Importance of IV knee graph of a Diode in soft clipping applications

[Steben]

That is quite intriguing.
How do the diodes have their response beyond 1.0V? Do they converge?

No, the softer diodes then start to produce a higher output.   The easy way to think about it is the soft diodes squash less.   Once two diodes hit the same point on the VI curve the softer one will continue to rise whereas the hard one will stay flat.

On the 1V DC curve, you can see the soft diode has a *higher* slope (around the point 1V in 600mV out) than hard diode so by extending the line the soft diode will rise-up more.

The "softest one" almost looks like a schottky with some extra resistance in series or a high treshold germanium.

It's definitely around that zone.

If I get time, I'll try show tweaking the slope with resistances isn't the same as having a diode which has the lower slope.   The slope the eye sees has the same problems as the knee the eye sees.    There is only a fixed slope because of the scale it is drawn on.  If you draw the curves on different scales you get different slopes.   For a reasonable upper limit in currents, the diode curves are always curving whereas with resistors the slope levels off.   The curvature is what produces the upper harmonics.    Especially the nasty harmonics which are far away in frequency from the fundamental.   A resistor will not produce harmonics.   If your guitar signal is 200Hz and the harmonics are 4kHz  that's a lot of harmonics away and only curvature can produce that.

I understand. Although on a given scale (voltage span) the effect of crossing curves comes close to added resistance, hence the higher output, dynamics and lower harmonic content.
That is why germanium and silicons sound different in the SAME circuit, but can sound alike in different circuits.

[Steben]

  2015

[Vivek]

Effect of "Compliance" in clipping












Based on plotting following equation

V=(abs(V(Out1))< clip)*V(Out1)
+(V(Out1)>= clip)*(clip+((V(Out1)-clip)*compl))
+(V(Out1)<=-clip)*(-clip-((-clip-V(Out1))*compl))

Where V is the output, Out1 is the signal that has to be processed, clip is clipping voltage, compl is compliance after clip

Compliance = 0.5 means

If there is a signal with 3V peak
and its clipped at 0.65V

then it means that the input signal has 3-0.65 volts more to go after clipping

But 0.5 compliance means that the output signal will have a peak of 0.65 + 0.5(3 - 0.65) V

[Vivek]

Boneyard, Carl Martin Plexitone

Why did designer make 5-7 mA go through LED in the feedback loop ?

What extra benefit did he get instead of only allowing 100uA into the LEDs ?

[Rob Strand]

V=(abs(V(Out1))< clip)*V(Out1)
+(V(Out1)>= clip)*(clip+((V(Out1)-clip)*compl))
+(V(Out1)<=-clip)*(-clip-((-clip-V(Out1))*compl))

So the way I look at this these days would be,

   V1 =  abs(V(Out1))< clip)*V(Out1)

which is just the clipped waveform. 
ie.  If V(out1) > +clip then V1 = +clip, if V(out1) < -clip then V1 = -clip  otherwise V1 = V(out1)

Then make the combination,

   V  = V1 *(1-comp)  + V(Out1) * comp

which is just a mixer or blender:
- when comp = 1 only the clean signal comes through.
- when comp = 0 only the clipped signal comes through.
- when comp = 0.5 equal amounts of clipped and clean signal are mixed together.

The important point here is this V is *identical* to the one you gave.   All I'm doing is writing it differently.

The second view makes it very clear what you are listening to - a blend of a hard clipped signal and the clean signal.  The character of clipped signal is nasty.  And adjusting comp does not change that nasty character.   So comp appears to soften but it's actually not.  All comp does is make the nasty signal softer.

A softer transition for the "clipping" would change the character of the clipped signal.    That would appear as a more rapid tapering of higher harmonics (in the same way that a triangle wave sounds less nasty than a square wave).  In the blend case the tapering of the harmonics is fixed and is determined by the hard clipping.

That is why germanium and silicons sound different in the SAME circuit, but can sound alike in different circuits.

And it's pretty much why this holds.    The parallel resistor is the only thing you can play with and it has limited scope to change things.

What extra benefit did he get instead of only allowing 100uA into the LEDs ?

Smaller R's make the clipping a little harder.

[Vivek]

Respect and Thanks, Sir Rob !!!!!


PS : Why have 2 LEDS in series in PLEXITONE ?

[Rob Strand]

PA : Why have 2 LEDS in series in PLEXITONE ?

My guess is since the non-inverting stage blends a bit of clean putting in two LEDs waters down that clean component.   The whole clipping stage is setting itself up for  a harder-clipped sound (but not necessarily nasty, maybe less hair at low drive).     The reason they can get away with two LEDs in series is because they have up'ed the supply voltage.    Two LEDs on a 9V set-up will clip the opamp in most cases.

[Vivek]

When this thread has run its course, I wish we had a synopsis post :

A) Adding more diodes in series does this :

B) Changing parallel resistor does this :

C) Increasing current through the diodes does this :

[Vivek]

PA : Why have 2 LEDS in series in PLEXITONE ?

My guess is since the non-inverting stage blends a bit of clean putting in two LEDs waters down that clean component.   The whole clipping stage is setting itself up for  a harder-clipped sound (but not necessarily nasty, maybe less hair at low drive).     The reason they can get away with two LEDs in series is because they have up'ed the supply voltage.    Two LEDs on a 9V set-up will clip the opamp in most cases.

Maybe 2 LEDS is the cause and higher voltage requirements is the effect

They decided on 2 LEDS in series because of ...........

