High- and Low-Pass Filter in Feedback Loop (GGG-Schematic Problems)

[GibsonGM]

If there is no capacitor at the output of the opamp, you don't have to "re-bias" the next stage...in other words, if the DC offset can be carried to the next input, then you don't re-bias.

Most times, the output of many devices such as opamps, BJTs, mosfets, tubes, and more will have a "DC blocking cap".  This cap...blocks DC (bias voltage) to keep it in the part of the circuit where it is doing work.  The cap also allows you to shape the tone between stages, and therefore is pretty important.    When it is present, this means you must treat the next stage separately, and bias it accordingly....

>>Bias point is simply the 'center point' that your signal rides around. It can go above this, and also go below this point (if you did not bias 'up', the negative part of the signal could not be amplified, as ground would be zero volts (in most circuits) and the signal must go below this when it crosses zero).   Try to keep 'ground' and 'bias point' mentally separate if you can - they are a little different than each other, although they can be in the same place at times, primarily if you are using a dual polarity power supply so that the signal CAN swing below zero - then it's ok to use zero as your ground, and will also be the bias point!   

[PRR]

> Quite everywhere where I'd expecting gnd is now bias.

In single-supply (battery) design, audio ground often rides on a half-supply bias so the opamps will swing both ways.

> do I have to bias the signal in front of every one

No and yes. In a no-gain circuit you can let the DC bias ride through. But many stomp-pedals are HIGH gain circuits. If you DC-couple the whole path the high DC gain and DC errors may throw the late stages to one rail or the other. Also we usually do NOT want high gain at low frequencies. It flubs the speaker and makes ugly distortion. We very often AC-couple every stage or few, to lose the DC and lessen the bass. After each AC coupling DC bias must be re-inserted.

[Rob Strand]

If you DC-couple the whole path the high DC gain a

Just to be clear.   

https://www.circuitlab.com/circuit/z7vfwk/screenshot/540x405/

DC gain occurs when you don't have cap in series with R2.   So in this case it would be wise to keep C.

However if you have a cap in series with R2 then you can remove C (and  R1) and DC couple the bias through.

If you DC couple many stages then in theory the worst-case DC offset on the last stage increases.  While not usually a problem for audio it can be an potential problem for other fields.

If you have DC across any pots (say DC coupled volume pot) then there is more chance for pots to become scratchy.

[antonis]

To help Rob to be more clear...
<DC gain occurs when you don't have cap in series with R2.   So in this case it would be wise to keep C.>
C "isolates" DC path to GND through R2 (also called Gain resistor) hence there is no gain for DC in this branch..
But DC appears through R3 (feedback resistor) which resistor value doesn't mind at all in this case 'cause it only acts as a wire (like just in voltage follower/buffer).
R2 has an open leg for DC so it can be safely omited..
For such a non-inverting arrangement, DC gain is unity..