Question about Parametric EQ "Q" value

El Creepo said:
Does anyone have a rough basic idea about how our "Q" value might translate to the "bandwidth/octave" value on the Furman PQ-4? It goes from 1/3 - 4 octave.

I found a site that has a bunch of equations and calculations but I have no idea if that translates to how it is done in the Axe FX.
Click to expand...

It does. It's a simple f0/(f2-f1) ratio. So your table shows the PQ4 values on the left and the AxeFX values on the right.

However......one of the features valued in the PQ4 is the way the filter changes from when you boost to when you cut. So, the bandwidth values on the front panel are only correct when boosting. The actual values are quite a bit different when cutting (or vice versa, sorry I don't remember).

P.S.
I looked it up. The Q when cutting ranges from about 1/10 octave to about 1/2 octave. So, use that range instead of the range on the front panel when cutting. Good luck dialing in your Dimebag tone :).
 
Last edited:
GlennO said:
It does. It's a simple f0/(f2-f1) ratio. So your table shows the PQ4 values on the left and the AxeFX values on the right.

However......one of the features valued in the PQ4 is the way the filter changes from when you boost to when you cut. So, the bandwidth values on the front panel are only correct when boosting. The actual values are quite a bit different when cutting (or vice versa, sorry I don't remember).

P.S.
I looked it up. The Q when cutting ranges from about 1/10 octave to about 1/2 octave. So, use that range instead of the range on the front panel when cutting. Good luck dialing in your Dimebag tone :).
Click to expand...
"Feature".

1581386150859.jpeg

This is due to a poor circuit design. Designing a PEQ using additive/subtractive bandpass filters results in this kind of response. IOW H(s) = 1 + Hbp(s)*k where Hbp(s) is a bandpass filter and k is a constant that adds (or subtracts) the response from the overall response.

The right way is to to put the poles and zeros on a coincident root locus. When you increase the gain the poles move towards the imaginary axis and the zeros move away and vice-versa. Most engineers don't know how to do this though as it's not a cookbook technique.
 
I don't know anything about the intent or competence of the Furman designers, or have an opinion on the actual results of this particular piece, but some EQs, parametric and otherwise, have asymmetrical boost and cut on purpose. I think the idea is that narrower cuts are less intrusive, while wider boosts are more natural sounding.
 
Absolutely. The "wide when boosting, narrow when cutting" principle is an important axiom when mixing. The idea is you're trying to get rid of a particular problem frequency when cutting, but you're trying highlight a frequency range when boosting. This is an example of something that mathematically seems wrong, but musically makes perfect sense.
 
FractalAudio said:
"Feature".

View attachment 63510

This is due to a poor circuit design. Designing a PEQ using additive/subtractive bandpass filters results in this kind of response. IOW H(s) = 1 + Hbp(s)*k where Hbp(s) is a bandpass filter and k is a constant that adds (or subtracts) the response from the overall response.

The right way is to to put the poles and zeros on a coincident root locus. When you increase the gain the poles move towards the imaginary axis and the zeros move away and vice-versa. Most engineers don't know how to do this though as it's not a cookbook technique.
Click to expand...
The rg 100es and Randall century 200 might be poor circuit designs too, but many of us pantera fans are awaiting their inclusion in the amp section 😏🤩
 
You must log in or register to reply here.
Back
Top Bottom