Craig Andertone Super Tone Control What Is It Actually Doing to the Signal?

[replaceablehead]

Just finished building the Craig Anderton Super Tone Control. I used the pedalpcb prefabricated circuit board. Lazy I know, but I'm up for a new toner cartridge and I'm just a little over toner transfer at the moment.

I feel like there's not enough info on how this thing works. I've read Electronics Projects for Musicians and I've read about synths with state variable filters, but the operation of this particular unit still seems a little arcane.

I understand how high, low and band pass filters work. What I can't understand intuitively is this idea of "mixing" them. What I would have expected was switchable, high, low and band modes, with frequency and Q adjustment. I would have expecting a full signal that gets cut.

Instead what you have is these pots that "turn up" the filters. I don't get it. How can I raise the volume buy turning up the filters?

I would assume to achieve a flat response I should set the frequency to it's minimum with the high pass turned up, or to it's maximum with the low pass turned up. This does seem to sort of work, but it's noisy and appears to give a slight boost. Is that just because there is some make up gain in the volume control?

Also am I to assume that the band pass is quite narrow? In band pass mode is it possible to widen the cut off frequencies? Or is the resonance control just shifting the response from narrow to really narrow with a boost? Or is the resonance moving from wide to super narrow?

Are the filters ever providing a frequency boost? I would expect them to cut only aside from the resonance providing a boost at the cut off. But on listening to the higher settings with the resonance down I can hear what sounds like a boost as well as a cut.

Also the high/low resonance switch, is it just setting the sharpness of the Q? Like from a sharp spike to a more mellow spike? Is that the idea?

It is undoubtedly an awesome pedal and I feel like more people would build it if they understood it.

[PRR]

> I would assume to achieve a flat response

This is NOT for giving a "flat" response; there are far easier ways to do that.

It is "semi-universal". It can do a lot of things you may not want.

KISS. Just turn ONE HP/BP/LP pot up, leave the others down. BP and Freq will give you a variety of honky shapes. LP gives a range of dull sounds. HP a range of shrill.

HP and LP together should give a dip.

Yes, there is gobs of gain in there.

I *think* all three HP/BP/HP pots equal gives "flat" response. Unless there is a null in the overlap.

[Mark Hammer]

The first edition of EPFM contained a slightly different version of the Super Tone Control.  What shows up as variable level controls in EPFM2 is simple toggles in EPFM1.  The circuit in EPFM1 also lacks the Hi/Lo resonance switch.  I gather closure of the Hi/Lo switch, in tandem with the voltage-divider action of the resonance pot and the 12k resistor the Hi/Lo switch grounds, reduces the amount of feedback.

In principle, dropping the resonance down to the barest minimum, should get you closer to "flat", although common sense says "flat" was what the bypass switch is for.

[replaceablehead]

Thanks so much for the response, they really helped confirm what I suspected.

although common sense says "flat" was what the bypass switch is for.

I cracked up when I read that! Yes, I suppose it might seem redundant, but I think a lot of users would like to understand at what point they're getting the least effected signal and that way they have a baseline from which to make adjustments. I mean, yeah I can use my ear, but it's nice to be able to confirm what you're hearing in order to train your ears.

I think there is a null in the overlap, but I'll try that out again.

So if I leave everything off and just turn up say the HP. As I turn it up, the frequencies above the cutoff are allowed to pass, but at a certain point it starts to sound like the frequencies above the cut off have gone above unity gain. Is that right? So what you have really is a filter with a bit of boost?

The fact that the first version had switches makes so much more sense.

Also, the width of the band filter is fixed? I mean when I increase the resonance it's just boosting the center frequency, and it's not bringing the cutoff's closer together?

Because it would be cool to be able to widen and narrow the bandpass filter, for example, I would like to be able to cut everything below about 120hz and everything about 6khz. Is that possible with this Super Tone Control?

[ElectricDruid]

Because it would be cool to be able to widen and narrow the bandpass filter, for example, I would like to be able to cut everything below about 120hz and everything about 6khz. Is that possible with this Super Tone Control?

No, not if it's a state variable filter. The bandwidth of the bandpass is "fixed" in that sense. Turning up the resonance increases the peak, and narrows it a bit, but it won't really change it hugely.

You'd need two filters for that, one set to do the highness and one to do the lowpass. Then by feeding them inverted control voltages, you'd have a "width" control. Plus you'd get two independent resonant peaks.

Now, that might get quite vowely...

[replaceablehead]

Right, I thought that was the case. I can see why they use these in synth applications, but not so much for recording.

