Son of Screamer vs Tube Screamer RC Filter in clipping section

[cdwillis]

Why are these different? Wouldn't one create a low pass filter and the other create a high pass filter?




I've never been a big fan of Tube Screamers. The way they suck the low end out of your signal and honked up the mids is good for lead guitar, but as a base tone it's very blah and anemic in my opinion. So I've been looking at the electrosmash analysis of the circuit and how I could modify it for more low end without getting farty. My idea was to swap the 4k7 resistor for a 2k2 (increase the gain a bit), then use a .1uf cap and a .68uf cap on a toggle to switch the knee from 723.8hz (close to stock TS808) to 106.4hz.

The .047 and 470k on the input of the op amp should let enough low end through. There's a 1k/.22uf low pass filter after the clipping section that's cutting high end above 723.8hz.

If this is all over the place, I apologize. I'm still very new and learning a lot about this stuff.

[GGBB]

Why are these different? Wouldn't one create a low pass filter and the other create a high pass filter?

The only notable difference is the low pass on the output in the son of screamer.

[cdwillis]

That's in the tube screamer circuit too, it's just included in the tone section on the Electrosmash divisions. Look at R4 and C3 on the TS section. It doesn't make a difference that it's reversed from that on the SoS?

[MikeA]

Regarding fattening the bass, there are some good insights at http://www.muzique.com/lab/fatt.htm .  Also the Aion Electronics Stratus PCB build doc (1590B version) shows mods for fat bass as well as flattening the mids.

[Steben]

Swapping the cap and resistor is a common mod on TS style soft clipping circuits.
parallel caps with a pot as well as a "bass control" as in the Timmy for example.

It's all taste. In the days of yore legends played with rangemaster as always-on pedal with even more low end cut.
Bass does not cut through the mix.
[harsh mode] They used a bass player [/mode]  

One has to understand most basic amps were at that time quite low endy, with the exception of treble channels.

[idy]

Swapping the order of the R and C has no effect. The two configurations are the same, you are tripping on an insignificant and arbitrary detail. Both have a path for frequencies above a certain point to be removed from the inverting input, forcing the op-amp to boost them.

They both have a high cut also, the tiny cap in the feedback loop, which allows highs above a certain freq. a clear path to the inverting input, cutting them.

[amptramp]

You have to look at what you are trying to do.  You don't want to clip high frequencies because that only creates higher harmonics which would make the bass guitar cut into lead guitar territory.  You don't want hard clipping low frequency signals because the input signal carries harmonic information that disappears when clipping occurs - this makes the signal muddy.  Imagine a low-frequency fundamental with harmonics riding on top of it.  Once the clipping occurs, the fundamental is prevented from getting any louder but all of the harmonics are removed during the time the signal flatlines on either high or low levels.  This makes the signal muddy.  Bad enough with a lead guitar but even worse with a bass guitar because the mud lasts longer due to the longer time constant.

I would at least double the capacitances of a Tube Screamer to get the bass frequencies through and probably go beyond that because you may have other pedals in series cutting low frequencies and I am all in favour of keeping the full frequency range until you need to apply a lower frequency limit.  If you have three pedals that are down 6 db at the 41 Hz minimum frequency, your signal is down 18 db at 41 Hz and that is bad news.

The best way to allow the signal to be clipped without creating too much mud is to put a resistor in series with the feedback diodes.  Indeed, the Son of Screamer uses two diodes in one polarity and one in the other and this looks like it will allow more signal in the two-diode direction except the two series diodes are germanium and the diode for the opposite polarity is silicon.  You have a lower threshold voltage for the two germanium diodes but a higher series resistance.  The result of this is that a lot of even harmonics will be created whereas the TS mainly generates odd harmonics.

The Rat also has a method of avoiding mud at low frequencies: the diode feedback is capacitively coupled so the clipping threshold gets higher at low frequencies.  This could work here with a capacitor in series with the back-to-back diodes.  If you had a separate cap for each diode, the cap would charge up and the bias would increase until the diode stopped conducting, meaning leakage current would dominate the response and that never works well.

One thing to remember is that both the TS and SoS have unity gain for the input signal even when driven hard into clipping and this is how they avoid that hard clipping sound that sounds like the speaker is being trapped from going beyond a certain amount in both directions.  If you ever hear a speaker hitting the enclosure, the sound is harsh and not something you want to emulate in electronics.

[Steben]

The Rat also has a method of avoiding mud at low frequencies: the diode feedback is capacitively coupled so the clipping threshold gets higher at low frequencies.  This could work here with a capacitor in series with the back-to-back diodes.  If you had a separate cap for each diode, the cap would charge up and the bias would increase until the diode stopped conducting, meaning leakage current would dominate the response and that never works well.

?
Where is that cap?

[willienillie]

For the record, R4 connects to Vref in the TS, not ground.

[ElectricDruid]

For the record, R4 connects to Vref in the TS, not ground.

Fair enough for getting facts straight, but since C3 blocks DC it doesn't really make any odds.

[amptramp]

Sorry, I got it mixed up - the pedal that uses capacitively coupled back-to-back diodes is not the Rat, it's the Electro-Harmonix Big Muff Pi in all its variations.

[Steben]

Sorry, I got it mixed up - the pedal that uses capacitively coupled back-to-back diodes is not the Rat, it's the Electro-Harmonix Big Muff Pi in all its variations.

true, it does need that cap already for DC blocking however

[amptramp]

Yes, the cap is needed for DC blocking but it also serves to reduce clipping at low frequencies.