Chebyshev filter

Thread Starter

bwd111

Joined Jul 24, 2013
117
Distinguishing features of a Chebyshev filter? Im thinking It allows ripple in the passband just because it doesnt have a maximally flat response over its passband. Pretty sure im correct thou
 
Last edited:

LvW

Joined Jun 13, 2013
1,761
Ripple in the passband
Ripple in the stopband
Steeper rolloff than Butterworth
Sorry - not 100% correct.
No ripple in the stop band;
Steeper roll-off near the pass band edge only - in general, the roll-off is determined by the order only.
In addition, pass band edge defined (mostly) NOT by -3 dB but by the ripple.
 

Papabravo

Joined Feb 24, 2006
21,228
Sorry - not 100% correct.
No ripple in the stop band;
Steeper roll-off near the pass band edge only - in general, the roll-off is determined by the order only.
In addition, pass band edge defined (mostly) NOT by -3 dB but by the ripple.
That is for a Chebyshev Type I
The rolloff for a Chebyshev type I is steeper than for a Butterworth of the same order.
The Chebyshev type II has no ripple in the passband and does have ripple in the stopband and it's rolloff is not as steep as the Type I.

http://en.wikipedia.org/wiki/Chebyshev_filter
 

LvW

Joined Jun 13, 2013
1,761
That is for a Chebyshev Type I
Yes - if somebody asks for Chebyshev response (without any additional information) I always assume Chebyshev I - and not the inverse Chebyshev function.

The rolloff for a Chebyshev type I is steeper than for a Butterworth of the same order.
Sorry - not correct. As mentioned before, the roll-off is determined by the filter order only (starting approx. one cade above the pass band edge).
Butterworth and Chebyshev (2nd order): 40 dB/dec.

Hint: Do not blindly trust wikipedia.
 

Papabravo

Joined Feb 24, 2006
21,228
Yes - if somebody asks for Chebyshev response (without any additional information) I always assume Chebyshev I - and not the inverse Chebyshev function.


Sorry - not correct. As mentioned before, the roll-off is determined by the filter order only (starting approx. one cade above the pass band edge).
Butterworth and Chebyshev (2nd order): 40 dB/dec.

Hint: Do not blindly trust wikipedia.
You have to trust them if you're too cheap to buy a book.
You could also build one and measure it, but even that has become too difficult for most.
 

LvW

Joined Jun 13, 2013
1,761
You have to trust them if you're too cheap to buy a book.
You could also build one and measure it, but even that has become too difficult for most.
And what about using your own brain?
A short look on the transfer function reveals that the denominator is the same for both functions - thus, the roll-off is the same, of course.
(It was the only purpose of my last sentence to remind you and other readers that it is not a good approach to pick-up any claims/statements found in books or the Internet without a critical review).
 

The Electrician

Joined Oct 9, 2007
2,970
And what about using your own brain?
A short look on the transfer function reveals that the denominator is the same for both functions - thus, the roll-off is the same, of course.
Are you saying that the denominator of the transfer function for a Butterworth filter is the same (identical) as the transfer function for a Chebychev filter of the same order?

Perhaps you meant to say "A short look on the transfer function reveals that the denominator is the same order for both functions"
 

LvW

Joined Jun 13, 2013
1,761
The_Electrician, thanks for the attempt to correct me.
However, what I had in mind is the 2nd order denominator in its general form:

D(s)=1+s/(wpQp) + (s/wp)^2

Of course, the pole Q (Qp) is different for both functions (and the same wp).
 
Top