BaxEq question

[richiyobs]

Hi i m trying to redraw this simple baxEQ.... found on google...
simulation is fine... i d like to know the role of Ra and Rb, the article is not complete!
from Ltspice i noticed they balance the bands level and also looks like the slope is changing to

thanx for your attention

best

 

[JohnRoberts]

Ra and Rb are like the input resistors feeding a virtual earth inverting op amp stage... So Ra relative to nominal 10k feedback resistors affects boost/cut.

That looks more complicated than the simple Baxandal EQ made famous, but maybe he designed more than one.

JR

 

[richiyobs]

thanx for explaining
is this a Barry Porter s design?

 

[JohnRoberts]

IDK, I've seen similar EQ used in consoles.

[edit- Barry Porter's iconic EQ design was a state variable full parametric EQ /edit]

JR

 

[richiyobs]

builded a proto and it works and sound great!
also measured into smaart to see phase plot..... very precise , specially when you find the flat position between gains, very very linear...i m using standard pots at moment.... hope that center dented pot can be of help!
need to calculate this flat condition for bands gain to have lorlin pot for a stereo build.... is there a useful trick to do this or multimeter in hands and check pot value according to its position?

Ra and Rb were selected according to psu headroom.... +&-18 .... so cut and boost will not exceed

thanx for attention
best

richi

 

[richiyobs]

Just to experiment I used it to filter the side-chain of a Jfet limiter instead of standard Hpf...Great results!
Could also set it to have a Tilt response and so the way the SC behaves.... A sort of key filter ... I m using to reduce distortion at low freq cause you can set the way to smooth with the filter shapes
Am I wrong or is it doable?
Thanx for attention
Best

 

[richiyobs]

For Gain pots... I was using log ones... So middle value was over the center...

 

[abbey road d enfer]

also measured into smaart to see phase plot..... very precise , specially when you find the flat position between gains, very very linear...

Shelving EQ's are almost alway (very few exceptions) 1st order. That means that the phase shift is less than 90°. Actually, phase-shift is directly related to the amount of boost/cut. In other words, two shelving EQ's with identical frequency response also have identical phase response, whatever any wicked marketing blurb would try to sell.

need to calculate this flat condition for bands gain to have lorlin pot for a stereo build.... is there a useful trick to do this or multimeter in hands and check pot value according to its position?

Calculation of values is actually very difficult, but you can use a simulatot, such as LTspice.

Ra and Rb were selected according to psu headroom.... +&-18 .... so cut and boost will not exceed

This does not make much sense. Ra & Rb determine the max amount of boost/cut. In order to relate it to headroom you need to know the signal level, which is an unknown.

 

[Brian Roth]

I don't know how to approach this.....

I had been reading Barry Porter's website approx. 20 years ago when I learned he had passed away. I quickly saved everything I could. Very soon it all had disappeared from the internet. I still have my collection (such as it is!) on my current hard drive.

On his website, Barry made more than a few mentions about passing his concepts into the public domain.

So, I'm in a quandary about posting his "salvaged" website materials. It's an interesting read and shows a few things he was designing just before he died. I checked Archive.org but they are missing many elements of Mr. Porter's website.

I also have several of his articles which are scans from "ETI" which I assume was a magazine from the 1980's era. They are very interesting.

Comments on how to proceed? I want to be respectful to other's IP rights.

Bri

EDIT: It seems that archive.org has ETI stored there. I guess we'll be OK if I upload the Porter article scans here.

 

[gyraf]

My route would be to bring it forward, share in a "limited" forum like the tech. documents section (afaik only members can read this part) - then be prepared to delete if anyone with certified complaining rights complains. I have no doubts that groupdiy would've been his sort of taste..

(ETI, Electronics Today International, was one of my favorite magazines during the 80'es. Would go to my library just to read this)

/Jakob E.

 

[Brian Roth]

My route would be to bring it forward, share in a "limited" forum like the tech. documents section (afaik only members can read this part) - then be prepared to delete if anyone with certified complaining rights complains. I have no doubts that groupdiy would've been his sort of taste..

(ETI, Electronics Today International, was one of my favorite magazines during the 80'es. Would go to my library just to read this)

/Jakob E.

OK:

https://groupdiy.com/threads/barry-porter-website-and-articles.81558/
Bri

 

[richiyobs]

Ra and Rb setted to full fill the meter on my console in boost mode.. Used +4dBu as reference signal... I have no other ways to calibrate it
6dB are more than enough to reach distortion...

But what about using this kind of filter in a detector?
Respect to a standard Hpf noticed that you can have a sort of extra threshold for the freq you are filtering out.... Reversed boosting, more sensible to a certain range....
Is this a wrong way to drive the side-chain?
Best

Edit: on Spice setted Ra Rb to have 15dB of cut and boost... Too much

 

[abbey road d enfer]

Ra and Rb setted to full fill the meter on my console in boost mode.. Used +4dBu as reference signal... I have no other ways to calibrate it
6dB are more than enough to reach distortion...

Still doesn't make much sense.
A +4dBu signal should reach "0" on a VU-meter? Then "full-filling" it would mean +7dBu (to reach the +3 - or 100% - mark).
Then how can you read +6?

But what about using this kind of filter in a detector?
Respect to a standard Hpf noticed that you can have a sort of extra threshold for the freq you are filtering out.... Reversed boosting, more sensible to a certain range....
Is this a wrong way to drive the side-chain?

In the side-chain, you will find there's no single EQ setting that works on every program.The most common approach is to provide several (often switchable) High-Pass frequencies.
The so-called "thrust" filter is the closest to a "one-size fits all" solution. It's a +3dB/octave filter, constituted of overlapping 1st-order filters. However it cannot be left in permanently on some programs.
A tilt filter has a similar form of response. It's a one-knob solution.
As often, there is more than one way to skin a cat, you must use a solution that answers your personal set of constraints.

 

[richiyobs]

Fine thanx for explaining the detector question....
I don t read 6dB on the meter... It s my choice to have only 6dB of cut/boost

 

[JohnRoberts]

But what about using this kind of filter in a detector?
Respect to a standard Hpf noticed that you can have a sort of extra threshold for the freq you are filtering out.... Reversed boosting, more sensible to a certain range....
Is this a wrong way to drive the side-chain?
Best

+1 to what Abbey said... When designing side chain EQ we are motivated to KISS to avoid too much complexity, and/or front panel controls. One approach that provides utility and simplicity is a "tilt" control, that looks like it sounds, alternately boosting one range while dipping the other. While not perfect this can dramatically improve sensitivity to enhance or ignore sound sources within the side chain. Since this EQ is not on the primary audio path it does not have to sound good, it just needs to shape the response effectively.

JR

 

[richiyobs]

Great... Thanx for all your support and patience
Best

 

[richiyobs]

Still doesn't make much sense.
A +4dBu signal should reach "0" on a VU-meter? Then "full-filling" it would mean +7dBu (to reach the +3 - or 100% - mark).
Then how can you read +6?

your re right ....I said that badly
at first turn on, Ra and Rb were setted to have 18dB of cut/boost according to simulation
and so it was....
the choice of only 6dB cuase it s enough for my needs and max line level i m sending to it... that s why i said reached distortion

thanx for attention
best