Convert this Vox wah from bandpass to low-pass filter?

[geo]

This Vox wah schematic is a very simple band-pass filter. Can it be easily converted to allow the low frequencies through without significant changes to the circuit?




I have some plans for this.

[ashcatlt]

Well it’s not a bandpass but a resonant lowpass. The only thing limiting the low end coming through is the coupling capacitors which really should be big enough to let the lowest frequencies from a guitar pass through. You could make them bigger without hurting much.

But now your title says you want it to be a highpass, which would cut low end, but then your post says you want to allow the low frequencies through, which it already does, so I’m not sure what you’re really hoping for. Converting this circuit to an actual highpass is not trivial and would pretty much mean just totally rebuilding the thing.

[geo]

Whoops, should have said low-pass filter. It's a wah, which are bandpass filters, right? But the circuit is somehow acting to effectively create a variable capacitor. I was hoping that if it could be converted to a low-pass filter, it could be used in the tone control for a guitar with active pickups, so that you could have a knob to adjust the elbow frequency, similar to the varitone, except more flexible.

[Chman]

You are correct the Wah Wah is a bandpass filter.

The peak frequency is determined by the 0.01uF feedback capacitor and the inductance.

The virtual value of this capacitor is effectively varied by the gain of the circuit.

So the 100k pot varies the apparent value of the capacitor and so varies the peak. (Think of it like

Miller feedback cap)

But you are wrong about the Gibson varitone which gives a notch in the frequency response,

not a peak or bandpass.

[sumgai]

ash is correct, it's variable-resonance low-pass filter.  Do we really have to do an in-depth analysis here? 

But the real point is..... geo, by putting this circuit (modified or not) into a guitar, with a knob in place of the pedal, it will sound like a "cocked wah" - is that what you want?

[reTrEaD]

Dec 8, 2021 23:47:42 GMT -5 sumgai said:

ash is correct, it's variable-resonance low-pass filter.  Do we really have to do an in-depth analysis here?

If you can, perhaps you should. Chman's description seems right to me.

[sumgai]

Doped out for the night, should be able to comply tomorrow.... or as soon as I'm able to think straight. Stay tuned.

[Chman]

If I'm not mistaken, I think what geo is asking for is something like the Gibson Varitone (which introduces a notch),

BUT instead of switching the capacitors to select different 'notch' frequencies, modify the Wah pedal to produce

effectively a variable capacitor, giving a fully adjustable notch frequency.

Sounds like a good suggestion to me.

[Chman]

I decided that I really like the idea of something to give me the same effect as a Gibson Varitone BUT fully adjustable (by potentiometer, not a switch).
Since the value of the inductor Gibson use seems to be somewhat vague (somewhere between 0.5H and 15H) with unknown Q and not readily available, I chose to avoid it.

I have designed, simulated and built the following circuit and I’m very pleased with the clean sound produced.

[MattB]

Jan 24, 2022 15:09:55 GMT -5 Chman said:

I decided that I really like the idea of something to give me the same effect as a Gibson Varitone BUT fully adjustable (by potentiometer, not a switch).

I agree, it's a great idea

I don't know much about active electronics, but I can follow a schematic, and this looks like it would be a fun project to build. Being able to precisely dial in the notch frequency should make this a lot more useable than having just five fixed settings. I've been planning on learning more about electronics so this is a good reason for me to start. Hopefully by the time I get round to building it I'll understand it a little better.

[Chman]

Go for it MattB.

I find that my prefered notch frequency is in the 400 to 600 Hz region,

but it's probably a matter of taste/sound.

First transistor has just a small amount of gain (x2) and to buffer the filter.

The filter itself is simply a Bridged-T circuit.

Second transistor makes up the signal lost in the filter section (approx x30)

Chman

[geo]

Dec 9, 2021 4:22:01 GMT -5 Chman said:

If I'm not mistaken, I think what geo is asking for is something like the Gibson Varitone (which introduces a notch),

BUT instead of switching the capacitors to select different 'notch' frequencies, modify the Wah pedal to produce

effectively a variable capacitor, giving a fully adjustable notch frequency.

Sounds like a good suggestion to me.

This sounds kind of beautiful, but what I was describing was a low-pass filter with a variable cutoff frequency. Essentially, a guitar tone control where you can adjust the capacitance to get any elbow frequency you choose. So maybe comparing to the varitone was a bad idea.

I'd love to hear what the notch filter you built out sounds like though. I can't picture it in my head. Imagine sweeping it while you held a note with a little vibrato...

[Yogi B]

Jun 30, 2022 12:29:12 GMT -5 geo said:

what I was describing was a low-pass filter with a variable cutoff frequency. Essentially, a guitar tone control where you can adjust the capacitance to get any elbow frequency you choose.

