Pros and Cons of OD circuits

[pdf64]

You're right that the cleanish sound you get with a TS when rolling back the guitar volume is rather muffled (part of this is due to the cable capacitance anyway).

That’s not it. High output pickups are needed to hear the effect I’m referring to; or a clean boost with low output pickups.

I like to bias my amps for almost straight slopes with a triangular signal at full output, so no CD.

But with overdrive, won’t bias shift and HT sag (and cathode bias ‘squish’ if appropriate) induce CD to manifest? Perhaps at the A to B transition, rather than zero crossing class B.

[Helmholtz]

But with overdrive, won’t bias shift and HT sag (and cathode bias ‘squish’ if appropriate) induce CD to manifest? Perhaps at the A to B transition, rather than zero crossing class B.

Probably, but doesn't mean that's something I'm after .
I never said that I like heavy power amp distortion.

[uneumann]

And please explain the picture shown.

The picture above is the transfer function for positive input. It's symmetric so negative inputs behave the same way. The lower line is the input and the upper bent curve is the output. You can see that a linear input produces the curved output. No matter... here is the response to a triangle input. The curved purple line is the output. You can see that the middle area (zero crossing) is stretched vertically. The triangle peaks are reproduced at the extreme pos/neg inputs, but the zero crossing area is distorted.

This is much harder to see in a sine wave - thanks for the suggestion to use a triangle.

As for freq or temporal effects - there are none. The small cap C21 can be ignored. It's simply a circuit with a distorted transfer function - but then so is a clipper.

[Helmholtz]

The picture above is the transfer function ...

Yes, that's the reason why I like to use a triangular signal. As long as frequency dependence can be ignored, the slopes represent the transfer function.

The picture above is the transfer function for positive input.

Not clear what you mean with positive input, as the picture shows ground-symmetrical AC signals (no offset).


Regarding the outpur signal, I see 3 different slopes, corresponding to 3 different gain levels::
- Starting with the steepest one determined by R27,
- slope decreases when the left side diodes start to conduct,
- even shallower slope when the right side diodes start to conduct and shunt the 30k resistors.

All in all the output signal doesn't resemble crossover distortion, where the transfer slope is shallower at or above zero crossings, then getting steeper again.

[uneumann]

Not clear what you mean with positive input, as the picture shows ground-symmetrical AC signals (no offset).

You asked about the first picture - that's what I meant by "picture above". That one is just the transfer function. It is unipolar - the input only ramps positive.
The second picture has the triangle input around zero - as requested.

Regarding the outpur signal, I see 3 different slopes, corresponding to 3 different gain levels::
- Starting with the steepest one determined by R27,
- slope decreases when the left side diodes start to conduct,
- even shallower slope when the right side diodes start to conduct and shunt the 30k resistors.

Exactly - that's what the transfer function shows. The gain decreases non-linearly as the amplitude increases.

All in all the output signal doesn't resemble crossover distortion, where the transfer slope is shallower at or above zero crossings, then getting steeper again.

Ahhh, but it is crossover distortion. It distorts around the zero crossing. It happens to be created differently and look differently than the push-pull CD you're familiar with.
The ear doesn't how it's made and harmonics are generated no matter how it's made. The fact that it happens at or around the zero crossing makes it crossover distortion. It's certainly not a clipper.
Even if you don't agree with calling it CD - you can see how this behaves differently than a clipper - which is really my main point. Call it whatever you wish, but let's agree on that. That's why I included it in the original post with the summary of circuits.

It seems we're getting deeper into this now so let me also talk about what this circuit sounds like. I happen to be playing with the circuit recently.
There are three main audible differences. One is that those little high freq wiggles on lower freq notes I was talking about earlier survive this circuit very well. In fact, not only do they get through the circuit, they manage to contribute to the OD tone because they also exist in the zero crossings. This difference is quite audible when compared to say a DCCF. Because they get through, you can push much more bass through this OD circuit without losing those string partials (harmonics). You don't have to cut the bass to get decent tone. Clearly, you still can cut bass - but you don't have to - and you keep note definition.

Another difference is that there is no compression at high levels. You never get to that point where turning up the guitar volume or amp drive doesn't make it louder. This is of course good and bad. If you want compression you need to add a clipper/compressor somewhere in the path - but you have that choice and you can control the degree of compression outside of this circuit.

Lastly, as a note decays in a clipper, it cleans up as the clipping stops. While that may be good for some, others want the note to stay overdriven - to sustain in OD. In order to do that with a clipper, you need lots of gain to keep the signal large as the guitar signal decays. Hi-gain is needed to get OD sustain with clippers. This circuit lets the note decay and keeps the OD distortion going as long as you're in the non-linear area. In fact the % distortion goes up as the note decays until the amplitude drops below the nonlinear region. The effect of this is that it sounds like there is much higher drive and gain than there is. This is good for keeping noise at bay. You don't need as much gain to get good sustained OD.

[Helmholtz]

That’s not it. High output pickups are needed to hear the effect I’m referring to; or a clean boost with low output pickups.

I only use vintage or vintage type PUs, so I can't comment here.

I just can't see, how a 10% or lower "clean signal content" could make a significant difference.

[Helmholtz]

Ahhh, but it is crossover distortion. It distorts around the zero crossing.

Well, I'd say the circuit is linear around zero crossing up until the diodes start to conduct.
You should get a perfect triangle with output low enough.

Completely different from typical CD where gain is lower around zero crossing and increases above.

Another difference is that there is no compression at high levels.

Then it's not for me as I want the compression/sustain and feedback.

