Connecting capacitor in overdrive to ground or Vbias?

[flo]

In many overdrives that use diodes in the feedback loop of a non-inverting opamp stage, there is a resistor and a capacitor in the "gain attenuation circuit part" that influence the gain for low frequencies. Sometimes the capacitor is connected to ground, sometimes to Vbias. I read that this can influence the sound of the overdrive but I don't understand how.

I do understand that the "Vbias making circuit", like a resistor voltage divider and capacitors, can present a higher impedance than ground but if it is decently decoupled I expect the AC signals to "regard" Vbias to be the same as ground. So why would this make a difference in the overdrive sound?

I have not tried nor heard this difference yet but I'm curious if others have ...

I've tried searching the forum about this subject but could not find any threads/subjects about this topic. I assume that this has been discussed before and that my searching skills are lacking. Is there someone that can point me to topics about this or can explain how connecting this capacitor to ground or Vbias can change the overdrive sound?

[R.G.]

It's discussed pretty much every time a new crop of DIYers discover the issue.

Pretty much to the extent that the Vbias is low impedance, it makes no difference. To the extent that the bias voltage has some significant impedance at any frequency of interest, it has to be taken into account.

In general, you can see how much difference it makes by computing the ratio of the Vbias to ground cap to the gain setting cap. The AC impedances of these two parts are proportional to the inverse of their capacitance, so if the Vbias cap is 10X the gain setter, it contributes only 10%. If it's 100x, it contributes only 1%. Since gain setting caps are usually less than 0.1uF and Vbias caps are usually 10uF or more, the ratio is nearly always greater than 100, and the difference between connecting the gain cap to real "ground" and the vbias point is less than 1% on an impedance basis. It's even less than that given the logarithmic nature of human hearing.

[flo]

Thanks for that explanation.
Now I'll try to figure out what happens if I use a "bad" Vbias that has fairly large impedance.
(which was probably also discussed before but I still can't find the relevant topic threads. )

[R.G.]

Thanks for that explanation.
Now I'll try to figure out what happens if I use a "bad" Vbias that has fairly large impedance.
(which was probably also discussed before but I still can't find the relevant topic threads. )

It does two things.
1. It lowers the gain that you're trying to get by introducing more impedance in what is almost always a gain setting resistor.
2. It injects the current noise into the Vbias network, and so the voltage generated by the current from the gain-setting resistor/cap through the Vbias impedance is added to everything connected to the Vbias network. Got a funny oscillation that won't go away? Hmmm...

There's really no good reason to connect it to Vbias unless connecting it to ground is physically difficult. Well, OK, there is one. At start up, the cap in the gain setting string has to charge up to Vbias. This puts a thump on the output of the amp which the gain-set cap serves as it charges up. Usually this is microseconds, but can be longer. Depend on the values of everything else. That last statement is not a hand-waving generalization, it's largely technically accurate.