Getting ready to build my mag/piezo switching box, review needed!

[BlueLdr]

Hey everyone, I'm looking for a second pair of eyes on my buffered mixer/switching project. To (re)summarize, I'm building a pedal to switch between/mix the mag/piezo signals from a stereo output guitar. This pedal will be the first thing in the signal chain, so I have buffers on all the inputs, which are more or less the same as the Klon's input buffer. The mixer is based on PRR's design here. Instead of each channel having its own volume, they're cross-faded using a dual gang pot. EDIT: All opamps are TL072s.



For the mixer routing, I picked up a few PedalSync MV-57s and DPDT relays from Small Bear. The digital logic is all 74HCT chips.


*Note: I moved the final 100K pulldown resistor to be between the 10uF cap and the 470 ohm resistor.

I've breadboarded it all out and it works nicely, but I wanted to get feedback before I call it finalized.

1. Are there any glaring issues or beginner mistakes that I overlooked?
2. The trimpot on the mixer opamp doesn't seem to do much. Is there a better way to ensure that the mixer as a whole is at unity gain?
3. I was having pops when switching, especially with the third relay that switches between the mixer and the "single" out. After re-breadboarding everything, the problem went away and it's now silent. But in the process of diagnosing it, I came across this AMZ article about electrolytic cap leakage, and that got me thinking, is it worth replacing all the 10u caps in the signal path with tantalum caps? I have to do one more parts order anyway so it would only be an extra dollar or two.

I also have the PCB designed in KiCad, which I have another set of questions about. Would it be ill-advised to share the KiCad project files here? If so, what would be the best way to share the PCB for review?

[amptramp]

I am not sure why you need SW1 P2.  Unless I am missing something, the MAG PRE RETURN could go directly to the input on SW1 P3.

The 1K feedback resistors around the op amps is a rather heavy load.  I would increase both resistors connected to the inverting input to avoid excess coupling through the power ground.  If you go up to 10K, you can reduce the capacitors connected to the inverting input to 1 µF and these could be film, eliminating those particular aluminum electrolytics.  The 30K trimpot might then be the right size to actually do something useful.

74HCT logic sounds good if you can drive the relays with it.  You might need a transistor amplifier stage but if it is working as it is, then that is not a problem.

You might prefer to have the switching after the op amp buffer stages for the mag and piezo inputs so you don't get noise amplified in an unused channel and added to the signal.  You should have the logic ground any unused inputs.

It usually pays to have a small capacitor on the order of tens of pF across the feedback resistors in the op amp stages, especially with a trimpot and capacitor to ground in the output stage.  The capacitance across the trimpot and cap would be fairly large and the capacitance to ground would tend to destabilize the amp by shunting the high frequencies to ground, allowing a rising gain with frequency that causes an unwelcome phase shift where the rising response meets the falling response of the op amp.  If you have an op amp like a 5532, this oscillation may occur above the audio range but it may reduce the headroom of the signal.

[BlueLdr]

Thanks for the feedback, I'll try out those changes! (Gotta add some 1u films to the order though)

I forgot to mention, all the opamps are TL072s.

74HCT logic sounds good if you can drive the relays with it.  You might need a transistor amplifier stage but if it is working as it is, then that is not a problem.

The relays are all driven by the PedalSync microcontrollers (datasheet here).