SWTC Tone Control / Treble Boost

[fuzzy645]

I'm rather intrigued by this, i particular the treble boost tone pot:  http://www.muzique.com/lab/swtc.htm

I have 2 questions:

1.  What would the effect be of (lets say) doubling the cap from .022 uf to .047, or even higher? Would it increase the treble boost?
2.  It seems to me this is a passive tone control, correct?  If so, then how can it possibly boost highs?



Here is the exact text from the site:

By moving the C1 capacitor, we have totally changed the response of the tone control circuit. With the wiper of R1 all the way to the right side, the frequency response is totally flat as if the capacitor is not in the circuit, which it isn't because the two ends of C1 are shorted together in that position. As the wiper is moved to the left end of the pot, proportional amounts of treble boost are added to the circuit.

With the new "Stupidly Wonderful Tone Control 2", the volume always remains the same but the amount of treble is controlled from flat to boosted as the pot is adjusted. There never is a low frequency rolloff with this circuit.

[CynicalMan]

1. Increasing the capacitor's value reduces the frequency that the treble boost starts at. You'll want capacitors smaller than 0.022u, though. Around 1n to 4.7n would work better.

2. It doesn't boost anything, but it sounds like a treble boost with a volume cut. With the control at minimum, the capacitor is out of the circuit and it acts as an attenuator (voltage divider). With the control at maximum, the capacitor bypasses the first resistor, letting high frequencies through intact, but still attenuating the bass. It's not a rolloff, though, it's more like a shelving filter response.

[fuzzy645]

1. Increasing the capacitor's value reduces the frequency that the treble boost starts at. You'll want capacitors smaller than 0.022u, though. Around 1n to 4.7n would work better.

2. It doesn't boost anything, but it sounds like a treble boost with a volume cut. With the control at minimum, the capacitor is out of the circuit and it acts as an attenuator (voltage divider). With the control at maximum, the capacitor bypasses the first resistor, letting high frequencies through intact, but still attenuating the bass. It's not a rolloff, though, it's more like a shelving filter response.

Ahh, that makes sense....thanks.  So its really  more of a treble retention circuit than a treble boost.   So if we use a 100K pot and the pot is fully counterclockwise (on 0) we have the full 100K of resistance (and no cap) and in reality the volume is actually cut at this point.  When we rotate pot clockwise towards 10 we are decreasing the resistance and therefore increasing the volume, while simultaneously attenuating lows and retaining highs above the cutoff point, hence the perceived treble boost. 

Another quick question: When I tried it with a 100K pot I didn't get too much swing. Would a larger pot be recommend, such as 250K or 500K?

Thanks!!

[Mark Hammer]

The arrangement shown is essentially a variable bright switch, and a bright switch is essentially a poor man's version of a "loudness" switch on a stereo.

The brightswitch on an amp provides a cap that straddles the input  and wiper of the volume pot.  It doesn't add any treble content, per se, but rather provides for differential attenuation. 

Normally, a volume pot turns down the level for all frequencies.  How much it does so is a function of how much resistance is on one side of the wiper versus the other.

When there is a cap connecting the input and wiper of a volume pot, it behaves like a zero-ohm resistor for higher frequencies (and larger value resistor for lower frequencies), so that there is always less resistance on one side of the wiper than the other for some frequencies.

Of course, if you've turned down (attenuated) those parts of the spectrum not passing easily through the cap, but the stuff passing through the cap is unaffected, the illusion is created of having more treble, since the bass is turned down more.

In that respect, the diagram would make infinitely more sense if the cap was shown straddling the wiper and left side of R1, rather than the wiper and right side.  It would make more sense, as well, if you thought of R1 + Volume pot as a larger-value pot that could never be set higher than a certain level.  If, say, R1=50k and Vol = 100k, then it behaves like a 150k pot that can never go higher than 2/3 (50k on one side of the wiper, 100k on the other).  If the wiper of R1 was now moved all the way to the right, such that higher frequencies ignored R1 and passed through C1, then for those higher frequencies, the circuit would behave as if there was only C1 in series with the volume pot.

make a little more sense, now?

[fuzzy645]

Mark - I understood what you said (and it makes sense), however the part I'm not fully getting is why you would want the cap on other side of the pot.  I have it hooked up as per the diagram now and as you rotate the pot clockwise (up towards 10) the perceived treble boost kicks in.

My only problem with the sound is the "swing" is not very large and the effect is on the subtle side. Currently I am using a 100K pot for both volume and tone, and I have tried both .022 uf and .047 uf for the cap.   As per the earlier reply, I will try with a much smaller cap value in the .001 uf range.

