Plug-n-Play Buffered Electrical Bypass Using CD4053 and Momentary SPST

[Andon]

Updated schematic attached, as well as initial PCB layout.  Following the theme of RG's advice above, I went ahead and added DC blocking caps to the effect send/return lines (can replace with a wire if not needed).  Those ICs and 1uF film caps take up a lot of space.  SMD should shrink that considerably, particularly if I use ceramic caps.  Ceramic or electrolytic for the DC-blocking caps in the signal path (as SMD film caps seem to be very expensive)?

You can use 0805-sized components just fine for most audio path applications, so you can change those resistors and caps to their 0805 footprints and save some space that way. The only time I leave electrolytic capacitors as-is is when they're in a power supply/conditioning setup (as you have here), or if it makes more sense due to their size/availability.

For the electrolytics you can also use SMD packages - 4mm x 5.4mm for the 47uF, and 6.3mm x 7.7mm for the 100uF. The 100pF, 100nF, and 1uF capacitors can all be 0805 package size, just make sure to use as high-of-quality as you can (X7R minimum, NP0/C0G ideally).

Those ICs are also available in SMD footprints as well, if you feel so inclined.

[matt_garman]

Yeah, this through-hole version is mostly a prototype.  After the design itself seems good/stable, I'll have some all-SMD boards made up; those should be fairly small.

I spent some time reading old threads here on what kind of caps to put in the audio path.  Looks like surface mount MLCCs might not cut it for measured HiFi circuits, but probably good enough for guitar effects.  It's one of those "three goals, pick two" scenarios: high capacitance, small size, ideal capacitor behavior.

Does anyone know what Boss uses for >= 1uF coupling caps in their SMD pedals?  If they are using 0805 (or smaller) MLCCs, then it's good enough for me.

[matt_garman]

It's a success!  I did a writeup on the PedalPCB forums.

I built it almost exactly per the schematic, with a couple minor changes:

• R4 and C3 were intended to be an RF filter; I jumpered R4 and omitted C3
• I used 22uF for C2 (VREF decoupling), as noted above, even 10uF is probably fine here

Everything worked fine on first power-up.  Thank you everyone that helped!

I have plenty of spare boards available, I'm happy to give them away for free, please send me a PM if you'd like me to mail you one or two.

[mictester]

Is the main advantage of CD4049 over CD40106 just the current sourcing/sinking ability? 

In practice, the Schmitt ICs (like the 40106 and the 4093) are better for your simple push-switch operated bistable.  The 4049 will work (as suggested above), but the switching thresholds of the IC are much less clearly defined, and switch-bounce can become an issue.  The Schmitt-input ICs have very clearly defined switching thresholds and tend to be more reliable long-term. 

Switching your LED with the 40106 gates is fine - it's well isolated from anywhere the audio is passing, but I always separate the logic from the audio with a small value resistor in the positive supply to the logic (I often used 33Ω, simply because I had a box of thousands of that value!).  I would also put a small electrolytic from logic + supply to ground after the resistor, just to have a bit of supply filtering.  I never got pops or clicks from logic-switched audio in lots of pedals (and other products) as long as there wasn't any DC offset being switched with the audio and as long as the layout was sensible.

There's one minor caveat that many people forget when they use CMOS for audio switching or routing - it's never good to expose the CMOS (or the inputs to FET op-amps for that matter) to the Outside World™.  I would always include some kind of bipolar transistor buffer or protection diodes to prevent static splats or even high induced voltages from nearby wiring from killing the sensitive parts in my gear.  It always adds a little complexity, but pays dividends in reliability!