series coupling caps?

[embotone]

I am adding this Kevin O'Connor effects loop to an amp I am building.  It will be feeding a .047uf coupling cap going into my phase inverter.  But that will leave me with two .047s in series.  I have used this loop before, it works well, but I don't remember what I did regarding this situation.  Advice on implementing it this way?. Or can I feed directly to pin 2 of my LTPI, and leave out the 2nd .047uf cap?  That is what I am thinking I should do.  Kevin's book is not clear regarding this, and I am, well, I guess I am smart enough to ask.

[HotBluePlates]

... two .047s in series.  I have used this loop before, it works well, but I don't remember what I did regarding this situation.  Advice on implementing it this way?. Or can I feed directly to pin 2 of my LTPI, and leave out the 2nd .047uf cap?  ...

Make the coupling cap into the PI much bigger, like 0.47µF. 

Being 10x bigger than the loop's cap, the resulting value is approximately the same as the loop's cap:
0.047µF || 0.47µF = (0.047 x 0.47) / (0.047 + 0.47) = ~0.043µF

[sluckey]

You need a cap on the input of the PI to protect the hi-Z bias. And you also need a cap to prevent B+ feeding back into the loop circuit via that 1M resistor. Don't worry about the PI input. Fender was fond of a .001 cap there.

[PRR]

If it is a Fender-like long-tail splitter, the input impedance is much more than the grid resistor (because bootstrapping). A 0.01u will be very ample. Maybe too ample for a hard-flogged amp, 0.001u will shave the flub.

But yes. There are 3 DC levels here, which must not mix, so you need 2 caps.

[embotone]

Thanks kindly for the advice.  I've not been here for a while, good to see some of the same folks.  Are you guys getting as old as I am?