How to refine the Son of Screamer ?

[HOTTUBES]

I have built another one of these Overdrives and find it to be kinda muddy & loose sounding
with little to no top end . I was kinda hoping it would have more high end sparkle , and a
tighter low end ....and more drive would be great as well .

Do the Overdrive guru's have any tips on how to get this circuit to Kick Ass !!

My build is totally as the Vero shows component wise , the Opamp is a JRC4558D , and the
clipper's are 4148's ....


Thanks for any help !!!!

[rockhorst]

All the standard Tubescreamer tricks still apply to the SoS circuit, so just search for that. Heaps of info. Some famous SoS-type pedals for inspiration would be Zendrive, Timmy, Eternity, Barber Silver and most other overdrives out there today. Have a look at those schematics for inspiration.

[petey twofinger]

i spent an evening with a similar 4558 based od , tried quite a few things , went with a combo for asymmetrical clipping , red led's / 1n34a . lost a little unec distortion using the led , but it was just the trick .

[HOTTUBES]

In this layout , what do the #4  & #5 caps do in the circuit ?
Does the #4 resistor somewhat control the gain in this circuit ?

[earthtonesaudio]

Let's see, input and output buffers, momentary switching, low battery indicator... oh crap what have I done!?

[Kesh]

In this layout , what do the #4  & #5 caps do in the circuit ?
Does the #4 resistor somewhat control the gain in this circuit ?

c4 forms a low pass rc circuit with r6. It rolls off frequencies above about 700hz. You could maybe swap it for a 100n for more upper mids and top end. c5 controls the tone control's 'pivot' frequency. About 3khz. R4 controls gain with the gain pot as part of a standard op amp non inverting amp (if you ignore the diodes) but as it's in series with a cap, changing it will also change tone.

[Mark Hammer]

High end sparkle is partly a product of lowering drive within the TS platform.



The use of the feedback loop for settng drive means that as gain is increased (51k plus a larger share of the pot's 500k), the top-end rolloff (dictated by the 51pf cap) moves in a downward direction, from around 61khz at min gain to 5.6khz at max gain.

That's not the end of it.  Because the higher drive generates a lot of undesirable harmonic content AND increases risk of hiss, AND because the architecture/strategy of the TS is to apply more gain to the mids and top than to the bottom (in order to achieve roughly equivalent clipping across the fretboard), the circuit includes additional lowpass filtering.

So, if you wanted more sparkle, you'd likey want to do a couple things:

1) Aim for lower overall gain.

2) Swap the diode pair for a diode quartet, Timmy-style, to produce a higher clippig threshold and less clipping.

3) Ease back on the compensatory lowpass filtering by reducing the .22uf cap in the 1k/.22 filter pair to something like .1uf.

4) Change the tone control from its current form (complementary treble boost/cut) to something more like a gentler rolloff Zendrive-style.

[HOTTUBES]

High end sparkle is partly a product of lowering drive within the TS platform.



The use of the feedback loop for settng drive means that as gain is increased (51k plus a larger share of the pot's 500k), the top-end rolloff (dictated by the 51pf cap) moves in a downward direction, from around 61khz at min gain to 5.6khz at max gain.

That's not the end of it.  Because the higher drive generates a lot of undesirable harmonic content AND increases risk of hiss, AND because the architecture/strategy of the TS is to apply more gain to the mids and top than to the bottom (in order to achieve roughly equivalent clipping across the fretboard), the circuit includes additional lowpass filtering.

So, if you wanted more sparkle, you'd likey want to do a couple things:

1) Aim for lower overall gain.

2) Swap the diode pair for a diode quartet, Timmy-style, to produce a higher clippig threshold and less clipping.

3) Ease back on the compensatory lowpass filtering by reducing the .22uf cap in the 1k/.22 filter pair to something like .1uf.

4) Change the tone control from its current form (complementary treble boost/cut) to something more like a gentler rolloff Zendrive-style.

Thanks Mark for the great tips .....

When you say a Diode Quartet , do you me two(2) pairs in series for clipping ?

Example :  1n34a + 1n4001                   LED + 4148
                1n34a + 1n4001        or        LED  + 4148 

[Mark Hammer]

...or two 4148s in parallel with two more 4148s.  Basically, just something to raise the forward voltage required a little more.  You'll end up with more output level as a byproduct of that too, which ain't such a bad thing.

A pair of red LEDs might not be so bad, but I wouldn't add any other diodes in series with them, since you'd bring the clipping threshold up so high, you'd get sporadic clipping.

[HOTTUBES]

Ok , here's what i've done so far , and were heading in the right direction !!

I have changed things to these values ...

C1 to .033uf to try and pull back on the low end .( was this the best place to do that )
C3  i had to remove this cap to get my sockets in place , so no C3 cap ! ( At the moment )
C4 to  .1 uf

Clipping Diodes are LED + 1n4001 for both .





