Signal drops for a second when engaged/disengaged

[Pmelius]

So after playing tonight with a couple dirt pedals that I built, I realized that a common issue is when I switch them on or off, there is a brief signal drop.  It doesn't seem to happen exactly every time either and the signal kind of "fades" back in. All of this happens in less than a second. 

I feel like I've read about this in the past, but can't seem to dig up what or where.  None of these pedals have a pulldown resistor.  I know that these help with switch pop (which I'm not experiencing), but do they also serve this function as well?

The "effect input grounded when bypassed" method was used on a couple of the switches, and a couple were just wired for switching and nothing else.

[antonis]

Power Supply filter cap charging..??

[Mark Hammer]

What Antonis means is that maybe the bypass switch cuts the power to the effect circuit in bypass mode?

[antonis]

What Antonis means is exactly what Mark asks.. 

[Pmelius]

What Antonis means is that maybe the bypass switch cuts the power to the effect circuit in bypass mode?

This could be it, now that I think of it...When I was wiring these switches, it might have been that I was also switching the power supply to the effect instead of just switching the LED ground on/off.

[Mark Hammer]

Yep, that's the culprit.  If there is any medium-large value electrolytic straddling V+ and ground (e.g., 47-220uf) to smooth out the power, that requires a few moments to charge up again when power gets reapplied.  Few, if any, commercial pedals cut the power to the circuit board in bypass mode.  The temptation is always there, since it would hypothetically conserve current.  But it "confuses" pedals, so it is typically avoided.

[teemuk]

I assume there isn't FET switching involved...

But if is, they commonly have a little "fade" implemented in design that prevents the "pop" sounds.