Son of Screamer Question

[Kearns892]

I built the son of screamer as found on Dano's site http://beavisaudio.com/bboard/projects/bbp_BoutiqueTubeScreamer_Rev1_1.pdf. It works fine except for one thing; this thing doesn't get near the saturation as the original tube screamer and not as much gain. I have tried increasing the drive pot to 1 meg, lowering just R4 and R5, all of which increase the gain, but do nothing to get that saturated sound. I'm running a humbucker equipped guitar through a Class A tube amp and have tried high volumes. Is this an inherent trait of this variety of TS or am I missing something?

Thanks   

[snarblinge]

I built one of these as my first pedal, really love it. but I think if you track down a schematic of a TS and compare that to what you just built you will find your answer. its not a full TS just part of it.

[Kearns892]

So you would say the omitted parts make the difference?

[ItZaLLgOOd]

When you say original tubescreamer what are you referring to? (TS 9 or TS808) Are you doing a side by side test?  The only sections missing are the input and output buffers.

I would try to take out D1 or D2, that should give you a little more clipping. Not sure how much though.  Something else might have gone wrong in your build. Double check all of your resistor and cap values.  I have put a 47k in R4 before. Not a whole lot of getty up.

[Kearns892]

I'm not doing a side by side test but I'm not going after some elusive sound I heard one time. From what I've heard of other peoples 9s and 808s and the net I'm not getting the overall distortion and saturation of the originals with the son of screamer. Basically I'm wondering if thats normal or if there is an error. I will check what you suggested though.

[BAARON]

Removing D1 or D2 won't add much more clipping.  It will barely be noticeable.

The parts removed in the Son of Screamer are just the input and output buffers.  They have nothing to do with the gain levels produced by the pedal.  The rest is the same, although this schematic seems to be missing one part: put a 22-47µf cap in parallel with R2, taking care that the negative lead is the one going to ground.  It may or may not make a difference.

If you want to really expand the high gain range, replace the following four parts:

R4 >>> 1k
R5 >>> 1k
C2 >>> 0.22µF
VR1 >>> 1M audio taper

Changing those 4 parts will increase maximum gain about 8.3x-9.6x (depending on the note's frequency), while leaving the minimum gain setting basically unchanged.  (For those who disagree with my math, I suggest calculating your gain by adding the resistance of R4 to the reactance of C2 because they act as one big impedance source rather than strictly as a high-pass filter, as we commonly assume...)

[anchovie]

I'm not doing a side by side test but I'm not going after some elusive sound I heard one time. From what I've heard of other peoples 9s and 808s and the net I'm not getting the overall distortion and saturation of the originals with the son of screamer. Basically I'm wondering if thats normal or if there is an error. I will check what you suggested though.

This could be down to the setups of the other people that you have heard clips from. The TS was designed to push (overdrive) a tube amp and not necessarily be the sole source of distortion. What amp are you running it into?

[sean k]

Excuse me if I'm wrong but my understanding of the first stage is that you don't want to increase the gain  so much as decrease it if you want clipping. The more you increase the gain by increasing the resistance the less effect the clipping diodes have. If you were to make it closer to unity, through the resistance, the more the waveform through the diodes would have precedence on the output waveform.

Therefore more diodes in series equals more clipping and if the feedback resistance is higher then more diodes need to be added in series for the same amount of clipping to occur at high gain. The original schematic then kinda makes sense if it is used with a tube amp so that when the gain is low, and the input tubes of the amp aren't being pushed into the B+ and ground, then the clipping occurs with the diodes which have more effect on the waveform out of the pedal but as the gain goes up the pedal output is cleaner and this then pushes the tube into it's limits.

So you've got to decide where you want your clipping to occur and whether or not gain in the feedback loop is what is needed or more clipping in the pedal. Thats if I'm right though.

[Ben N]

The lack of an input buffer may make a difference if you are using a low-impedance opamp. To check for this, plug your SOS in after a Boss pedal to see if that changes your performance. A simple solution would be to use a jfet-input opamp, like the ubiquitous TL072 (which is, IIRC, what Jack recommended in the first place).

[BAARON]

Excuse me if I'm wrong but my understanding of the first stage is that you don't want to increase the gain  so much as decrease it if you want clipping. The more you increase the gain by increasing the resistance the less effect the clipping diodes have. If you were to make it closer to unity, through the resistance, the more the waveform through the diodes would have precedence on the output waveform.

Therefore more diodes in series equals more clipping and if the feedback resistance is higher then more diodes need to be added in series for the same amount of clipping to occur at high gain. The original schematic then kinda makes sense if it is used with a tube amp so that when the gain is low, and the input tubes of the amp aren't being pushed into the B+ and ground, then the clipping occurs with the diodes which have more effect on the waveform out of the pedal but as the gain goes up the pedal output is cleaner and this then pushes the tube into it's limits.

So you've got to decide where you want your clipping to occur and whether or not gain in the feedback loop is what is needed or more clipping in the pedal. Thats if I'm right though.

I think you're confusing "gain" with "output volume of the pedal," because he wants more distortion happening in the pedal itself, not in a pushed tube amp.
Having done a lot of experiments with the clipping diodes in TS circuits, I can tell you that there's a simple relationship between the number/forward voltage of your clipping diodes:
- higher forward voltate/more diodes = less distortion, but LOUDER (more output)
- lower forward voltate/fewer diodes = more distortion, but QUIETER (less output)

[Kearns892]

Alright, I've been comparing Son of Screamer schematics and it turns out eh one I built doesnt use a biased voltage source just 9 volts, how would that affect the sound?

[anchovie]

Yes it does. The bias voltage is created by the junction of R1 and R2.

[slacker]

The bias voltage needs a big cap to ground from where R1 and R2 join.  
The cap makes that point look like ground to AC so the gain is then set by R4, R5 and the value of VR1. Without the cap you've basically got the parallel resistance of R1 and R1 in series with R4 so the maximum gain will be 1 + (547/54.7) = 11 instead of about 100.

[Brymus]

Also you can make the drive pot as big as 2meg like the SRV808 but then you will be exceeding the voltage rails of the opamp and start getting opamp clipping mixed with the diode clipping.
Also the diodes make a difference too. I like a 1n4001 with 2xLED or 2x1n4148 for asymetrical clipping,some like symetrical clipping from just 2 diodes.

[BAARON]

Also you can make the drive pot as big as 2meg like the SRV808 but then you will be exceeding the voltage rails of the opamp and start getting opamp clipping mixed with the diode clipping.

No, you won't.  You won't start getting voltage rail clipping from the op-amp until the voltage swing of your guitar's signal + the total combined forward voltage of all the clipping diodes reaches the supply rails.  A bigger gain pot won't magically push the signal beyond what the clipping diodes will allow.

[Kearns892]

The bias voltage needs a big cap to ground from where R1 and R2 join.  
The cap makes that point look like ground to AC so the gain is then set by R4, R5 and the value of VR1. Without the cap you've basically got the parallel resistance of R1 and R1 in series with R4 so the maximum gain will be 1 + (547/54.7) = 11 instead of about 100.

So like 10uf?

[BAARON]

Yes. 10µF would be big enough.

[Kearns892]

Hmm when I put the 10 uF cap in the drive didn't change I just got a squealing noise over the signal.

[BAARON]

Are you sure it's in the right spot?  It needs to be in parallel with R2, with the negative lead going to ground.

[Hupla]

Are you sure it's in the right spot?  It needs to be in parallel with R2, with the negative lead going to ground.

I just wanted to let you know that this does work. THANK YOU. It now sounds like every tubescreamer I ever heard.