But the Opamp will clip with 9V supply and 2 series LEDS in the loop

Hence they needed to up the supply voltage

[iainpunk]

When this thread has run its course, I wish we had a synopsis post :

A) Adding more diodes in series does this :

make the amount of clean, compared with dirty, smaller (harder clipper)

B) Changing parallel resistor does this :

change the gain, add more clean input to the output

C) Increasing current through the diodes does this :

make the knee harder

[Vivek]

When this thread has run its course, I wish we had a synopsis post :


B) Changing parallel resistor does this :

change the gain, add more clean input to the output

I should have said :

B) increasing the parallel resistor does this :

[Steben]

PA : Why have 2 LEDS in series in PLEXITONE ?

My guess is since the non-inverting stage blends a bit of clean putting in two LEDs waters down that clean component.   The whole clipping stage is setting itself up for  a harder-clipped sound (but not necessarily nasty, maybe less hair at low drive).     The reason they can get away with two LEDs in series is because they have up'ed the supply voltage.    Two LEDs on a 9V set-up will clip the opamp in most cases.

Higher treshold means less clean component and indeed this reflects in harder clipping character.
Less "softer than amp drive" character of for example the Tube Screamer. let's say IMHO low clipping tresholds in overdrives only work as opamp clipping protection in a pedal that is used as booster and not as main drive tone. Most valued overdrives have one thing in common : higher treshold. The Plexitone is in fact a really good example and sounds great with single coils to get that snappy strat-in-amp break-up timbre without clipping the opamp (which happens almost all the time in diode to ground clippers).

[Vivek]

Thanks !!!

I know somebody had posted the math behind "More diodes means less of clean signal is added with the clipped signal"

Could you please post it again.

[iainpunk]

When this thread has run its course, I wish we had a synopsis post :


B) Changing parallel resistor does this :

change the gain, add more clean input to the output

I should have said :

B) increasing the parallel resistor does this :

higher gain, the knee doesn't change, but the amount of clean sticking out the top is more.

cheers, Iain

[iainpunk]

Thanks !!!

I know somebody had posted the math behind "More diodes means less of clean signal is added with the clipped signal"

Could you please post it again.

can i ''visualize'' it for you, without the numbers?
if the clipping threshold is surpassed, the gain is 1, AKA you add clean signal
if the clipping threshold is 0,6V and the gain is extremely high, there is basically a addition of the 0,6V square wave to the clean signal.
if the clipping threshold is 2V and the gain is extremely high, there is basically a addition of the 2V square wave to the clean signal.
in comparison, there is less clean on the 2V square than the 0,6V square because the gain after the threshold is 1. that's why i don't recommend putting a signal larger than 1/3 the clipping threshold into soft clipping high gain, because it will sound like clean with a bit of distortion going on in the background, instead of real soft clipping like in tubes. this is also why i don't think its wise to use compliance resistors in soft clipping, it just makes it worse.

cheers, Iain

[Steben]

Thanks !!!

I know somebody had posted the math behind "More diodes means less of clean signal is added with the clipped signal"

Could you please post it again.

can i ''visualize'' it for you, without the numbers?
if the clipping threshold is surpassed, the gain is 1, AKA you add clean signal
if the clipping threshold is 0,6V and the gain is extremely high, there is basically a addition of the 0,6V square wave to the clean signal.
if the clipping threshold is 2V and the gain is extremely high, there is basically a addition of the 2V square wave to the clean signal.
in comparison, there is less clean on the 2V square than the 0,6V square because the gain after the threshold is 1. that's why i don't recommend putting a signal larger than 1/3 the clipping threshold into soft clipping high gain, because it will sound like clean with a bit of distortion going on in the background, instead of real soft clipping like in tubes. this is also why i don't think its wise to use compliance resistors in soft clipping, it just makes it worse.

cheers, Iain

Unless you add clipping in front of the opamp stage, which sets the maximum amplitude added on top of the treshold and effecticely acting as a top clipping element. it makes the stage "immune" to heavy signals

[Vivek]

So, given all the discussion here

What is the best way of making an Amp in a Box with 3 or 4 clipping stages ?

what should each stage be ?

[Steben]

So, given all the discussion here

What is the best way of making an Amp in a Box with 3 or 4 clipping stages ?

what should each stage be ?

Probably you refer to gainy type? not shiny fender in a box?
I suggest 2 asymmetrical clipping stages with capacitive coupling followed by tone control and finally a symmetrical clipping stage.

[iainpunk]

So, given all the discussion here

What is the best way of making an Amp in a Box with 3 or 4 clipping stages ?

what should each stage be ?

2 simple jfet stages (biased at 6V for asymetry) and one jfet/mosfet cascode at the end (with high feedback and biased to be symmetrical).
(i dislike feedback clipping opamps, a lot)

cheers, Iain

[Fancy Lime]

So, given all the discussion here

What is the best way of making an Amp in a Box with 3 or 4 clipping stages ?

what should each stage be ?

If there was a single best way to build any specific type of pedal, this forum would be a lot less interesting. It totally depends on what sound you are after. And even then there are probably several equally good and equally simple/complicated topologies that get you there. I for one like diodes in the NFB path of an inverting opamps better than a non inverting because I find it easier to tune the clipping characteristics. Matter of taste and personal journey, though. I also rarely find more than two clipping stages necessary, unless I am looking for extreme or experimental sound, which is probably not the case for an amp in a box type of pedal.

Well, that wasn't helpful at all, was it? Sorry. Anyhow, the best way for you is your way, grasshopper.

Cheers,
Andy