Tubescreamer tone control - why?!?

[ElectricDruid]

I've got a question about the Tubescreamer tone control - why?!?

It doesn't seem like it really *does* that much, or at least not much beyond what a simple passive pot+cap would have done. So I suppose what I'm asking is why do you think they went for this more complicated active design? And what's that weird-looking thing with the pot between the two op-amp inputs all about?! It just seems like a really odd thing to me, and I can't imagine why they came up with it or did it that way. Any insights or opinions welcome!

Here's the circuit:


And here's what it does:

[antonis]

It doesn't seem like it really *does* that much, or at least not much beyond what a simple passive pot+cap would have done.

[ElectricDruid]

Thanks Antonis, but I've done my research. I'm just not *convinced* by this control.

Where's this supposed "bandpass" filtering effect? It doesn't really show up on Electrosmash's graphs, and it doesn't really show up on mine either. The effect is feeble, frankly.
And the "second order lowpass filter" is a fancy way of saying we've got one RC lowpass (R7/C5) followed by another. Amazing!

It seems to me like the BMP tone control does more with less. It looks like this design is trying to do something similar (pan from highpass at one end to lowpass at the other) but it just doesn't seem to be as effective.

Maybe it's time to start tweaking the sim and see if we can't make it a bit more *exciting*.

[Radical CJ]

I think the simple reason for the modest effect of the tone control is that that is all that's needed for the purpose of dialing in a good sound from a guitar... that, and it uses up the other side of the dual opamp.

Consider the Metal Zone, it has a highly adjustable eq, but perhaps too powerful for practical use. Its best not to move any of the eq knobs more than a few minutes from 12 o'clock!

[ElectricDruid]

Ok, I've done some more experiments in the sim. The results are (I think) quite interesting. In my view, there are two problems with the design.

The first problem is basically C5 from the Electrosmash schematic, that RC lowpass on the way in. It is set at 720Hz (incidentally, the same frequency as the highpass in the drive stage) so it completely flattens any treble ahead of the tone control, leaving the tone control little to work with. If we remove that cap, here's what the tone control does:



So...not bad, right? It's a baxandall-style response, with roughly 15dB of cut and boost. That should do something. That brings us to the second problem - the pot response. If it was baxandall, the flat setting would be in the middle. Here's what actually happens:



Basically, the whole action of the pot goes from cut to zero, and then when you get to the extreme treble end, you get some boost. It makes no odds whether you use a log or a lin pot, I tried it.
Of course, this doesn't actually *matter* that much, because as soon as we put that C5/220n cap back in, we completely flatten the whole thing anyway and what's left doesn't do much - this is the actual response as designed:



Gagh! Seems to me a baxandall-style treble control would have been a better idea. It'd give the same overall tone, but much better pot response. And backing off the C5 value significantly would allow the tone control to have room to breath. In short, I think there are definitely some tweaks to be had from this tone circuit.

CJ, I do largely agree, and perhaps we don't need +/-15dB of treble (certainly not +15dB of treble!) but there's really no reason for the pot *response* to be so mucked up. You could easily tweak resistors at the ends of the pot in the standard baxandall circuit to give you -15dB to +5dB, for example. And C5 could be slightly higher than 720Hz and give the tone control some space without having to let everything through. At the moment, it doesn't look like the tone control really does a lot. For example, at 10KHz, the response over 95% of the pot travel varies from -22dB to -18dB (ignoring that one extreme setting when you get to the far-treble end). 4dB is nothing. Allowing for that extreme-treble setting, we get -22dB to -8dB, a more useful 14dB range. Except that we can't sweep that 14dB range smoothly at all. It's just hopeless!

[antonis]

Thanks Antonis, but I've done my research. I'm just not *convinced* by this control.

I'm pretty sure you're well informed about it Tom but, honestly, I can't get into your query..

Mr S. Tamura (at very late of '70s) might considered that an unused amp serving for an active filter configuration might be good for a market target..

P.S.
What puzzles me is the ridiculously low (even for bi-polar inputs op-amp) bias resistor(s) value (10k) which severely downgrades C5 (and C6 at fully Bass side) impedance at low frequencies..
(probably designer took it into account for his calculations..)

[Lino22]

Guys there is a high pass filter formed in the feedback loop on the treble side. Why does it work like a high pass filter when it shunts high frequencies to the ground? Sorry for a noob question.

[antonis]

It does and it works so..
(just check green framed in my schematic above..)

edit: R8 + C6 are set in the denominator of the Gain formula 1 + R11/(R8 + C6)
The higher the frequecy the lower the C6 impedance hence the higher the gain - and vice versa..

If you replace C6 with its respective impedance for certain frequencies, you'll verify its operation as HPF..
(same stands for feedback resistor shunt capacitor, C4 in Electrosmash schematic, but it works as LPF due to its placement inside NFB loop..)

[Mark Hammer]

The Boss OD-1 is essentially a Tube Screamer, with a 2+1 diode complement rather than 1+1.  The 2nd op-amp stage provides a fixed treble-cut function, with only a Gain and Volume control, rather than the 3 knobs of the TS and OD-1 derivative of the SD-1. 

