.
 Summary
The Butterworth filter response is very flat and has a
roll-off rate of –20 dB.
The Chebyshev filter response has ripples and
overshoot in the passband but can have roll-off rates
greater than –20 dB.
The Bessel response exhibits a linear phase
characteristic, and filters with the Bessel response
are better for filtering pulse waveforms.
 Summary
A filter pole consists of one RC circuit. Each pole
doubles the roll-off rate.
The Q of a filter indicates a band-pass filter’s
selectivity. The higher the Q the narrower the
bandwidth.
The damping factor determines the filter response
characteristic.