And cleaning up at lower levels is essential to me, as it allows to vary the sound via the attack. Light picking: quasi clean, hard picking: distorted.

[uneumann]

Well, I'd say the circuit is linear around zero crossing up until the diodes start to conduct.
You should get a perfect triangle with output low enough.

Completely different from typical CD where gain is lower around zero crossing and increases above.
Then it's not for me as I want the compression/sustain and feedback.
And cleaning up at lower levels is essential to me, as it allows to vary the sound via the attack. Light picking: quasi clean, hard picking: distorted.

We can disagree about the term... but of course it cleans up at low levels. Think about it. You yourself say (above) that at very low levels there is no distortion.
The curves are smooth, so the transitions from clean to distortion and back again are also smooth. Sustain and feedback are still there. I never said they were not. How did you infer that?

Also, the point of my explanations and comments are to simply explain something. I'm answering your questions as clearly as I can.
These are technical points - not audiophile opinions. I'm not saying your have to agree or love anything.
However, until you've tried it, I don't really see how you can formulate a strong subjective opinion about how it sounds.

[Helmholtz]

We can disagree about the term... but of course it cleans up at low levels. Think about it. You yourself say (above) that at very low levels there is no distortion.

If we agree that the response is linear at low levels, you can't have (crossover) distortion at low levels.
I understand that "crossover" typically refers to zero crossing.

Sustain and feedback are still there. I never said they were not. How did you infer that?

My comment above (obviously) referred to your statement that "there is no compression at high levels".
If so, I doubt the circuit would increase sustain like other distortion circuits with progressive compression.
Increased sustain is a consequence of compression.
Your circuit certainly has some overall compression, so maybe my evaluation here is premature.

And I definitely want the signal to clean up at decay or with softer picking.
But that's only me .

[Mick Bailey]

The way to experiment with crossover distortion is to add a pair of back-to-back diodes in series with the signal path. These can have a bypass pot to vary the depth of the effect (similar to Peavey's diode noise reduction circuit). The Boss HM-2 also uses diodes in the same way. You can experiment with diode types to give different levels of crossover. I had a distortion circuit breadboarded with a variable crossover feature (as well as varying asymmetric clipping). It was interesting and as spin off got rid of hiss and noise.

[teemuk]

My comment above (obviously) referred to your statement that "there is no compression at high levels".
If so, I doubt the circuit would increase sustain like other distortion circuits with progressive compression.

There's a "folding gain" -type of action in the circuit: At output amplitudes below diode Vf the gain is - as usual - defined by the feedback ratio, and typically high-ish. Above Vf the forward biased diodes sort of act like switches and the gain drops to about unity (or some other value if we introduce series resistance to diodes-). If you plot transfer function you can see how it "folds" instead of brickwalling, i.e. "clipping". So yes, compression is evident.

The instantaneous gain compression is what provides the "soft clipping". We could achieve a similar effect also with inverting amps and easily too: just introduce ample series resistance to diodes, enough to limit gain to about unity.

But problem of this type of operation is that if we pump the output level too high majority of signal amplification now takes place at the compressed part of the transfer curve, which is linear but just lower gain. The low amplitude parts of the signal on the other hand are then amplified too excessively and the "crossover region" thus becomes distorted.

So we really should limit the input signal to proper range to get this circuit to work like archetypal diode clippers. Wink. Peavey's "Saturation" generation of amps feature a clever voltage divider for the diodes that effectively increases Vf in relation to output signal and thus allows the circuit to operate at wider levels of output signal amplitudes than about 1 to 2 volts.

[pdf64]

I only use vintage or vintage type PUs, so I can't comment here.

I just can't see, how a 10% or lower "clean signal content" could make a significant difference.

Any PAFs or P90s types in there?
The output of even regular vintage types of those can, at least with the initial note attack and decay, match or exceed the 1.2Vpk-pk of the clipper

[Helmholtz]

I think the triangular signal response measured at the output of the "clipper" stage of a TS doesn't look completely different (might even have measured/scoped that long ago).
Of course gain ratios are different and the middle section is missing, but there is a triangular "roof" as well and the stage never really clips.

I think the reasons are:
- The added input signal,
- The instantaneous large signal gain can't drop to zero (more exactly one) as the dynamic (AC) resistance of the diodes has a lower limit given by the internal series resistance typically greater then 100R. So there will always be a minimum large signal gain between 1 and maybe 2, thus preventing actual signal clipping.

Looking at the output signal, it seems that the gain around the top is about 2. This would means that there's about 50% clean signal content in the peak.

[Helmholtz]

Any PAFs or P90s types in there?
The output of even regular vintage types of those can, at least with the initial note attack and decay, match or exceed the 1.2Vpk-pk of the clipper

Yes and yes.

Nevertheless I like my TS with a Fender amp as it is, no desire to modify. (Drive control at 11:00). Nothing wrong with SRV's strat tone.
Sometimes I prefer the Zendrive with an LP, but the circuit principle is the same. But then I miss the integrated buffer.
I've experimented with different component values some 10 years ago but finally settled with original values.
The TS is certainly not perfect (doesn't sound like a JTM45 with KT66s ) but I like it better than all the maybe 2 dozens of OD pedals I've tested.

Most other overdrive/distortion pedals produce nasty IM effects with double-stops.

BTW, meanwhile I've come to the conclusion that the percentage of added clean signal must be larger than I thought.

[pdf64]

Absolutely nothing wrong with a TS, unless (for me) it’s used with higher output pickups and a dead clean amp.
I use mine all the time to push an ‘on the edge’ amp into overdrive.