The OD-1 is, to my ears at least, a pleasing tone, although my own build has a switch to add some bottom, or remove some top end in addition to adding bottom.  Bear in mind that, as near as I can tell, the design of the gain/clipping stage, with the ground-leg bass cut and complementary output treble cut (what Tom shows as R4/C4, and Antonis shows as R7/C5) was intended to provide roughly equivalent clipping across the entire fingerboard, by attenuating the gain for wound strings and attenuating the harmonic content of unwound strings and higher notes.  In their absence, low notes would growl more, and also0 have more sizzle than notes played higher up on the unwound strings.

In turn, the Tone control appears to be intended to compensate for the tone of pickups/instruments that are bottom-heavy or conversely top-heavy, by disadvantaging treble at one extreme and bass at the other.  If the highpass filtering in the Gain stage, and lowpass filtering in the Tone stage are well-suited to the instrument (as in the OD-1) there isn't much purpose to the addition of a Tone control.

So, to respond to Tom's  question: In some cases, the Tone control is not only useful, but necessary, and in other cases it is clearly superfluous.

[Lino22]

edit: R8 + C6 are set in the denominator of the Gain formula 1 + R11/(R8 + C6)
The higher the frequecy the lower the C6 impedance hence the higher the gain - and vice versa..

Thank you Antonis, i've learned a new thing today

[Mark Hammer]

Hard to believe it's 15 years old already, but interested parties should read this:  http://www.muzique.com/lab/tstone.htm

[merlinb]

And what's that weird-looking thing with the pot between the two op-amp inputs all about?! It just seems like a really odd thing to me, and I can't imagine why they came up with it or did it that way.

That's a standard tone topology used in graphic EQs, probably done that way to make tweaking easier at the design stage since you can do a lot with it, including making it into a bandpass. Maybe they even tried that option but just didn't end up choosing that in the end. They actually did build a better EQ into the SD9.

[marcelomd]

Basically, the whole action of the pot goes from cut to zero, and then when you get to the extreme treble end, you get some boost. It makes no odds whether you use a log or a lin pot, I tried it.

IIRC that pot is W or S taper. So more "sensitive" near the extremes.

[PRR]

It is not any Baxandall. It is one unit of one of the two main ways to build Graphic Equalizers. Steve Dove classed this as "swinging inputs", but the internet has not been kind to Dove's writings. Your image is badly drawn-- conceptually there are two buses boost & cut, there are multiple pots across the buses, and a tuned circuit on each pot wiper.

http://www.geofex.com/article_folders/eqs/paramet.htm#graphic_eq
https://i.stack.imgur.com/HTy03.jpg
https://circuitscheme.com/tag/graphic-equalizer-schematic

[PRR]

> the internet has not been kind to Dove

Ah. Much hunting needed but the old scrolls have been recovered.

https://worldradiohistory.com/Archive-All-Audio/Archive-Studio-Sound/80s/Studio-Sound-1981-06.pdf  --pg 76
https://worldradiohistory.com/Archive-All-Audio/Studio-Sound.htm  --- the magazine

Also the collected Doves, but with errors:
https://www.google.com/books/edition/Handbook_for_Sound_Engineers/t__vBgAAQBAJ?hl=en&gbpv=1&dq=swinging+outputs+EQ&pg=PA1017&printsec=frontcover

[ElectricDruid]

Mark,

Thanks for the AMZ link. I'm glad it's not just me that thinks this thing could use a tweak, and that others have been here before.

Merlin, PRR,

Thanks for pointing out that it's a graphic-EQ-style circuit. I hadn't spotted that, but now you mention it, I can see it. S'funny, it just doesn't look like it with only one band! lol.

Since those graphic EQs are usually flat in the centre with equal cut and boost, I think something weird might be going on with the pot response in my circuit in the sim. I'll have to investigate that, because that definitely doesn't look right. I'm getting a defiantly lop-sided pot response.

[iainpunk]

replace the cap and resistor from the tone pot with a gyrator and capacitor and make it a boost+cut for any frequency band desired.
you could even insert the gyrators from the HM-2 circuit to get that kind of sound. while youre at it, swap the C103 and C105 (47n in opamp feedback loop and the 220n filter cap of the first static filter) to get more fizz through less filtering and more low end fuzz through more gain at low frequencies.

cheers

[PRR]

> usually flat in the centre with equal cut and boost, I think something weird might be going on with the pot response in my circuit in the sim. I'll have to investigate that, because that definitely doesn't look right. I'm getting a defiantly lop-sided pot response.

C4.

[Elektrojänis]

If you wanted to have second order low pass on the bass side there would need to be another resistor in the signal path between the caps. Two caps in parallel equals just a bigger cap, so the cutoff point gets lower. And that resistor between one of the caps and ground probably just makes the rolloff even less steep at some frequencies.

[anotherjim]

That 720Hz LPF pre-filter might make sense if there has been boosting of high frequencies in an earlier stage, due to distortion maybe. What happens if you input square waves into the filter?