What is this diode actually doing?

[carboncomp]

Um....what is the diode actually doing in the MK3 circuit read about 4 different explanations online so thought I would be better off asking here!

And follow up, what ramifications are there in changing forward voltage?

[bluebunny]

Is it doing that clever temperature-compensation thing?  The transistor starts wandering in one direction when things warm up, and the diode wanders in the other direction.  The wanderings cancel out and everyone is happy.

Anyway, someone far more qualified will be along shortly to give you the real explanation.   

[GibsonGM]

I've seen them there for that purpose in other designs, so that seems highly likely, Marc!

[FSFX]

Um....what is the diode actually doing in the MK3 circuit read about 4 different explanations online so thought I would be better off asking here!

The 1N270 diode does a variety of things there. As mentioned by many, the temperature dependant reverse leakage current will try to compensate for temperature induced changes in leakage current of the transistor. That is a DC effect of the diode. The diode will also affect the way the signal is distorted by that transistor. Forget about things like Vf in this case as it is a germanium diode, not a silicon one, and will be acting a bit like a voltage dependent resistor even at very low signal levels, well below what you would normally associate with the Vf of a diode. I recently posted a short article analysing this as it seems to be something that gets misunderstood quite often. 

[Rob Strand]

The same question came up a week ago.

[carboncomp]

Um....what is the diode actually doing in the MK3 circuit read about 4 different explanations online so thought I would be better off asking here!

of a diode. I recently posted a short article analysing this as it seems to be something that gets misunderstood quite often.
[/quote]

Could I get a link, would love to read that.

[FSFX]

Could I get a link, would love to read that.

Although this relates to the series diodes in the Boss HM2, the information regarding the difference between germanium diodes and silicon diodes is still relevant.

www.fleetingspider.com/files/HM2_Diodes.pdf

[PRR]

The same question came up a week ago.

Yes, it did. Around April 16.

Q3 is not even static biased, so "bias compensation" seems dubious. In dynamics, C2 passes signal peaks +/-. Without D1, the transistor's base diode would quickly "grid-block", turn-OFF. D1 makes things more symmetrical.

[FSFX]

Q3 is not even static biased

This subject seems to come up all the time. I was reading some of your responses to this type of question from over 12 years ago on this forum.

As regards biasing of Q3, it is a germanium transistor and relies on its leakage current for biasing, as I am sure you are well aware. As the leakage current very much depends on the actual transistor used and the temperature, I doubt that it would be possible to match the leakage of a germanium diode like a 1N270 to compensate for the transistor's leakage. In fact using a similar transistor to Q3 wired as a diode gives far better temperature compensation but even that is limited to about +/- 10 degrees C around the typical ambient temperature.
As with many pedal circuits, someone must have thought it would be a good idea to try using it there and, as you say, it makes the charge/discharge path to the coupling capacitor a bit more symmetrical but, in practice, it doesn't actually do much with such a low value coupling capacitor and the impedances around that part of the circuit.
Having the diode does attenuate the input a bit so affects the clipping as well as the bias, however, just putting a 33k resistor in there instead of the diode actually produces a very similar result as regards bias and distortion and gives about the same gain and output level for the stage.