Parallel looper issues update

[Spacecommandant]

I was having issues with a parallel looper design and asked for advice a while back to help eliminate switch popping and circuit noise. The attached file [SCHEMATIC_Version1] shows a pop-free version; the switch popping finally disappeared after removing all pulldown resistors. The guitar signal was too quiet, though, and the circuit had a high noise floor. Although the schematic doesn't show it, I was using a voltage doubler.
The second file [SCHEMATIC_Version2] was a redesign using lower noise op amps and other changes. The switching itself remained the same. This version has switch popping issues.
The pulldown resistors at input/output/4PDT lug 4 were added later to try to eliminate the popping but didn't help. I tried lug 6 to Vref as in Version 1 but it didn't help. The LED being removed doesn't help either. I lowered all capacitors in the audio path to 100n to temporarily hear what that did when switching the 4PDT and SPDT(Tails) on and off. There was still a significant pop when either were turned to the on position.
I should note that I used an audio probe to determine that there is *almost* no pop going through the sends, mostly through the circuit output and output jack. I added a 100n capacitor after lug 2 (not shown) and this reduced the pop when switching the 4PDT to the off position.
If there's anything else worth trying, I'd appreciate any suggestions!

[Aentons]

My understanding is that almost all switch popping is DC voltage present on the switch. You might take a multimeter and probe inward starting at the output jack to see if you detect any.

[matmosphere]

I didn't read the earlier thread, but I'm having a difficult time telling what is going on with your bypass setup. Are you trying to have the option for tails with true-bypass? If that is the case I'm tempted to say you need to add a buffer in there somewhere.

I found a setup for something like that with a delay. Maybe this will help.

http://vulcanofx.blogspot.com/2015/03/analog-fet-switch-daughter-board-for.html

[Spacecommandant]

I didn't read the earlier thread, but I'm having a difficult time telling what is going on with your bypass setup. Are you trying to have the option for tails with true-bypass? If that is the case I'm tempted to say you need to add a buffer in there somewhere.

I found a setup for something like that with a delay. Maybe this will help.

http://vulcanofx.blogspot.com/2015/03/analog-fet-switch-daughter-board-for.html

Yes, the circuit is true bypassed when tails isn't engaged.
I temporarily bypassed pole 1 of the 4PDT (connecting R5 directly to R27) to see what would happen and the popping vanished so I've experimented with this in mind.
I moved the bypass connection after R5 to just after C1 instead, and added a buffer before the 4PDT pole 1 and this has reduced the popping (after plugging in power, the popping volume drops to a more reasonable level after 7-10 seconds).
I'd like to experiment with the circuit you've linked. I'm going to try to integrate it into pole 1.
Thanks

[Spacecommandant]

I'm wondering if there's anything else that could help reduce crosstalk between channel A and B? There's still a faint signal of channel B when panned to channel A and vice versa, although it can be quite noticeable with loud effects in the chain. I was advised to raise C29 to 220u in order to get the impedance at ground to the optimum level and it did help somewhat but I'm hoping there are other ways to help. I've tried isolating ground/Vref paths where possible.