L303.A.R1
A1
APPENDIX
SUMMARY of BUTTERWORTH
FILTERS
The Butterworth filter is constructed with passive elements; R, L, and C. The voltage
transfer function of any filter is defined as the ratio of the output voltage to the input
voltage, expressed in the terms of the complex frequency variable s as follows,
H(s) = V
0
(s)/V
IN
(s) = N(s)/D(s)
(1)
The transfer function can always be expressed as the ratio of two polynomials, N(s) and
D(s).
The roots of the numerator are the zeros and the roots of the denominator are the
poles of the transfer function.
We will consider a low pass filter for our discussion.
The low pass filter, as depicted in
Fig. 1, will pass low frequency signals up to the corner (or break) frequency, f
c
. Then, the
output of the filter will drop at a rate of 20n dB/dec, where n is the order of the
denominator.
The ideal filter has perfect response. To obtain the maximum possible slope,
for a real filter, we let the order of the numerator be zero, i.e., N(s) = A. This will result in an
allpole transfer function. For an effective filter the order of the denominator n must be as
large as possible.
Fig. 1
Low Pass Filter Characteristics.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
L303.A.R1
A2
Hence, the Butterworth transfer function of an nth order low pass filter can be written as,
H
n
(s) = A/B
n
(s)
(2)
where the denominator B
n
is called the Butterworth polynomial of order n and is defined in
Eq. 4 below.
Transfer Function Magnitude
The magnitude squared of the transfer function, with s = j
ω
, can also be expressed as
follows,
H
n
(j
ω
)
2
= H(j
ω
) H(j
ω
) = A(
ω
2
)/B(
ω
2
)
(3)
Since A
n
(s) = A, then A
n
(
ω
2
) = A. The denominator can be written as a polynomial in
ω
2
,
B
n
(
ω
2
) = B
0
+ B
2
ω
2
+ B
4
ω
4
+ . . . + B
2n
ω
2n
(4)
where B
n
(
ω
2
) is the Butterworth polynomial of order n.
We see from Eq. 3 that the transfer
function is an allpole function, no zeros.
If we let B
0
= A
0
then the transfer function at
ω
= 0
is H
n
(0) = 1.
If B
2n
= 1/
ω
c
2n
, then we have the special case shown below. The transfer
function of the Butterworth filter of order n can now be written,
H
n
j
ω
(
)
2
=
A
0
B
0
+
B
2n
ω
2n
=
1
1
+
ω ω
c
(
)
2n
or
H
n
j
ω
(
)
=
1
1
+
ω
ω
c
2n
=
1
1
+
f
f
c
2n
(5)
Eq. 5 is a simple formula for calculating the magnitude of the Butterworth transfer
function in terms of the corner frequency! It is easily plotted in Maple. See Fig. 2. We see
the properties of the Butterworth filter; at low frequencies the transfer function is 1, and at
high frequencies the rolloff is 20n dB/dec. Also at
ω
=
ω
c
the magnitude of the transfer
function is 1/
√
2 (or 3 dB) for any n. These are simple properties of the Butterworth filter.
This is the end of the preview.
Sign up
to
access the rest of the document.
 Spring '10
 GERBER
 Lowpass filter, Butterworth Filter

Click to edit the document details