Behringer Autoquad - modding HP filter slope

[Gertius]

Ok, thanks again for the help!

I downloaded LTSpice, it looks awesome (and it´s free)! Can´t believe I hadn´t heard of it sooner!!

...the response has been dead on with the calculator result.  It gives more data that any other place I've found online.

Ok, I´ll give it another try as well. One thing that bothered me was the tiny rendering of the frequency response, I didn´t get comfortable with that.
But maybe LTSpice can help with that aspect.

Btw, abbey, entering "k" and "n" worked for me. Maybe you left a space? I entered for example 101k.

Ok, now I´m positive that I can figure out the right component values for the mod.
With both the Japanese Site and LTSpice I should be able to manage.

I´ll report back when I have news about my progress.

 

[emrr]

That calc does have 2nd order options with gain, just not 3rd order. 

 

[abbey road d enfer]

Btw, abbey, entering "k" and "n" worked for me. Maybe you left a space? I entered for example 101k.

It works for the component values, but not for frequency; for 1kHz you have to type 1000Hz and 1000000Hz for 1MHz.

 

[emrr]

Here's a quick comparison of unity gain Sallen-Key filters against the Intelligate filter, to show the slope/shape difference.  Intelligate is red.    The panel markings for frequency on the Intelligate are a joke, and not to be trusted. 

 

[Gertius]

It works for the component values, but not for frequency; for 1kHz you have to type 1000Hz and 1000000Hz for 1MHz.

Ah yeah, sorry, I missed that bit in your original post.

Here's a quick comparison of unity gain Sallen-Key filters against the Intelligate filter, to show the slope/shape difference.  Intelligate is red.    The panel markings for frequency on the Intelligate are a joke, and not to be trusted.

Thanks for still posting that, emrr! Very revealing. Maybe the markings on the Intelligate´s front panel still mean the -3dB point, which differs from the Butterworth implementation? Or are they way off?

Anyway, I´m looking forward to making the changes and listen to the difference, but it could be a while...

 

[emrr]

I looked at the -3 point relative to panel marking, and it was far off.  In my limited experience this has been true of most continuously variable commercial filters.  The panel markings are rounded "feel-good" numbers.

 

[Gertius]

Hi guys,
ok, so now I have played around with LTSpice a bit and have figured out the component values for the 2nd order HPF mod.
I have attached the plot of the original Autoquad filter (blue), an ideal Butterworth (green) and the modded 2nd order Behringer with around 4dB of gain (red), like abbey road d enver proposed.

I have tested the modded "Behringer with gain" for the whole R range (2k-102k => 30Hz- around 1.5kHz) and the shape stays pretty much the same.

I can´t believe how great LTSpice is!! I just drew the 3 circuits and was able to do comparison plots in no time! (ok, the UI feels a bit clumsy at first, until you get the hang of it). And the best thing is, it gives very similar results to what emrr and I measured in reality.

Anyway, just wanted to share my progress. I will continue to do the 3rd order next, and then start soldering

Cheers!
Christian

 

[Gertius]

... and the schematic in LTSpice
(PM me if you are interested in the LTSpice file)

 

[Gertius]

Ok, I have made some progress here.
I have designed the mod with all the posted ideas in mind, first in LTSpice, and then I modified the actual device.
I modified both one of the 3rd order and one of the 2nd order channels to butterworth response for listening comparisons.
Putting the voltage divider behind the LPF worked like a charm. Having it at the input indeed made the filter curve vary a lot with the position of the pot.

In the following posts I´m attaching the schematic for the 3rd order mod, as well as the measurements comparing the original filter with a 2nd order butterworth, with a 3rd order butterworth. Please note in the schematic that 47909R is the maximum value of the pot 47k + the 909R from the schematic.

The 3rd order butterworth sounds really great, and really different to the original filter.

I´m seriously considering doing the 3rd order mod in all my 24 channels. I´m looking at 200 Euros for parts and probably 1-2 weeks of work.
I have ordered another sample pot that is now 16mm in diameter, but unfortunately a four gang. I´ll see if that fits first, and if yes I can make the decision.
Boy, it would be quite some work though!

Anyway, already the experiment was great fun, and lots to learn here! Thanks again for all your input!!
I was really happy to experience the differences between the filter types first hand in analog.

Great thread!

Cheers,
Christian

 

[Gertius]

And the measurements.

Green: original Autoquad filter
Yellow: 2nd order Butterworth mod
Red: 3rd order Butterworth mod

 

[emrr]

Nice!  Thanks for the update. 

 

[abbey road d enfer]

I have ordered another sample pot that is now 16mm in diameter, but unfortunately a four gang.

Pls note that you can remove one wafer by "uncrimping" it on most multiple gang pots.