1
\$\begingroup\$

I've developed a circuit and custom PCB for use in a guitar pedal effects chain. This takes the input signal and switches it between 3 distinct external loops, (A, B, C) such that you can arbitrarily switch between loops to three different pedals, and switching to any one will bypass the other two loop circuits. Only one loop can be active at a time.

I've set this up with solid-state relays so that say, a momentary switch press on A will

  1. turn on the A circuit relay and latch it open, which routes the signal to that loop.
  2. break the latch ground for the B and C relays, unlatching any that are on and switching the signal to bypass those loops.

I had some test boards printed and put everything together, and while the circuit switching fundamentally works as expected, there are some issues I discovered that make it not really usable in the current state:

  1. Whenever any of the switches are pressed, there is a loud "pop" sound, louder than the relative volume level would be normally. I can't imagine it's good for the amp or anyone's ears.

  2. The actual output signal volume is extremely attenuated to the point that I can't really use the switch even with the popping issue, because the relative volume of other effects (when I bypass this completely) would be way too loud if I turned up the amp to compensate.

It's worth noting that each effect return has a 100k volume pot to trim levels between the three effects, however the low volume problem doesn't change at all even when all loops (and thus the volume pots) are bypassed.

The signal input and output, and all three loop grounds are grounded to the aluminum case, as well as each volume pot. The 9v power negative is not grounded to the case, as it doesn't touch the signal chain ever. the 9v circuit only turns on the relays and LEDs. The DC power jack is plastic and is not grounded to the case (center negative).

Any clues as to what might be causing these issues? I've searched around and there are several possibilities it seems so I'm not totally sure which if any of them are the problem.

schematic

\$\endgroup\$
4
  • 1
    \$\begingroup\$ So even with all the loops bypassed, the signal is treated to three diode drops...this can't be good. \$\endgroup\$
    – user156429
    Commented Jun 19, 2020 at 3:48
  • \$\begingroup\$ @TylerStone can you explain that? I'm not a professional at this \$\endgroup\$ Commented Jun 19, 2020 at 17:34
  • \$\begingroup\$ Removing the 1N4448 diodes should restore at least some of the signal that is being attenuated. Trying to get my brain wrapped around your switiching setup, and it appears as though a bypassed loop's "output" will be connected to the output of whatever device is last in that loop's chain...I would probably approach this by creating a three-output JFET buffer that sends signal to all three loops, and have the relays switch between the three "returns". \$\endgroup\$
    – user156429
    Commented Jun 19, 2020 at 19:31
  • \$\begingroup\$ Regarding the popping - if you're absolutely sure there isn't any DC hanging on any of your audio signal lines, the first thing I'd try is putting a reverse biased diode across the relay coil(s). An RC snubber can sometimes be more effective, but you'll have some 1N4888s to use anyway. \$\endgroup\$
    – user156429
    Commented Jun 19, 2020 at 19:36

1 Answer 1

1
\$\begingroup\$

The problem is the diodes on the audio path, it simply does not work if there is even one diode on the audio path, since audio is AC signal.

Simply arrange the DPDT relays to either bypass the audio, or route it via the send/return connections. You should use both relay poles for this so you can't drive the LEDs with relay, but there might be ways to drive the LEDs with other ways from the foot pedal switches.

\$\endgroup\$
2
  • \$\begingroup\$ normally you can switch the LEDs with the foot switch, but these are momentary, so I need some other kind of latching mechanism to keep them turned on \$\endgroup\$ Commented Jun 23, 2020 at 22:44
  • \$\begingroup\$ looks like switching the send return is going to be the best option here, and I'll probably just need more relays to control the LED. \$\endgroup\$ Commented Jul 19, 2020 at 18:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.