Recently I've been thinking about something quite similar though in a slightly different context (variable simulated cable capacitance), and thus looking at capacitance multiplier circuits. The topology of the most intuitive version is below:

The triangles being the generic buffer symbol, as I'm currently unsure as to what is best to use for that function. Preferably I'd use op-amps, but to date I've got weird resonances in my SPICE simulations and don't know enough to properly explain/eliminate them — so have currently had better results with discrete circuits.

The theory is that when the potentiometer is turned towards lug 1, the output of the righthand buffer is equal (or very close to equal) the input voltage, thus the voltage across the capacitor is a very small fraction of the normal 'expected' voltage (between the input and ground), making the capacitor appear much smaller than its actual value. As the pot's wiper is rotated towards 3 the output voltage decreases, therefore both the voltage across the cap and its effective capacitance increase to their normal level.

Ideally the effective capacitance varies between zero and the value of the capacitor and is directly proportional to the position of the pot's wiper (assuming it has a linear taper). Though, in reality, going all the way to zero is difficult as it requires that the gain of the buffers is precisely one.

[sumgai]

I'm tempted to throw a wet blanket over this party, because of the simple fact that capacitors have a time constant, and that doesn't change with frequency, nor with surrounding component values.  Look it up.  Put simply, if one wishes to change the effective frequency of a cap (and do recall that time is the inverse of frequency), then they can clock the input signal to a given cap - that'll change things up, in a hurry!  Otherwise, one is depending on resonance values, and that's a crap shoot, IMO.


As to the original question, yes, the lower frequencies can be allowed through the circuit.  With one exception, you can increase the value of most any cap in the circuit, but do remain aware that changing one component can adversely affect the overall tonal output.... because it might/will interact with other components.

The exception is the 4.7mfd coupled in parallel with the two resistors and the inductor.  Decrease that value, and a greater amount of lower frequencies should show up at the output.

HTH




sumgai

[geo]

Jul 1, 2022 15:38:39 GMT -5 sumgai said:

I'm tempted to throw a wet blanket over this party, because of the simple fact that capacitors have a time constant, and that doesn't change with frequency, nor with surrounding component values.

Capacitors have a time constant when paired with a resistor! With zero resistance, the time constant would be zero. So there is a surrounding component whose value can change that time constant.

[thetragichero]

for this notch filter idea, instead of something like a twin t and then a bunch of makeup gain, wouldn't feeding a variable gyrator into the inverting input of an op amp (to me i always think of the boss heavy metal tone controls) be a little more... elegant?

[sumgai]

Jul 4, 2022 4:41:19 GMT -5 geo said:

Jul 1, 2022 15:38:39 GMT -5 sumgai said:

I'm tempted to throw a wet blanket over this party, because of the simple fact that capacitors have a time constant, and that doesn't change with frequency, nor with surrounding component values.

Capacitors have a time constant when paired with a resistor! With zero resistance, the time constant would be zero. So there is a surrounding component whose value can change that time constant.

Ah, I see that you're prepared to show us a 100% resistance-free circuit.  Eager anticipation only begins to describe my feelings at this moment.

Look geo, I don't throw out statements like this lightly. I spent nearly all of my 4th decade as a college professor of everything electrical and electronic. The challenge is upon you to prove your statement makes mine suspect. But I won't be "arguing my case", that'll be up to the rest of the Nutz, as I have a tough enough time arguing with my wife these days.  Summer is already over, it lasted only a week at my location, and I still haven't finished painting the house.  But hey, Happy Independence Day to you, sir.

And to all the rest of the Nutz as well!




sumgai
Chief Resident of non compos mentis ville.

 

[geo]

Jul 4, 2022 10:28:21 GMT -5 sumgai said:

Jul 4, 2022 4:41:19 GMT -5 geo said:

Capacitors have a time constant when paired with a resistor! With zero resistance, the time constant would be zero. So there is a surrounding component whose value can change that time constant.

Ah, I see that you're prepared to show us a 100% resistance-free circuit.  Eager anticipation only begins to describe my feelings at this moment.

I certainly cannot reduce the resistance of the circuit to send that time-constant to zero... but I could increase it.

[Yogi B]

Jul 1, 2022 15:38:39 GMT -5 sumgai said:

Put simply, if one wishes to change the effective frequency of a cap (and do recall that time is the inverse of frequency), then they can clock the input signal to a given cap - that'll change things up, in a hurry!

I was being somewhat conscious of what's likely to already be in people's parts drawers but I guess it isn't much of a stretch for that to include a 555 timer — assuming that using that for a variable duty cycle 50kHz+ oscillator (plus one or two transistors for the actual switching) would be adequate.

