Something for the breadboad - a hard or soft clipping circuit snippet

[slacker]

Here's a simple little circuit I've been playing about with recently, not really a finished pedal although it sounds alright as drawn.



With the wiper of the 100k pot in the middle you've got basically a clean sound and as you turn it towards the outsides you get two different sorts of clipping.
Turning the wiper towards the top gives you a TS style non inverting stage with "soft" clipping.
Turn it the other way so the wiper moves towards the bottom and it gives you an inverted "hard" clipping stage.

The 47n cap sets the bass rolloff frequency in both "modes" and the diodes can be tweaked to taste, all the usual tricks LED, mosfets, asymmetrical or whatever should work.

There's a couple of things wrong with it at the moment. With the wiper of the 100k in the middle there's a volume drop and there's a bit of a dead zone either side where not much happens. Making the the first stage into a booster might help with that.
The 680k resistor which sets the maximum gain could probably do with being bigger, I got my gain calculations wrong but couldn't be bothered changing the schematic, 1M - 2M2 would probably be better.

Enjoy

[petemoore]

  INteresting configuration.
  Pot to one side feeds more into the + of the second opamp, turned the other way then the - input of second oA gets more signal input.
  With the wiper of the 100k in the middle there's a volume drop and there's a bit of a dead zone either side where not much happens.
  Toward the middle settings, it looks like both the +/- inputs see the same phase of input [at similar or same amplitudes], and cancellation results.
  Is the second opamp's +input bias intended to be '10k / DC' coupled to the first opamps output? 

[earthtonesaudio]

Cool.  I dig it.

[slacker]

  Toward the middle settings, it looks like both the +/- inputs see the same phase of input [at similar or same amplitudes], and cancellation results.

Yeah, that's what's happening, the gains for the inverting and non inverting sides aren't exactly the same though so you never get complete cancellation. I'm hoping that giving the first stage a gain of 10 or so will reduce the effect or even make the middle settings into a usable booster, I haven't had time to try it yet though.

  Is the second opamp's +input bias intended to be '10k / DC' coupled to the first opamps output? 

Yeah the bias for the second stage comes from the output of the first stage.

Cool.  I dig it.

Thanks

[snap]

you could invert the signal on the lowermost leg before the pot.

[earthtonesaudio]

How about a single transistor phase splitter as input, send the emitter to the noninverting input and the collector to the inverting input, something like that?  Then capacitively couple and bias the op-amp separately.

[George Giblet]

Yeah, good one slacker.

[Ben N]

How about a single transistor phase splitter as input, send the emitter to the noninverting input and the collector to the inverting input, something like that?  Then capacitively couple and bias the op-amp separately.

You beat me to it. Added benefit: The output remains in the same phase relationship with the input, in case you ever use the thing in parallel with another effect or incorporate Sparkledrive-type clean mixing. (Hey, I think we just freed up an opamp--looks like a mixer to me.)

Neat idea, Ian.

[sean k]

What if you used a dual gang pot and the other 100k was in the feedback loop of IC1 and it had 5k resistors from the outer lugs to the centre lug. That means you'd have more resistance at the centre setting, 10k, but only 5k, in parallel with 100k which wouldn't bring it up much, at each end. That way you have a semi log increase in volume from each end to the centre... maybe?

[sean k]

Like this?

[Ben N]

Anything new on this front?

[slacker]

I haven't had chance to play about with it much.

I've changed the input buffer into a booster (220k from pin1 to pin2 and a 47k and 47n cap to ground from pin2) and that's taken care of the volume drop in the middle of the pot and makes a reasonable boost. This has also pushed the maximum gain up to about 250 - 300, which is similar to a Distortion Plus or a "hot" Tube Screamer.
If I get chance sometime next week I'll do a quick soundclip.

I don't think inverting the signal into one side will work because it's basically a differential amplifier so you just get maximum gain all the time no matter where the pot is set. That's what happens in the simulator anyway, I haven't actually tried it for real yet.
Using a transistor phase splitter and then decoupling it like earthtonesaudio suggested gets some interesting results though, not what I wanted but it might be worth investigating as a separate idea. Basically depending on where you put the "hard/soft" pot as you turn it towards pin5 the decoupling cap from the emitter starts to see a low resistance path to ground and starts acting like an emitter bypass cap which increases the gain on the collector so you get a hotter signal going into the inverting "hard" side of the opamp.

Sean's idea looks interesting, if I can find a dual pot I might try it out.

[slacker]

Here's a little soundclip of the circuit, this is with the first stage modded as explained in the last post.
The clip starts with the clean amp then with some low gain on the "soft" side, then maximum gain on the "soft" side and finally maximum "hard" gain.

http://www.aronnelson.com/gallery/main.php/v/slackers-stuff/album170/hardsoft.mp3.html

[John Lyons]

The clips sound nice.
So the gain changes in the clip all came from the 100K pot right?
I was a little confused by what you said about min and max gain on
the hard/soft sides since there is only one pot.
Thanks for the clip.

john

[Ben N]

Sounds good, Ian--this could really add to the functionality of the various opamp + clipper mainstays.

[slacker]

Cheers for the comments chaps.

So the gain changes in the clip all came from the 100K pot right?

Yeah John the 100k pot controls gain and clipping type. Basically "soft" is from 12 o'clock anti clockwise to 7 o'clock and "hard" is from 12 o'clock clockwise to 5 o'clock. The gain of whatever type increases as you move away from 12.
In the clip the low gain "soft" is about 9 - 10 o'clock, then the pot's at 7 o'clock and finally 5 o'clock.
About 11 - 12 o'clock you get a cleanish boost. There's a bit of dead spot between 12 - 1 o'clock where you get very low gain distortion that's not very interesting, but apart from that all the settings are usable.

[WGTP]

http://www.aronnelson.com/gallery/main.php/v/WGTP/OPtion.jpg.html?g2_imageViewsIndex=1

Interesting approach.  I tried something similar that avoided the phase reversal problem by "panning" between a TS (diodes in FBL) setup and a Dist+ (diodes to ground) approach.  It was based on John Hollis diode switching idea, using a pot rather than a switch. 

The Boss OS-2 deals with it by panning between 2 different distortions.  Cool stuff.   

[earthtonesaudio]

In a way it one-ups the OS-2 by doing both things with one set of diodes. 

Sounds good!

[John Lyons]

Thanks for the rundown Ian, I see what is going on now with the pot settings.
Very neat!


john

[earthtonesaudio]

I think this circuit would be a good candidate for a huge tone shaping section.  Like a rotary switch for different diode combinations plus a 4-band parametric EQ.  ...15 knobs.