Adjust gain, analog model of LP filter

DSP, Plug-in and Host development discussion.
juha_p
KVRist
480 posts since 21 Feb, 2006 from FI

Post Fri Jun 01, 2018 5:18 am

I have this "analog model" of one pole LP filter:

Code: Select all

w0 = 2*pi*fc;
b = 1;
a = [1 w0];
I would like to drop the filter say -3dB ... is it OK to do by just multiplying the transfer result?

Example (Octave code, fc = 700Hz):

Code: Select all

G = 1/sqrt(2);
AnalogLP1 = tf(b, a); 

% Results:
% Transfer function 'Analog' from input 'u1' to output ...
%
%         1
% y1:  --------
%     s + 4398
%
%Continuous-time model.
%
% without gain normalization filter is at -72.866 dB 

AnalogLP1 = AnalogLP1 * a(2) * G

%Results:
% Transfer function 'Analog' from input 'u1' to output ...
%        3110
% y1:  --------
%     s + 4398
%
%Continuous-time model.
%

Ivan_C
KVRian
1065 posts since 11 Aug, 2004 from Breuillet, France

Re: Adjust gain, analog model of LP filter

Post Fri Jun 01, 2018 7:37 am

If you multiply your analog transfer function with a constant, you get a new transfer function which is absolutely the same but with the magnitude amplified or attenuated by that constant for every frequency...

juha_p
KVRist
480 posts since 21 Feb, 2006 from FI

Re: Adjust gain, analog model of LP filter

Post Fri Jun 01, 2018 8:57 am

Thanks for the info.

Return to “DSP and Plug-in Development”