Pre-post clipping bass control

[PBE6]

Just thinking about bass filtering on distortion effects as discussed many times by Mark Hammer and others, trying to come up with something that would cut bass from the front-end but add it all back in on the back-end. Here's what I came up with:



This is just a sketch, so it's missing the usual coupling capacitors, pull down resistors, etc.. but the basic idea is:

1. Start with a buffer, then a shelving high-pass filter to cut bass, and then another buffer or gain stage to fix impedance issues.

2. Clip the signal.

3. Add bass back in with a non-inverting gain stage.

By adjusting the bottom resistors in each respective voltage divider, you can adjust how much gain is cut and then added back in without affecting the cutoff frequency. (The diagram isn't that clear, but the two potentiometers are meant to be two halves of the same ganged pot.)

In theory, so long as all the resistors and capacitors match up between stages, the bass boost will be equal and opposite to the bass cut, essentially just "shifting" the bas content from front-end to back-end. This should reduce fuzziness while maintaining bass content in the final signal. Of course, intermediate gain stages and clipping may make the balancing act a bit more complicated...

Is this an old idea? I'm going to try breadboarding it anyway to see what happens.

[GibsonGM]

I think it's AWESOME to breadboard just about ANY idea you have, PBE!

The concept of reducing bass prior to clipping is certainly not a new one, as you referenced Mark's writings on it.  Same thing with wanting to increase bass response later on...where would we be without that?   I haven't seen this particular idea for 'putting it back in', however!    The dual-ganged pot makes your idea pretty simple (hey, 1 knob!).    Ya, in practice it might get tricky, and there will be things you'll have to tweak, of course.   You'll probably have to work out where to put bias voltage, and so on.   Mostly this is done "on the bench", set and forget.  Your idea gives you more options, esp. if using different guitars/amps.

Have you 'simmed it' in LT Spice or anything?  Could be educational.

Go for it, I'd really like to know how it turns out for you!  There are no bad ideas, and no single way to get to the same place.

"OA3" could be a gain stage rather than a buffer as you want good level for clipping - not sure if you just drew it that way for simplicity but thought I'd suggest it...

[ashcat_lt]

I'm wondering why you wouldn't use an active stage for the bass cut also.  Like, why not just put that filter upside down in the feedback loop of one of the opamps.  You could then get rid of one of the opamps and you'd end up with a very familiar looking distortion unit followed by a complimentary bass boost.

[PBE6]

Sounds like an interesting alternative, what exactly do you mean by putting the first filter "upside down" in the feedback loop of an opamp? Right now both filters are identical (in the audio range) and have opposite and canceling effects, but they're both "right side up" from my perspective.

Just thought of something, two connected bass-only Baxandall loops might do the trick as well, and would allow the bass to be front end loaded too. Maybe a modified version of that is what you're talking about?

[aron]

I think Mark mentioned a few times about high end roll off as well. Pre-clipping.

[ashcat_lt]

Sounds like an interesting alternative, what exactly do you mean by putting the first filter "upside down" in the feedback loop of an opamp? Right now both filters are identical (in the audio range) and have opposite and canceling effects, but they're both "right side up" from my perspective.

Everything's upside down when put it in a negative feedback loop!  What you actually need to do is re-arrange those same components so that they form a shelving low pass and then stick it in the feedback path so it turns into the shelving treble boost that you actually need.  See any opamp-based od/dist pedal.

[Brisance]

Only thing, why not use a TL072 for two stages, otherwise you end up with 3 ICs

[GibsonGM]

He's just setting up the concept right now - he's probably going to use a dual opamp for sure.  Maybe TWO, and find a use for the extra one

[Gus]

Something I posted here before

Note the high pass is a little different I like this type of highpass before the circuit bass with a fuzz
This was posted for circuits that use Ges with low output that is why the last stage is x2

IMO this is a good idea buffer, eg, dist/fuzz, buffer/eq etc.

[GibsonGM]

That's a good one, Gus. I've learned a lot from your ideas over the years    PBE has some things to think about now!

[PBE6]

I'm looking for a symmetrical shuttling of bass from pre- to post-distortion, so the topology has to be symmetrical - hence the HP shelving and LP shelving boost idea. Mixing and matching topologies runs into asymmetry problems when you're trying to control the amount of bass shift.

A modified version of the Baxandall control is symmetrical and easy to control with a ganged pot, and only uses 2 opamps:



The PRE and POST pots would be controlled by the same ganged pot. The GAIN pot provides the boost. The simulation shows the bass shifting to be symmetrical, which achieves the original goal. We'll see if it sounds any good sometime this week.

[ashcat_lt]

So I simmed your first scheme, and I don't think either of the filters you've drawn really work the way you intend, and it definitely doesn't do what you're asking.

First let's look at the boost section.  When the pot is at 0R, it turns into a treble booster.  I haven't had enough coffee to explain except that it has to do with the LPF created by R5 and C3.  If you swap the pot and R5, it works great!

Now that you've got that, if you then pull out C3 and replace it with C2, you end up with a treble boost which works exactly opposite to the bass boost.  Use this for the first opamp stage, and...we'll get to that in a minute.

The bass cut control in the sim does almost nothing for the first 90% of its (linear) travel.  You might say you could use a log pot or whatever, but that doesn't change the fact that it just plain doesn't have the same effect for a given rotation as  the boost, so they don't actually end up at unity.  Also, when you turn the pot all the way down, you've shorted the entire thing to ground, and should get silence.  Swapping the fixed resistor with pot helps here, too, and actually ends up with a fairly flat response (much less than a db variation) throughout the sweep.

With the two active filters in series, it comes pretty close to flat too, but there is (at worst) almost 2db bump right around the cutoff at a point somewhere around 5% of the pot's resistance - very close to 0 on the knob.  You'd probably have to build it and hear it to know if that's a problem.

Hope this helps.

[PBE6]

Aww crap, you're right! Got too excited and only checked one end. Ok thanks, I will give these modifications a try.

[PBE6]

.