Also, somewhat tangentially, is it possible to simulate negative capacitance via the switched cap method?

As to the original question, yes, the lower frequencies can be allowed through the circuit.

I, perhaps, should also have been explicit in noting that what I was describing is also pretty much the underlying principle of typical the wah circuit. (Essentially just expanding on Chman's first reply to this thread.) And therefore a statement of intent as to how I'd aim to mod the original circuit.

Jul 4, 2022 7:13:19 GMT -5 thetragichero said:

for this notch filter idea, instead of something like a twin t and then a bunch of makeup gain, wouldn't feeding a variable gyrator into the inverting input of an op amp (to me i always think of the boss heavy metal tone controls) be a little more... elegant?

For me the first thing that comes to mind is JHS's Haunting Mids or maybe the old Catalinbread Varioboost. Both of which are a similar arrangement (but based around an inverting stage) and already feature the ability to sweep the centre frequency of the notch/peak.

(Also, speaking of the the Haunting Mids — which I believe uses both a TL072 & an NE5532 — I swapped out the former for the latter in my SPICE sims and got much better results. And in terms of circuit complexity you have to admit that it's tough to beat: 4 resistors, 4 caps, a dual op-amp, and a pot.)

Jul 4, 2022 12:11:26 GMT -5 geo said:

Jul 4, 2022 10:28:21 GMT -5 sumgai said:

Ah, I see that you're prepared to show us a 100% resistance-free circuit.  Eager anticipation only begins to describe my feelings at this moment.

I certainly cannot reduce the resistance of the circuit to send that time-constant to zero... but I could increase it.

As sumgai is hinting at there's always going to be some resistance (even the cap's own equivalent series resistance), unless your pickups are wound with specialist alloys and you're playing at temperatures of only a few degrees kelvin (too cold, even for newey ).

The potential issue with directly altering the time-constant with a variable resistor is that'll only sweep a first order filter and thus likely won't sound very 'exciting'. The appeal of a wah isn't just the variable roll-off: since it's a resonant filter, it's also the accentuation of a narrow band of frequencies. Also, unless the capacitance you're adjusting remains directly in parallel with the pickup you'll have the problem of either destroying or competing with the pickup's own natural resonance.

[geo]

Jul 5, 2022 2:06:41 GMT -5 Yogi B said:

Jul 4, 2022 12:11:26 GMT -5 geo said:

I certainly cannot reduce the resistance of the circuit to send that time-constant to zero... but I could increase it.

As sumgai is hinting at there's always going to be some resistance (even the cap's own equivalent series resistance), unless your pickups are wound with specialist alloys and you're playing at temperatures of only a few degrees kelvin (too cold, even for newey ).

The potential issue with directly altering the time-constant with a variable resistor is that'll only sweep a first order filter and thus likely won't sound very 'exciting'. The appeal of a wah isn't just the variable roll-off: since it's a resonant filter, it's also the accentuation of a narrow band of frequencies. Also, unless the capacitance you're adjusting remains directly in parallel with the pickup you'll have the problem of either destroying or competing with the pickup's own natural resonance.

I certainly understand that there's going to be non-zero resistance. I was just working my way to the point that there is a time-constant which I can adjust upwards by increasing resistance and the time constant should change exactly the same as if I had increased capacitance.

This isn't for a wah. I'm not interested in creating a bandpass or notch filter; I'm looking to create tone control with an adjustable elbow frequency. So instead of swapping out the tone caps, you'd have a trim pot you could adjust to move the elbow frequency.

[Yogi B]

Jul 5, 2022 5:22:37 GMT -5 geo said:

I was just working my way to the point that there is a time-constant which I can adjust upwards by increasing resistance and the time constant should change exactly the same as if I had increased capacitance.

What I was getting at is that though that is true, it is only true when you have only resistance & capacitance present, and that isn't the case in a guitar.

In normal circumstances a pickup's inductance has greater responsibility for how an external load will affect the treble frequency response than its resistance. (For example, at 1kHz, the magnitude of the inductive impedance of a typical strat pickup is roughly twice that of its resistance — and more like 3 or 4 times with a humbucker)

Sure, you can easily add extra series resistance to a pickup to decrease the cutoff frequency of a tone cap (this being the cause of the problematic tone control behaviour with '50s wiring), but this isn't simply the same as having a larger cap. You'll also unavoidably change the response from one which is primarily a result of an interaction between inductance & capacitance to that which is primarily a result of resistance & capacitance.

[geo]

Unless you buffer the signal, similar to the diagram you shared.