Scribd
Upload
40 views14 pages

Analog Butterworth Filter Desing Example Student

The document outlines the design process for an Analog Butterworth filter, detailing the necessary specifications, design steps, and verification methods. It includes a specific example of designing a low pass filter for audio applications, with calculations for passband and stopband gains, pole placement, and transfer function formulation. The design is verified using MATLAB to ensure the filter meets the specified frequency response criteria.

Uploaded by

22u1829
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
40 views14 pages

Analog Butterworth Filter Desing Example Student

The document outlines the design process for an Analog Butterworth filter, detailing the necessary specifications, design steps, and verification methods. It includes a specific example of designing a low pass filter for audio applications, with calculations for passband and stopband gains, pole placement, and transfer function formulation. The design is verified using MATLAB to ensure the filter meets the specified frequency response criteria.

Uploaded by

22u1829
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 14

Analog Butterworth

Filter Design Example


Vishweshkumar Aithal
Assistant Professor,
Dept. of ECE,
KLS GIT, Belagavi-590008.
Contents:
• Analog IIR Filter design steps
• Butterworth design with a specific example
• Verification of the design
Analog IIR Filter Design Steps:
1. List the specs:
Pass band edge Ω𝑝 in rad/s
stop band edge Ω𝑠 in rad/s
pass band gain 𝐾𝑝 (or attenuation 𝐴𝑝 ) in dB
stop band gain 𝐾𝑠 (or attenuation 𝐴𝑠 ) in dB,

Pass band edge fpass in Hz ⟹ Ωp = 2π × fpass in rad/s

Stop band edge fstop in Hz ⟹ Ωs = 2π × fstop in rad/s


Analog IIR Filter Design Steps (cont.):
Pass band ripple δp ⟹ K p = −Ap = 20 log10 (1 − δp ) in dB

Stop band ripple δs ⟹ K s = −As = 20 log10 (δs ) in dB

2. Use appropriate magnitude approximation (usually mentioned)


Butterworth/ Chebyshev.
3. Devise the prototype LPF, estimate the order of the filter, and place
poles accordingly to get the prototype LPF transfer function.
4. Apply suitable spectral transformation (low pass to low pass, low
pass to high pass, low pass to band pass or low pass to band stop).
Butterworth (Analog) Design:
• An audio application requires the bass frequencies to be separated and
to be fed to sub woofer. Design a filter which passes all the frequencies
up to 250Hz (only bass frequencies). Filter should have monotonic pass
band and monotonic stop band. Filter should offer maximum
attenuation of 3dB in passband and minimum attenuation of 20dB in
the stopband. Take stopband edge at 1000Hz.
• Solution: The low pass filter has to be designed.
• Specifications list:
Passband gain K p = −Ap = −3dB
Stopband gain K s = −As = −20dB
Passband edge frequency Ωp = 2π × 250 rad/s
Stopband edge frequency Ωs = 2π × 1000 rad/s
20 log
Devise
H jΩ
Normalized Lowpass
20 log
Prototype
H jΩ
10 N 10 a

0dB 0dB
Kp
𝐬 Kp
𝐬→
𝛀𝐩

Ks Ks

1 Ω′s Ω Ωp Ωs Ω
Normalized Lowpass Prototype 1 rad/s → Ωp rad/s Desired Analog Lowpass Filter

Ω′s rad/s → Ωs rad/s


Ωs 2π × 1000
Ω′s = Ωp = = 4rad/s
2π × 250
Butterworth (Analog) Design cont.:
• Normalized prototype low pass filter:
Passband gain K p = −Ap = −3dB
Stopband gain K s = −As = −20dB
Passband edge frequency Ωp = 1 rad/s
Stopband edge frequency Ω′s = 4 rad/s
Order Estimation:
10−0.1𝐾𝑝 − 1 100.3 − 1
log log
10−0.1𝐾𝑠 − 1 102 − 1
𝑁= = = 1.655390 ⟹ 𝑁 = 2
Ω𝑝 1
2 log 2 log
Ω′𝑠 4
Butterworth (Analog) Design cont.:
• Placement of poles:
• N = 2, hence, two poles need to be placed
πk π π πk π π
Sk = 1∠ + + ⟹ Sk = 1∠ + +
N 2N 2 2 4 2
Where, k = 0 to 2N − 1 ⇒ k = 0,1,2,3
First two poles lie on left half of s plane (for 𝑘 = 0 and 𝑘 = 1).
Next two poles lie on right half of s plane (for 𝑘 = 2 and 𝑘 = 3).
So 𝑘 = 0 and k = 1 are sufficient for a stable filter.
1 1 1 1
S0 = − +j S1 = − −j
2 2 2 2
Butterworth (Analog) Design cont.:
• The transfer function of normalized Butterworth low pass filter is
given by:
1
HN S = H2 s =
s − s0 s − s1
1
H2 s =
1 1 1 1
s+ −j s+ +j
2 2 2 2

1
H2 s =
s2 + 2s + 1
Butterworth (Analog) Design cont.:
• Spectral transformation:
• Apply lowpass to lowpass transformation.
• The cut off frequency is given by:
Ωp
Ωc = 1
100.1Kp −1 2N

(we match the pass band exactly and over-satisfy the stopband)

2π × 250
Ωc = 1 = 2𝜋 × 250.2969852 ≈ 1572.66 rad/s
100.3 − 1 4
Butterworth (Analog) Design cont.:
Ha s = H2 s |s→Ω
s = H2 s s
|s→1572.66
c

1
Ha s = 2
s s
+ 2 +1
1572.66 1572.66

2473259.48
Ha s = 2
s + 2224s + 2473259.48
Design Verification:
• To find the magnitude response let s = jΩ in Ha (s) and take the
magnitude:

|2473259.48|
Ha (jΩ) =
| jΩ 2 + 2224 jΩ + 2473259.48|

2473259.48
Ha (jΩ) =
2473259.48 − Ω2 2 + 2224Ω 2

• Ha jΩ Ω=Ωp =2π×250 = 0.707969 ≈ −3dB


• Ha jΩ Ω=Ωs =2π×1000 = 0.062526 ≈ −24dB < −20dB
Design Verification in MATLAB:
clear all;close all;clc;
b = [0 0 2473259.48];
a = [1 2224 2473259.48];
n = 2*pi*linspace(0,1.1*10^3,10^5);
[h,w]=freqs(b,a,n);
figure, plot(w,20*log10(abs(h)),'LineWidth',2);
xlabel('Frequency in rad/s');
ylabel('Magnitude of H[s] in dB');
title('Magnitude response plot of H[s]');
grid on;
Design Verification in MATLAB:

You might also like