freestompboxes.org

Merlin started an interesting thread over at DIYStomp yesterday. Link: https://www.diystompboxes.com/smfforum/ ... c=107560.0. Basically, he setup a clipping circuit and ran some tests using a 1kHz sine wave through various diodes. His findings were that, with the exception of forward voltage, all diodes clip in roughly the same manner. He also ran the test at 250Hz and said that the graphs looked the same, though it's not posted. The point was that EQ and gain have more to do with tone than the particular choice of clipping diode, and that new designs should be more focused on the tweaking the former rather than the latter. However, it was still noted throughout that changing the diodes in an existing design will indeed alter the tone. So I posed the following question: Excepting forward voltage, what exactly causes the tonal difference between diodes? This was the answer I received. Merlin's response is in bold.

I understand that, but what about the difference between 4148s and 4001s?  They both have a FVD of around 0.7V, right?  Yet in my own experimentations with a boutique TS circuit I built, I found the 4001s to have a much more mellow sound.

The 4001s will have more junction capacitance, which might make a different if you have a 1Meg feedback resistor and no other capacitor there.

So this is my question to you: What the hell is up with capacitance, tone, and clipping characteristics? I've seen it used to describe the relative hardness/softness of the 'knee', but Merlin's findings have me very confused. They suggest that the 'knee' may me more of an aural phenomenon rather than a physical property of the diode. I also read some university PowerPoint slides on junction capacitance, and one thing that caught my eye was that the diodes still leak a bit of current through the cathode after they switch from on to off. In that split second of time between states then, would a diode with a 15pf junction capacitance more or less be acting as a discharging 15pf capacitor? I'm building a Rat for my EE100 final project, and I've got to explain this stuff beyond the oscilloscope. I asked this on the other forum, too, but the thread's gone cold .

Tall Steve wrote:Merlin started an interesting thread over at DIYStomp yesterday. Link: https://www.diystompboxes.com/smfforum/ ... c=107560.0. Basically, he setup a clipping circuit and ran some tests using a 1kHz sine wave through various diodes. His findings were that, with the exception of forward voltage, all diodes clip in roughly the same manner. He also ran the test at 250Hz and said that the graphs looked the same, though it's not posted. The point was that EQ and gain have more to do with tone than the particular choice of clipping diode, and that new designs should be more focused on the tweaking the former rather than the latter. However, it was still noted throughout that changing the diodes in an existing design will indeed alter the tone. So I posed the following question: Excepting forward voltage, what exactly causes the tonal difference between diodes? This was the answer I received. Merlin's response is in bold.

I understand that, but what about the difference between 4148s and 4001s?  They both have a FVD of around 0.7V, right?  Yet in my own experimentations with a boutique TS circuit I built, I found the 4001s to have a much more mellow sound.

The 4001s will have more junction capacitance, which might make a different if you have a 1Meg feedback resistor and no other capacitor there.

So this is my question to you: What the hell is up with capacitance, tone, and clipping characteristics? I've seen it used to describe the relative hardness/softness of the 'knee', but Merlin's findings have me very confused. They suggest that the 'knee' may me more of an aural phenomenon rather than a physical property of the diode. I also read some university PowerPoint slides on junction capacitance, and one thing that caught my eye was that the diodes still leak a bit of current through the cathode after they switch from on to off. In that split second of time between states then, would a diode with a 15pf junction capacitance more or less be acting as a discharging 15pf capacitor? I'm building a Rat for my EE100 final project, and I've got to explain this stuff beyond the oscilloscope. I asked this on the other forum, too, but the thread's gone cold .

Maybe, but I think that the time taken for that 15pF cap to dis-charge might mean that it's effects are inaudible, or above the range of frequencies that the human ear can detect, personally, i think that the shape of the knee has more effect on the tone of the clipping, remember that the knee of a Germanium Diode is softer than a Silicon Diode (the 4000 series of diodes like the 1N4001 for example), softer clipping generates less upper harmonics so the tone won't be as buzzy as it would be for sharper clipping, also, the forward-voltage of the diode will affect the tone subtly by causing the clipping to be slightly asymmetrical........

DrNomis wrote: Maybe, but I think that the time taken for that 15pF cap to dis-charge might mean that it's effects are inaudible, or above the range of frequencies that the human ear can detect, personally, i think that the shape of the knee has more effect on the tone of the clipping, remember that the knee of a Germanium Diode is softer than a Silicon Diode (the 4000 series of diodes like the 1N4001 for example), softer clipping generates less upper harmonics so the tone won't be as buzzy as it would be for sharper clipping, also, the forward-voltage of the diode will affect the tone subtly by causing the clipping to be slightly asymmetrical........

Thanks! The knee and its effects are what I'm confused about, especially after the test. The test results seemed to diminish its significance. More junction capacitance = slower turn-on time = softer knee = less high end response, right? Also, in a feedback loop, if the diode does act like a 15pf cap even for just a split second, wouldn't that add to the overall capacitance of the smoothing capacitor?

Perhaps I didn't understand your question right but let me put some of my understanding to it :

The time the parasitic cap needs to decharge during polarity change is equal to the time it needs to charge in opposite direction, so the additional leakage
makes it look like a normal cap parallel to an ideal diode.

If this is true or even audible I don't know,
but I compared different Vf diodes in different gain settings and there where some of them which sounded exeptionally different but not for their Vf but for their capacitance which stood out before others.

That is exactly what I was looking for! Thanks! So what you're saying is that parasitic capacitance in the diode acts as a sort of bypass cap across the diode?

Tall Steve wrote:That is exactly what I was looking for! Thanks! So what you're saying is that parasitic capacitance in the diode acts as a sort of bypass cap across the diode?

In a way it does, but, there's also the resistance that the diode presents to the parasitic capacitance when the diode is forward-biased, which is probably a few ohms, this means that the parasitic capacitance will dis-charge almost instantaneously, I'm guessing that the resistance the diode presents when reverse-biased is probably infinite (ie, virtually open-circuit), a Silicon diode typically needs a forward-bias voltage of about .6V across it before it will conduct, whereas a Germanium diode only needs about .3V, now the Germanium diode doesn't suddenly switch-on like a Silicon diode does, it gradually turns on because the electrons in the Valence-Shell of a Germanium Atom are not as strongly held in it as they are in a Silicon Atom, so, they require less energy to knock them free, this is why Germanium diodes have a soft turn-on knee.....

Fantastic! Thank you guys so much!