Author |
Message |
Electromagnetic Wave
Joined: Apr 28, 2013 Posts: 302 Location: Kebek
G2 patch files: 38
|
Posted: Wed Feb 26, 2014 5:47 pm Post subject:
Filter Nord / FM Lin, how to find the peak frequency |
|
|
Hi,
I adapted the scala patch to be used on a filter. I want to change the frequency (Hertz) for each value (0-127). Now I try to find a way to determine the exact peak frequency (accuracy : 0.01 Hz) for each value.
I have no problem to find the frequency of an oscillator. Regarding the filter I thought about using a white noise as a sound source followed by a filter (bandpass, slope : 24 dB/Oct, maximum resonance : 127). But it does not work. Too much movement in the noise.
Another way would be to use a sine which is set to the desired frequency. Scanning the spectrum with the filter up to act on the amplitude of the sine.
But this way I work twice. I have to adjust the frequency of an oscillator for each value before adjusting the filter.
Do you have any other ideas of what I could use as an audio source?
I added a patch with a filter and a white noise to demonstrate what I am trying to do (Active the patch and play with the "Constant2" knob).
Thanks
Description: |
|
Filesize: |
204.22 KB |
Viewed: |
660 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
Description: |
|
Download |
Filename: |
FilterTuning12.pch2 |
Filesize: |
5.26 KB |
Downloaded: |
4176 Time(s) |
|
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Thu Feb 27, 2014 4:07 am Post subject:
|
|
|
Not sure if this applies to all filters ... some will keep resonating after having seen an initial small value, so you don't need the noise to stay present then.
When it does not keep oscillating it may still work with an impulse repeated every once in a while ... depending on how fast you can measure the frequency.
Depending on the filter the resonance frequency may depend on the set Q, something to keep in mind maybe. _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
Electromagnetic Wave
Joined: Apr 28, 2013 Posts: 302 Location: Kebek
G2 patch files: 38
|
Posted: Thu Feb 27, 2014 6:30 pm Post subject:
|
|
|
Quote: | Not sure if this applies to all filters ... some will keep resonating after having seen an initial small value, so you don't need the noise to stay present then. |
Yes! Thanks ! It works much better with a slope of 12 dB/Oct ... it does not work well with a slope of 24 dB/Oct ...I don't understand but this is not a big problem. I can probably use a delay with feedback to keep the signal.
A little reading :
Quote: | With digital filters you have to give the filters a bit of a push to get them oscillating. This is true for analog filters as well, but in analog systems there is always a bit of noise around to do the job. In a digital system you have to explicitly provide this. |
http://www.cim.mcgill.ca/~clark/nordmodularbook/nm_filters.html |
|
Back to top
|
|
|
|