The more I play around with this thing, the more I think filters are the most underutilized effects in the guitar world. I mean sure we have wahs and treble boosts, but with so many guitarists chasing the biggest fattest tone possible, it seems like there's a world of rarely explored timbres. I want my guitar to be razor thin and metallic and this seems to deliver a lot of that.

It's got me thinking about all the other kinds of filters out there.

[Mark Hammer]

Although bopping back and forth between engage and bypass should, in principle, allow for comparison against a flat response, differences in overall level between the two states can corrupt our perception, or at least make it harder to be certain.

I will suggest nudging the feedback resistor in the last stage up from 33k to 39k, for a little more compensatory oomph, and replacing the 10k terminating resistor on the output with a 10k log pot.  That way, you can set the level of the STC to be roughly comparable to bypass.  The increase from 33k to 39k is to compensate for the sorts of notch or bass cut that drops overall level, while the volume pot is for those cases where too many humps produce a much louder signal.

[ElectricDruid]

The more I play around with this thing, the more I think filters are the most underutilized effects in the guitar world. I mean sure we have wahs and treble boosts, but with so many guitarists chasing the biggest fattest tone possible, it seems like there's a world of rarely explored timbres. I want my guitar to be razor thin and metallic and this seems to deliver a lot of that.

It's got me thinking about all the other kinds of filters out there.

Totally agree - there's more filter possibilities than typically see the light of day.

For "metallic", you could try the "static flanger" sound, AKA comb filter, AKA very short delay with lots of repeats. That puts a series of peaks and troughs in the frequency response (hence "comb filter" - that's what the response looks like) that makes the input sound like it's played down a metal tube. Change the delay time to change the dimensions of the tube, and change the amount of feedback to change the hardness of the surface. Since Phasers also produce notches, a static phaser could also give you some of the same effect, especially if it has resonance.

Ring modulation would be another way to get more "metallic". Metallic sounds often have significant amounts of anharmonic content, and that's what ring modulation can give you - that "chiming" or "clanging" sound. The trick there is getting the balance right so it sounds good and not just like backing your car into a concrete bollard (another rather less pleasant metallic sound).

[Mark Hammer]

Back in the day, I built one into a wah shell.  Makes a great sweepable filter.  If you're familiar with the older tune "One Thing Leads to Another" by The Fixx ( https://www.youtube.com/watch?v=JHYIGy1dyd8 ), you can absolutely nail that sound, sweeping the highpass downward.
This guy did a great job on his build, both visually and in terms of functionality.  He included envelope control.

[replaceablehead]

I think I saw your build on another thread Mark, was it the wah housed in the machine foot pedal that a pawnshop had mistaken for a wah? That was pretty sweet. I wanted something more like a parked wah, but I have been thinking about expression pedal possibilities.

I've got a Mooer E-lady with filter matrix mode that is apparently a type of comb filter. I don't find it all that usable those, it's fairly extreme. On day I'll have to build a ring mod, everyone always demo's the most extreme settings, but I've been curious to find out what milder settings sound like.

Although bopping back and forth between engage and bypass should, in principle, allow for comparison against a flat response, differences in overall level between the two states can corrupt our perception, or at least make it harder to be certain.

So what are we saying the flattest response setting is with the stock setup? Is it all filters maxed with the resonance at minimum? Reading back through I feel like we never totally clarified that. I can accept close enough as good enough, I just want to have some idea of the starting point.

[Josh?]

One day I'll have to build a ring mod, everyone always demo's the most extreme settings, but I've been curious to find out what milder settings sound like.

Slightly off topic, but I think this demo nicely showcases how a ring mod can be used for a less in-your-face sound if you haven't seen it yet:

[replaceablehead]

That demo is exactly what I wanted. Thanks for sharing.

I feel like sometimes you just have to build it to really understand it.

[Mark Hammer]

I think I saw your build on another thread Mark, was it the wah housed in the machine foot pedal that a pawnshop had mistaken for a wah? That was pretty sweet. I wanted something more like a parked wah, but I have been thinking about expression pedal possibilities.

Nah.  That wasn't me.

Although bopping back and forth between engage and bypass should, in principle, allow for comparison against a flat response, differences in overall level between the two states can corrupt our perception, or at least make it harder to be certain.

So what are we saying the flattest response setting is with the stock setup? Is it all filters maxed with the resonance at minimum? Reading back through I feel like we never totally clarified that. I can accept close enough as good enough, I just want to have some idea of the starting point.
[/quote]
Given the method of adjusting the contribution of each filter output, in tandem with the way that varying the resonance inserts/decreases humps into the  response, I think it's pretty hard with the circuit to be able to set the controls in some visibly identifiable way that would yield a strictly flat response.  That's partly why I suggest installing a volume control, as a means to at least factor out volume differences when trying to dial in something closer to flat.