What you are describing (with the cap on the other side) reminds me a lot of a "treble bleed" circuit in a guitar's volume pot, as is the case in the following diagram. In the treble bleed highs are retained as you lower your volume pot. 

[Mark Hammer]

Good instincts.  It is exactly the same as the treble-bleed bypass cap in a guitar.

As for moving the cap to the other side, it doesn't have to be there.  Works the same on either side.  I just suggested it so that one might look at the R1 resistance on the volume-pot side of the wiper, and the volume pot itself, as a kind of composite volume pot.

The audibility of the "swing" will depend on the ratio of R1 to the volume pot resistance, and the value of the cap.  Keep in mind as well, that since the circuit provides differential attenuation, just like with bright switches on amps and bypass caps on guitars, there is no audible effect when everything is maxed.  We tend to notice the bright switch effect on our amp because we keep the volume setting below max most of the time.  I have an old tweed 5F2-A Princeton, and the "Tone" pot on that is essentially a variable bright cap stradding the volume pot.  Crank the volume up (and a 5W amp does give one that luxury!)m, and the tone control quickly stops doing anything for treble.  Turn the volume down to 3 or 4, and the tone control has a much more noticeable effect.

Same thing with this circuit.

[fuzzy645]

Good instincts.  It is exactly the same as the treble-bleed bypass cap in a guitar.

As for moving the cap to the other side, it doesn't have to be there.  Works the same on either side.  I just suggested it so that one might look at the R1 resistance on the volume-pot side of the wiper, and the volume pot itself, as a kind of composite volume pot.

The audibility of the "swing" will depend on the ratio of R1 to the volume pot resistance, and the value of the cap.  Keep in mind as well, that since the circuit provides differential attenuation, just like with bright switches on amps and bypass caps on guitars, there is no audible effect when everything is maxed.  We tend to notice the bright switch effect on our amp because we keep the volume setting below max most of the time.  I have an old tweed 5F2-A Princeton, and the "Tone" pot on that is essentially a variable bright cap stradding the volume pot.  Crank the volume up (and a 5W amp does give one that luxury!)m, and the tone control quickly stops doing anything for treble.  Turn the volume down to 3 or 4, and the tone control has a much more noticeable effect.

Same thing with this circuit.

Gotcha...thanks.

So I tried it with a 2200 pF and that seemed to work better than the .022 uf or .047 uf cap. However, I suspect since the value of both pots is identical at 100K I'm not getting the most swing. I assume I should reduce the value of the tone pot to either 50K or even 25K while leaving the volume pot alone at 25K to get more swing, is that correct?  Or is it the opposite, and maybe I should increase the value of the tone pot up to 250K??

[Mark Hammer]

Since the goal is differential attenuation, greater effect will be achieved if the tone pot is the higher value, since that will make the attenuation of the non-bypassed part of the spectrum greater.

Consider making the volume pot 50k, and the tone pot 100k.

[fuzzy645]

Since the goal is differential attenuation, greater effect will be achieved if the tone pot is the higher value, since that will make the attenuation of the non-bypassed part of the spectrum greater.

Consider making the volume pot 50k, and the tone pot 100k.

Hey Mark, this little treble booster tone pot is sounding great now!!  I have been experimenting with values ranging from 2200 pf to 8200 pf and several different values sound good.  Question:

Is there a formula one can use to determine the cutoff where frequencies are attenuated?  It seems like with the 8200 pf, the cutoff frequency is lower than when using a 2200 pf cap.  I was hoping I can mathematically sort of pinpoint the magic  cutoff frequency.  Is it 1 / (2 * pi * R* C)?  If so, then with a 50k pot using a 2200 pf cap we get

1 / 2 * pi * 50000 ohms *0.000000002 farads = 1591 hz , which should be a "perceived" boost of highs

changing to a 8200 pf cap we then get

1 / 2 * pi * 50000 ohms *0.000000008 farads = 397 hz, , which should be a "perceived" boost of midrange and highs

...and then getting really crazy and using  a .022 uf cap we get:

1 / 2 * pi * 50000 ohms * 0.000000022 farads = 144 hz., which should be a perceived boost of just about anything we hear on a guitar, no???

I'm sure I'm missing some key detail (I know source impedance and load impedance), but just wondering if I'm barking up the right tree.

Again, the thing sounds really cool (so there is no issue / troubleshooting, etc..),  I'm just purely being academic on this one.