Double-Pole BMP Tonestack: Can it be done?

[thehallofshields]

I once read somewhere that the Sound Engineer's 'Rule of Thumb' for EQ was:

Cut Narrow, Boost Wide.

Now, we all have our gripes that we can't find and in-between of Mid-Scoop Character, and Mid-Range Presence with standard BMP Tonestack.


LPF: -6db/Octave at 400hz
HPF: -6db/Octave at 1600hz

Now, my dream would be to keep some characteristic scoop around 1khz, but tighten it up with something like:


LPF: -12db/Octave at 800hz
HPF: -12db/Octave at 1200hz

I finally read a textbook chapter on RC Time Constants and can do the basic math, but I'm a long way from complex filters. I'm sure someone will come along and tell me that Impedance issues make it impossible to pan between Second Order Filters, but that's why I'm here.

How realistic is this idea?

[knutolai]

How about 2. order Sallen-Key LPF and HPF in parallel with a blend between the outputs?

[thehallofshields]

So use a Dual-Opamp for the job? I guess that would solve any funky Impedance issues.





To be honest I don't think I could figure out how to Bias those without a Dual-Supply...

[Mark Hammer]

It is certainly feasible in both active and passive fashion.  Naturally, the gain-recovery output stage would likely need to be goosed a little to compensate for the passive loss through an entirely passive 2-pole tonestack.

I guess the more critical question is whether such an arrangement would actually be desirable.  As is, the 7:00 max-bass position of the tone control is very muffled and only redeemed by the fact that the slope is 6db/oct, such that some mids and highs get through.  A steeper slope might remove all audible definition from the max-bass position and lead one to avoid using it altogether.  Likewise, even the 5:00 max-treble position has some bottom and guts to it.

That said, the OP's gut instinct to move the corner frequencies closer to their meeting point is, to my mind, an appropriate one.  However, perhaps a more usable control would be produced if the additional stages were not at the same corner frequency as the original ones, leading to a compromise gentler slope.  So, f'rinstance, 6db/oct at 500hz and a second single-pole at 1khz.

[midwayfair]

There's absolutely no need to go to a second order filter if you're just trying to get a deeper scoop. You can tweak the values of the caps and resistors (including the pot's value) to make the cut as deep or as shallow as you want it. It'll slightly change how the sweep acts, but I've fiddled with it in Duncan's tone stack to get even a -18dB cut that was pretty narrow. I highly suggest downloading that program and experimenting.

[anotherjim]

Bias for the Sallen-Key on single supply is easy.
The High pass, change the ground of R2 to Vcc/2 supply.
The Low pass can either get it's bias by DC coupling from a preceding stage or via a resistor feeding Vcc/2 to the input side of R1 and that should be after an AC coupling cap.
       C   R1
In  -¦¦-WW-
         ¦
        Z Bias resistor 10k-100k? Depends on how much you can load the source.
         ¦
     Vcc/2
ASCII art isn't my strong point 

[PRR]

> pan between Second Order Filters

The "problem" is that this *may* give an infinite notch.

The 1st-order filter has 90 degree phase-shift at infinity and 45 deg near the corner.

The 2nd-order filter has 180 degree phase-shift at infinity and 90 deg near the corner.

Comparing/combining *two* *2nd*-order filters, you find 45+45+45+45 = 180 degree phase difference. Which cancels.

You wanted a "scoop", instead you have a deep axe-cut.

Go ahead and try it. Certainly won't blow-up. Some mis-matching may ruin the null, for a non-infinite notch. There's also interaction between the two sides which is too much figuring for my matchbook.

If you wanna get fancy, you might buffer, one side inverting. This might flip the null into a peak, but I'm not sure.

Maybe ignore the fact that the BMP TS "can" scoop, set it up for minimal scoop, then build a scoop-only stage with adjustable depression.

[thehallofshields]

There's absolutely no need to go to a second order filter if you're just trying to get a deeper scoop. You can tweak the values of the caps and resistors (including the pot's value) to make the cut as deep or as shallow as you want it. It'll slightly change how the sweep acts, but I've fiddled with it in Duncan's tone stack to get even a -18dB cut that was pretty narrow. I highly suggest downloading that program and experimenting.

I've got the TSC Intalled. Were you able to get the narrow Q and deep Notch on the Big Muff setting with the default 1k Z?

[thehallofshields]

That said, the OP's gut instinct to move the corner frequencies closer to their meeting point is, to my mind, an appropriate one.  However, perhaps a more usable control would be produced if the additional stages were not at the same corner frequency as the original ones, leading to a compromise gentler slope.  So, f'rinstance, 6db/oct at 500hz and a second single-pole at 1khz.

I know this is old news, but people often prefer to bypass the tone control all together to reclaim their Mid-Range and component values for a Flat-Mid response has also become popular in the Muff Boutique market. Many complain that the these kill the scooped character.

My instinct is to tighten the bandwidth of the notch, from 400 -> 800 on the Low Pass, and from 1800 -> 1200 on the High-Pass. Of course to compensate we'd want to deepen the notch.

[thehallofshields]

It is certainly feasible in both active and passive fashion.  Naturally, the gain-recovery output stage would likely need to be goosed a little to compensate for the passive loss through an entirely passive 2-pole tonestack.

I've got a TL072 doing the input and output, so gain is not an issue.

Online, there are many calculators for Second Order RC Filters. Of course these are the Opamp Sallen Key type. Could I use one of those to get my values for a passive network?

[thehallofshields]

I've also noticed everyone seems to ignore that the Output Z of a CE BJT is much higher than the default 1k in the Duncan TSC.

I'm guessing that this is ignored because the end result for the Muff is Pre-Clipping Bass Reduction and Post-Clipping Bass Emphasis. -It all comes full circle.

[PBE6]

I just had to throw this in, it looks like an excellent and simple solution for developing a tailored BMP tone notch..and it's from NASA..just sayin'!!

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19800021082.pdf

(And honestly, I wouldn't be surprised if RG or PRR either wrote it or had an original copy just lying around somewhere, lol)

[thehallofshields]

Damnit PBE6 would you get out of here with your space filter.

This isn't rocket-science.

[thehallofshields]

Reading now. This is pretty cool. I feel like I'm reading declassified Area-51 files.

[thehallofshields]

Maybe ignore the fact that the BMP TS "can" scoop, set it up for minimal scoop, then build a scoop-only stage with adjustable depression.

Yes. I think several weeks back you linked me a page on the Twin-T Notch. I didn't get to the point of learning how to build one, but my plan B was to run a Tonestack with Flat-Mids, a simple Buffer, then a Notch Filter before going into the Gain Recovery Stage.

[amz-fx]

http://www.muzique.com/lab/dtone.htm

Also see: http://www.muzique.com/lab/atone.htm

regards, Jack