560 Q variation

[recnsci]

Does anyone have idea (or even better, reliable information) how much Q varies in API 560 from "flat" to full cut/boost?

btw, in my world Q is "damping" part of denominator of transfer function, i.e. Butterworth has Q of 0.707

Best Regards
Urosh

 

[helterbelter]

Are you asking for a dB vs Q vs Freq equation ? I can't help with that, but maybe others know....

 

[recnsci]

Are you asking for a dB vs Q vs Freq equation ? I can't help with that, but maybe others know....

actually, I would just like to know what's the span of Q variation, like "Q is reduced 2.7 times from flat to full cut/boost". My best guess based on little info I gathered is that Q goes from between 2 and 1 to about 0.6-0.4

 

[Core]

Maybe Automated Processes have further information?

 

[JohnRoberts]

Does anyone have idea (or even better, reliable information) how much Q varies in API 560 from "flat" to full cut/boost?

btw, in my world Q is "damping" part of denominator of transfer function, i.e. Butterworth has Q of 0.707

Best Regards
Urosh

AFAIK there is not universal agreement on how to even measure and specify the "q" for a boost/cut EQ section. While there is wide agreement for what Q means in simple HP/LP/BP filters, but boost/cut EQ sections are formed by adding or subtracting a BP section to a dry signal. So different EQ topologies give different families of curves for a given BP section. 

Rane has published a article on their website about the variations between sundry 1/3rd octave graphic equalizers, which contrary to what you might expect are not all the same.

I submitted a request to the AES standards committee a few years ago, and while they concede a problem no clarification has been forthcoming.

This is a problem in professional sound reinforcement when loudspeakers manufacturers try to publish generic crossover curves for use with different DSP platforms. The filter curves all agree, but any corrective EQ can give different results depending upon the platform specific definition of Q.

Which is a long way for me to say I don't know... .

JR
 

 

[abbey road d enfer]

What JR says.
I deliberately don't use Q notion because it just doesn't work for EQ's. It works only for RLC bandpass filters. The academic definition of Q suppose asymptotic zero response and narrow BW. Eq filters never obey the 1st condition and 2nd condition is not applicable for less than 3dB boost/cut.
On a mathematical POV, the relation between BW and Q, Q= Fc/BW is applicable to a 2nd-order filter, which transfer function is expressed by
1/as²+bs+1 (low-pass), bs/as²+bs+1 (band-pass) or as²/as²+bs+1 (high-pass).
The transfer function of an audio equalizer is more complex than that; it is mathematically a bi-quad, whose expression is as²+(1+k)bs+1/as²+bs+1
K is directly related to the amount of boost or cut
Some authors think that the Q of the numerator is the one to take into account, which basically means considering the Q at extreme boost or cut. In practical terms, it's acceptable as long as the amount of boost or cut is higher than 12dB.
But this evidently does not answer your question as to how Q evolves with the amount of boost/cut.
IMO, there is no answer to that. I could give an answer to: "How does BW evolves with the amount of boost/cut?", but that would be according to my definition, which may not be someone else's.
But then, even if we agreeed on how BW evolves, that would not give anybody the right to "convert" it in Q values.
As JR says, there are as many definitions of EQ bandwidth as there are designers.
Back to the essence of your topic: does it really matter? What do you expect in terms of practical conclusions?