Tube Screamer-Warm clipping How to ?

[HOTTUBES]

I'm seeking the forum's help here on how to get the most  " Tube like " or " warm " clipping etc ??
I'm not after Gain , or increased Volume ...





Thanks for any Advice !!

[mistahead]

Clipping - in that circuit the diodes are in the opamp feedback network aren't they?

There are diode mods available which replace the stock with dpdt'd options including 2x1 Red led (asymetrical, softer) and 1 red with 1 Si for similar, but (and someone will correct me if I'm wrong) the opamp clipping is a key part of the circuit and sound too - hence why folks spend so long testing the varied 4558's (I think are dual opamps right?) out there...

Might be worth doing a once over of geofex TS tech page - he had some thoughts if I call.

[SmoothAction]

Tubescreamers already utilize soft clipping, it's kinda the whole point of it all.. It sounds like you're trying to change the clipping characteristics of the screamer. What diodes are you using in the feedback path?

[HOTTUBES]

This is a setup that sounds great imo , but when i turn up the gain pot past half , it distorts in a bad way etc etc ...

D1 - 1n4001 + 1n34a
D2 - 1n4001
D3 - 1n4001 + 1n34a


If i could keep this diode config , but kill the distortion , it would be great imo .
It's very nice until about half way on the gain pot , then it all goes for shit and stars to really completely distort !

I'm using a stacked set of JRC 4558D's ...

[mistahead]

DPDT with a set of "softer" diodes is about all I can suggest - Ge or LEDs might give you much less of the abrasive distortion you are getting.

Stacking opamps would increase the gain/boost/signal voltage?

Gah - what's the volume control method on the TS? Voltage divider on the inside edge of the output buffer, tap a line from the otherside of the output buffer to confirm you're not overdriving it perhaps...

[psychedelicfish]

You could try a 6AL5 twin diode tube, check out merlin's spark gap OD thread for info on how to do it.

[defaced]

This is a setup that sounds great imo , but when i turn up the gain pot past half , it distorts in a bad way etc etc ...
...

It's very nice until about half way on the gain pot , then it all goes for sh*t and stars to really completely distort !

I'm using a stacked set of JRC 4558D's ...

Judging by your description and the text from RG's explanation of the clipping stage, it sounds like you might benefit from more series diodes to increase the clipping threshold. 
http://www.geofex.com/Article_Folders/TStech/tsxtech.htm

[pappasmurfsharem]

1n34a had a very low threshold so it clips more easily.

Try two red leds.

[teemuk]

Build something else than a Tube Screamer.  


No. I'm dead serious.

The clipping of that design is already extremely soft, in fact, way softer than clipping of many real tube-based circuits.

However, if the tone doesn't match your individual and totally subjective idea of what is "warmth" and "tubelikeness" then you probably need to abandon the design completely because there are no minor mods that could ever change that. You probably want to change A LOT more than just clipping characteristics.

[george]

As others have pointed out, RG has the answer:

[samhay]

This is a setup that sounds great imo , but when i turn up the gain pot past half , it distorts in a bad way etc etc ...

How about replacing the gain pot with a smaller value pot?

[ashcat_lt]

Tubes get hot when they run, but there isn't anything particularly warm about their sound when distorting.  A bare 12AX7 pushed to overdrive can get just as harsh and spitty and nasty as any opamp.  The "warmth" comes from tone shaping before and (mostly) after the tubes. 

A TS is generally pretty nasal.  If you're looking to warm it up, maybe look at raising the value of either or both of the caps in the feedback loop, or mess with the tone control values some.

[merlinb]

Where do people get the idea that the TS is a soft clipper?? Diodes in the feedback loop make the quintessential hard clipper. Ge diodes are a 'bit' softer, but we're still very much in solid-state territory. LEDs or multiple diodes are no better than the stock silicon diodes in terms of hardness, but they do allow you to get more output volume before the clipping happens. There will also be more low-pass filtering since the gain pot will be set to a greater resistance before clipping occurs, which may increase the impression of warmth, unless the feedback cap is also changed.

The classic way to make this into a soft clipper is to put a resistor in series with the feedback diodes (or even a different resistor for each diode). If you make it variable then you have some control over the softness. Something between about 10k and 100k ought to be what you're looking for.

[pappasmurfsharem]

Where do people get the idea that the TS is a soft clipper?? Diodes in the feedback loop make the quintessential hard clipper. Ge diodes are a 'bit' softer, but we're still very much in solid-state territory. LEDs or multiple diodes are no better than the stock silicon diodes in terms of hardness, but they do allow you to get more output volume before the clipping happens. There will also be more low-pass filtering since the gain pot will be set to a greater resistance before clipping occurs, which may increase the impression of warmth, unless the feedback cap is also changed.

The classic way to make this into a soft clipper is to put a resistor in series with the feedback diodes (or even a different resistor for each diode). If you make it variable then you have some control over the softness. Something between about 10k and 100k ought to be what you're looking for.

Just curious here. How would a resistor for each diode make a difference? since it's on the same track wouldn't they just become parallel and be that same as a single resistance?

Crap you said in SERIES. I need more coffee.

So what is the logic there, it send an attenuated signal to the diodes so it gets clipped less? Similar to how playing softer clips less?

[Mark Hammer]

Me, I don't even know what people mean by "warm clipping".  That doesn't mean what they seek is unattainable via solid-state means or that it is unreasonable.  I just don't know what it is they mean by that descriptor.

[FiveseveN]

I just don't know what it is they mean by that descriptor.

If I had to bet, I'd say it has nothing to do with distortion per se and everything to do with pre- and post-distortion filtering. But that's old news.

[Paul Marossy]

Tubes get hot when they run, but there isn't anything particularly warm about their sound when distorting.  A bare 12AX7 pushed to overdrive can get just as harsh and spitty and nasty as any opamp. 

+1

[Digital Larry]

Where do people get the idea that the TS is a soft clipper?? Diodes in the feedback loop make the quintessential hard clipper. Ge diodes are a 'bit' softer, but we're still very much in solid-state territory. LEDs or multiple diodes are no better than the stock silicon diodes in terms of hardness, but they do allow you to get more output volume before the clipping happens. There will also be more low-pass filtering since the gain pot will be set to a greater resistance before clipping occurs, which may increase the impression of warmth, unless the feedback cap is also changed.

The classic way to make this into a soft clipper is to put a resistor in series with the feedback diodes (or even a different resistor for each diode). If you make it variable then you have some control over the softness. Something between about 10k and 100k ought to be what you're looking for.

Check the TS schematic.  It is a non inverting stage.  The diodes there do not do clipping like they would in a non inverting stage.  They are part of a "piecewise-linear" feedback switching network.  When the output voltage minus the (-) input voltage exceeds the diode forward voltage, the stage turns into a voltage follower with Av = 1, not approaching zero.  If you put a resistor in series with the diode pair you could get higher gains after the diode starts conducting.  Also the corner between the two gain zones would be softened.

When you put a resistor in series with diodes that are being driven from an op-amp output to ground, you are softening the clipping corner because the current is reduced through the diodes and you don't go around the curvy corner of the I/V relationship quite so fast.

[teemuk]

Where do people get the idea that the TS is a soft clipper??

Oscilloscope?

[gritz]

I think that calling it simple soft clipping is a touch misleading. At high input signal levels a sine wave may look pleasingly rounded, but remember that the opamp's gain snaps from over x100 (with the drive control maxed) to x1 over a very small range of output. In a recent thread on the subject someone (I can't remember who, sorry) likened it to crossover distortion and yeah, it's a pair of very sharp kinks in the transfer function.