0
\$\begingroup\$

In order to reach a steep attenuation for an analog antialiasing filter, there are issues if a high Q notch filter is followed by a low order lowpass filter (or vice versa)? The notch resonance could make a brickwall filter near the cutoff frequency while the low pass would handle the higher part of the spectrum.

\$\endgroup\$
1
  • \$\begingroup\$ No this wouldn't work how you expect - have you tried superimposing a notch response with a low pass response and drawing the resultant combined frequency response. \$\endgroup\$
    – Andy aka
    Jan 9, 2016 at 18:54

2 Answers 2

3
\$\begingroup\$

This in essence is exactly how an elliptic filter works.

Most filter design packages that will design 'normal' Butterworth and Chebychev filters will also design elliptic filters, these have a Cheby passband, with added zeroes in the stopband.

If you try to randomly add zeroes to an existing low pass response, you probably won't get the response you want. If you do the proper elliptic design, then you get to choose the passband ripple and the stopband attenuation.

\$\endgroup\$
0
0
\$\begingroup\$

There is no easy solution to making a brickwall filter. A high-Q notch will indeed provide a steep cutoff, but it will by its nature have a very narrow rejection band, followed by an equally steep high-pass function. Since the rejection band is narrow, it won't do a whole lot to improve the function of the low-pass.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.