[replaceablehead]

So no matter what there's always a filter engaged. I can set the cut-offs so low or, so high that it's letting almost all the existing frequencies through, but no matter where I set the knobs somethings always getting cut somewhere.

It seems obvious now, it's only that in the book Craig say something about it being "easy to set back to a flat response" or something like that. Is he just talking about hitting the footswitch?

What's with the sound diminishing as you get into higher frequency settings? Is that because there's not very much high-end content in the original signal to pass?

[Mark Hammer]

That may be what Craig is implying.  I can't say for sure.

As for the sound "diminishing", any time you remove significant portions of the spectrum, there will be the sense that the overall amplitude has been reduced.  That will be true of removing the mids and highs with a lowpass setting, but even moreso for a highpass filter since the brunt of the guitar signal lives in the low end.

[replaceablehead]

Yeah, that's exactly what I thought, but I just thought I'd ask in case there was anything else coming into play in the higher settings.

Does anyone know how the STC compares to the Systech Harmonic Energizer? It's a bandpass filter with aggressive Q settings right?

Are all these filters more or less the same thing? I mean wah pedals often have inductors, does this create a very different type of filter, or are they all very similar?

Sorry for all the questions, I feel like I'm not giving enough back. All I can say is it's a great pedal and more people should try it out.

[snk]

Hello,
I have built both the Super Tone Control and the Systech Energizer.
They do not sound the same at all

The STC is a multimode filter, gathering HP, BP, LP into a single unit, and allowing you to dial in individual volume for each filter. It is a very interesting unit, and a great tool for shaping your sound, but if you are not familiar with filters you may need to spend some time with it as it allows for a wide range of sounds.
It's a quite clean sounding filter, imho.

The Energizer is "only" a bandpass filter with lots of character. It is a more rough sound shaper, with a great sound, and prone to go wild at any time.

So, both are filters, but 1-they do not have the same sonic character, and 2- they do not have the same features, so there is little overlap between them.

One thing to take into consideration while dealing with your issue with the STC :
1- having 3 filters at the same time at once means that you will have crossovers. One filter keeps the lows, one filters keeps the mids, and one filter keeps the highs, but at the crossover points you may have bumps or dips in the frequency range. In theory it might be flat, but in practice it's quite unlikely
2- Beware of the resonance : if your goal is to reach "frequency flatness", be sure to make your tests without any resonance (as resonance will increase the frequency amount around the cutoff frequency).

[replaceablehead]

mmm its interesting how filters can be so different sonically. I suppose a lot of filters that are designed for guitars are not high fidelity like the ones used in studios.

One thing I've noticed, it appears in the original schematic and the schematic that pedalpcb used, the resonance control is backwards so to speak. So when its full clockwise there is no resonance and you turn it counter clockwise to increase the resonance. I've seen a few comments scattered over the internet to this effect, but I thought I'd restate it here so people building this thing don't go tearing their hair out thinking they've miss-wired the pot.

[ElectricDruid]

mmm its interesting how filters can be so different sonically. I suppose a lot of filters that are designed for guitars are not high fidelity like the ones used in studios.

True, at least in my experience. A lot of the "synth filter" designs go a lot further in aiming for audio quality, whereas a lot of the "FX" designs seem mostly to minimise parts and distortion is an acceptable price to pay. That said, different filters *do* sound different regardless - a transistor ladder is *not* the same as an OTA filter, etc etc. People overplay this stuff, but it is true to some extent. It's just *more* true the harder you push things, I guess!

One thing I've noticed, it appears in the original schematic and the schematic that pedalpcb used, the resonance control is backwards so to speak. So when its full clockwise there is no resonance and you turn it counter clockwise to increase the resonance. I've seen a few comments scattered over the internet to this effect, but I thought I'd restate it here so people building this thing don't go tearing their hair out thinking they've miss-wired the pot.

Yes. This is not universal, but it is a common way to lay out the feedback paths in a state variable filter. The "Resonance" pot is not actually "resonance" but rather "Damping" and acts to *cancel* feedback that is already produced by the feedback path from the lowpass integrator stage. Hence more damping = less resonance = the pot works backwards.

This is not entirely universal - I've seen a Roland modular synth filter that had a resonance pot in the lowpass path, so it has the "more feedback=more resonance" response, but for SVF's in general, the pot from the bandpass stage and hence the  "reverse wiring" is more common.

[replaceablehead]

I just thought of another burning question. Has anyone got any idea what the sweep-able frequency range actually is on this thing?