What's All This Stuff About Germanium Diode Leakage?

[YourGuitarist]

So LMK if I'm re-hashing here but the leakage is the failure of the diode to act as an open circuit when "off" ?

AND

The series resistance is the failure to act like a short when it is on?

I would think Vf is easily worked around by scaling input amplitude into the diode circuit.

 

[Chuck D. Bones]

"Failure" is such a harsh word. There are no "ideal" components.

Scaling works in some circuits and not others because scaling is not always a simple option. Replace the the Ge diodes with Si in the distortion+ and the dynamics change. To get close to the original headroom/saturation profile, we'd need to increase the opamp's power rail.

The Burns Buzzaround and derivatives depend on the Vf and leakage of the Ge diode. Si will not do. EQD tried it with the Tone Reaper.

 

[darwin999]

The series resistance is the failure to act like a short when it is on?

I would think Vf is easily worked around by scaling input amplitude into the diode circuit.

In the power polarity detection part of the circuit, then thinking about diode characteristics in terms of the "classic I-V plot" is fine.

But for the signal path part of most effects pedals, the typical voltages are often modest and the typical currents are often very small. So you're way down in the foot of the curves, so now you need to put a magnifying glass on that "classic I-V plot". That's where the deviations from 'open circuit' and 'short' become important, and you need to think about things in more detail because they matter. I suggest looking at some good diode data sheets where they show the forward and reverse characteristics in log plots that go down to very low voltages and currents.

 

[darwin999]

Note added: sadly there aren't good data sheets on germanium diodes that go down to low voltages and currents, especially in reverse bias. But people here have posted data and also discussed the impacts, especially @Chuck D. Bones