I must say it's sounding alot better , but i still would like more high freq.

                     

[HOTTUBES]

Yep , sounding badass so far !!
just need to figure out how to squeeze some more top end out of her !!


Thanks for everyone's help so far ....

[HOTTUBES]

Why does the VOL pot effect the drive so much in this circuit ?

I get very little Grunt out of this pedal until about 75% on the VOL , then i really get to hear the drive come alive ...

[ashcat_lt]

What are you running it into?

With the diodes you've got in there you're getting 4V or more output, and probably not a whole lot of distortion from the pedal itself.  I'd guess you're hearing clipping from downstream of this pedal when you turn up the volume pot.

[Kesh]

What are you running it into?

With the diodes you've got in there you're getting 4V or more output, and probably not a whole lot of distortion from the pedal itself.  I'd guess you're hearing clipping from downstream of this pedal when you turn up the volume pot.

Agreed.

The clipping wont be from the diodes, unless you used IR diodes or something, and then still not much. You've effectively turned it into a booster, which is fine for overdriving an amp.

[HOTTUBES]

It funny you mention that , i can remove the diodes completely , and still have a fair amount of gain on tap .
Why does this circuit still work with out the diodes in place ?

[Kesh]

because it is over-driving whatever comes after it.

the diodes reduce gain but add distortion by clipping off the signal. without diodes it will give about x100 gain, that's 40dB, less what all the tone elements take out, so about 35dB for middle frequencies, tailing off to much less for high and low frequencies.

you may also be clipping the circuit's op amp if your guitar is hot enough, and/or you have gain set to max.

[Mark Hammer]

The diodes simply determine at what signal amplitude clipping will occur and how much gain is required to produce it.

It is the case that a great many circuits will produce distortion even without any diodes that are intended to deliberately produce distortion.  You need to keep in mind that many op-amps will not be able to swing to more than a half to one volt within the "rails".  So, a chip that can swing up to a volt within the rails, and powered by 9vdc can produce an amplified signal that replicates the input signal up to +/-3.5V AC.  The input signal is often 100mv peak-to-peak.  Multiply that by 35x and you have +/-3.5v.  The Tube Screamer provides for a minimum gain of just under 12x and up to 118x.  So, well before you reach maximum gain on the knob, you will have exceeded the chip's ability to cleanly replicate the input signal.  Many folks here have noted that they can stick the clipping diodes in a drawer across the room, and their 741-based Distortion+ can't deliver a clean tone, no how, no way, at ANY gain setting.  And some units, like the Black Cat distortion, rely exclusively on pummeling a dual op-amp without nary a diode in sight.

In many respects, a lot of chip-based distortions are "double clippers".  We might normally think of circuits like the Big Muff as being such double clippers, given that they have two explicit clipping stages.  But in many respects, the way that op-amps are used often exceeds their capacity to cleanly reproduce the input.  So we end up clipping a signal that is already clipped.  That may well be part (though not all) of why people can sometimes find that swapping one dual op-amp for another delivers a more pleasing tone to their ears.

[Kesh]

on a slightly different tack. one thing i don't understand about the ts is why the emitter followers have their bases biased from the 4.5V line.

i guess it doesn't matter on the original circuit much, but with diode mods it means less headroom and the output buffer may also clip when diodes' clipping thresholds and gain are up. and ugly clipping that will be.

...
So, well before you reach maximum gain on the knob, you will have exceeded the chip's ability to cleanly replicate the input signal.
...

IIRC it's a log pot, so only the very last part should clip

[HOTTUBES]

I think i have been a bit fooled during my play tests with this vero build .
From the start i have been thinking it's got far too much BASS cause something sounds like its been distorting
alot like when the signal has too much low end etc , but in this case maybe its a problem with my Diode set up
as you have pointed out ?

I have socketed the C1 cap , and have changed this cap down to a .01 uf thinking it was alot of low end creating
this distorted sound , but even at the .01 uf value its still present .


It seems i have always liked the Boss SD1 , so i'm shooting for that sorta sound , but i just not experienced enough
with this circuit yet to know how to get it in that ball park , In the SD1 circuit if you want more low end we swap out the C3 cap , or if you want to change the mids play with the Value of C4 etc etc .... but with this Vero i'm kinda lost as too what cap or resistor does what ...


Thanks for the Help Guys ....
it sucks being a Noob !

[R.G.]

on a slightly different tack. one thing i don't understand about the ts is why the emitter followers have their bases biased from the 4.5V line.
i guess it doesn't matter on the original circuit much, but with diode mods it means less headroom and the output buffer may also clip when diodes' clipping thresholds and gain are up. and ugly clipping that will be.

It's because the original TS was not designed to be modded. It was designed to be as inexpensive as possible to build while meeting its performance objectives with the stock/no mods circuit. I suspect the original designers had no clue that cloning would happen or that it could possibly be as pervasive as it is today. If they did, I want to hire them as prophets.