0% found this document useful (0 votes)
43 views

Active Filters 3

This document discusses parameter scaling and sensitivity issues with the Sallen-Key filter configuration. It proposes using component simulation as an alternative design technique, where inductors in a passive filter design are replaced with active generalized impedance converter (GIC) circuits. The GIC can simulate a grounded inductor using two op-amps, allowing the design of higher-order filters without sensitivity problems.

Uploaded by

maithuong85
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
43 views

Active Filters 3

This document discusses parameter scaling and sensitivity issues with the Sallen-Key filter configuration. It proposes using component simulation as an alternative design technique, where inductors in a passive filter design are replaced with active generalized impedance converter (GIC) circuits. The GIC can simulate a grounded inductor using two op-amps, allowing the design of higher-order filters without sensitivity problems.

Uploaded by

maithuong85
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 17

Parameter Scaling

 Most filter designs are given in a normalised


form, i.e. for a cut-off frequency of 1 rad/s.
 To transform for an arbitrary frequency, the
component values must be scaled.
Design Example – Low Pass Filter

1
H s   0 
1

1
R 2C1C2 s 2  2 RC2 s  1 2
R C1C2 RC
 02 1 C  C1C2
 2 
s  2 0 s   0  s  2 2s
2
C2
   1   RC2 0 
 0  0 C
Normalised Un-normalised
 0  1 rad/s  0  500 rad/s
  0.38   0.38
R 1  R  10 k
C1  2.6 F 1 1
C  4  200 nF
C2  0.38 F R 0 10  500
C1  C /   200 / 0.38  520 nF
C2  C  200  0.38  76 nF
Sensitivity
 When designing high-order filters (e.g. six or more
poles), damping ratios of individual sections can be
very low.
 The actual damping ratio of the Sallen and Key filter
is sensitive to the open loop gain of the op-amp.
 This sensitivity is especially notable when the cut-off
frequency is high (implying the open loop gain is
low).
 Detailed sensitivity calculations are required to
quantify this effect…(ref. Sedra & Smith)
Sensitivity Example

  0.1, 0  1000 rad/s


10
A0  1000
A0  200
1 A0  100
Gain

0.1

0.01
100 1000 10000
Frequency [rad/s]
Overcoming Sensitivity Problems
 In high order filters, the compounded errors
due to the sensitivity of the Sallen & Key
configuration could be significant.
 Solutions:
 Use better tolerance components.
 Use a different configuration
 Component Simulation
 Operational Simulation
Component Simulation
 Filter design by component simulation can be broken
down into two stages:
 Design a passive LCR network that realises the desired
transfer function.
 Replace all the inductors with equivalent active networks
(impedance convertors).

E.g. Passive
high pass filter
Impedance Convertors
I1 I2

Impedance
V1 Convertor V2

V2  V1
I 2  I 1
Voltage controls voltage. Current controls current.
Conversion works in either direction – bilateral.
Impedance Conversion
I1 I2

Impedance
V1 Convertor ZL V2

V2
I2 
ZL
V1
 I1 
ZL
V1 Z L
Z IN    K  ZL
I1 
Using Impedance Convertors
 The impedance appears to be multiplied by the
convertor circuit.
 Several types of convertor:
 Positive Impedance Convertor (PIC) : K is real and positive.
 Negative Impedance Convertor (NIC) : K is real and
negative.
 Generalised Impedance Convertor (GIC) : K is complex and
may be frequency dependent.
 For example,
 ZL = a resistance, R
 K=j
 ZIN = j R
 i.e. input looks like an inductance.
Generalised Impedance Convertor
Op-amp analysis
rules:
V  V
VOUT  anything

 v1  v3  v5
v2 & v4 are arbitrary.
Consider currents flowing from left to right:

v1  v2
I Z 1  iin   v1  v2  iin Z1
Z1

v2  v3 v2  v1  iin Z1
IZ 2   
Z2 Z2 Z2
v3  v4
IZ3   v3  v4  I Z 3 Z 3
Z3
v4  v5 v4  v3  I Z 3 Z 3
I Z 4  iout   
Z4 Z4 Z4
but, I Z 3  I Z 2
  iin Z1  Z 3 iout Z1Z 3
 iout     
 Z2  Z4 iin Z 2 Z 4
Loaded GIC

Z1 Z 3
IZ5  iin
Z2Z4
VZ 5  vin  I Z 5 Z 5

 Z1 Z 3 
 vin   iin Z5
 Z2Z4 
vin
vin Z1Z 3 Z 5
Z IN  
iin Z2Z4
0V
Simulated Grounded Inductor
Z1 Z 3 Z 5
Z IN 
Z2Z4
R R R

1
R
sC
 sCR 2

The input appears


to be an inductance
of CR2 Henrys.
Five Pole High Pass
Filter Example
Summary
 The Sallen & Key second order filter configuration is
highly sensitive to component tolerances.
 The resulting errors are particularly problematic for
high order filters.
 Component simulation is an alternative filter design
technique.
 It simply replaces inductors in a passive filter by
active equivalent circuits.
 The GIC configuration can simulate a grounded
inductor using two op-amps.

You might also like