What is the consensus on GND (0V) to PCB

[bluelagoon]

What is the consensus on GND 0volt to the PCB board, when there is no battery snap attached only a Direct DC in power jack in option??
As the usual for when there was a battery was to connect the ground via the input jack Right Channel Ring connect point which then shunts to PCB GND., but only when a lead jack is inserted.
Is there still any merit in following same procedure when there is no longer a battery powered option??
Or would it it now be best to take GND direct to the PCB,  via the shortest route?

[duck_arse]

shortest route. that way you bypass a couple of mechanical/electrical failure points. [on a  powered pedal board, do you ever pull your audio plugs from the jacks at power-down time, or just switch off at the wall?]

[ElectricDruid]

+1 agree with Duck. I never bother with the sticking-a-jack-in-powers-it-up thing these days. It was always a pain in the neck to have to unplug everything anyway. Power goes from the DC socket to the board directly and keeps everything tidy.

[R.G.]

First, finding a consensus on anything in a pedal is hard to do.  8-)

I like power ground directly to the board, but for pedals I've designed where there is a need for enabling power when a cord is plugged in, I like to use either a PNP transistor or a MOSFET to turn on power to the board by using the second stereo jack contact to pull down the base/gate of the transistor. You get the same action (power on when a plug is inserted) but don't have the odd situation of your power supply running through the ground wires of the input jack, where any disturbance (like ground currents in the power lines) causing noise by being selectively diverted into the input jack wires.

[ElectricDruid]

First, finding a consensus on anything in a pedal is hard to do.  8-)

<panto voice>OH NO IT ISN'T!</panto voice>

[Rob Strand]

The advantage of connecting the PSU directly to the PCB, preferably to a big cap is it can prevent PSU noise *generated by your circuit* getting into the audio paths.

If your circuit has noise currents on the PSU rail and you wire the DC jack to the input socket then those noise currents will be flowing down the signal ground (between the jack and your board).   (Ideally you want to stop that noise current getting out in the first place.)

If your circuit doesn't have noise currents on the PSU you could argue that making terminating everything at the input socket puts the star ground point at the input of the box.   

Neither situation covers all the cases of being noise free with a shared/daisy chained PSU.   In the second case the noise current can still flow on signal ground between pedals.  It's more likely the first case will work out better for bad cases and if you did have to use a separate PSU then it's more likely to keep the noise out of the audio.

[R.G.]

Rob and Tom, you might be interested. I did some analysis on whether where you put which pedal on a daisy chain matters. Of course, it does. You want to put the ground-noisiest pedals closest to the power supply on the daisy chain. This minimizes the foot-ohms and connector-resistances between the noise source and the other pedals which presumably pollute less. It seems to work out on some ad-hoc testing. We maintain a rogues' gallery of bad actors to test with and it kinda works out.

[antonis]

But we don't build ground-noisy pedals, do we..??

[Rob Strand]

Rob and Tom, you might be interested. I did some analysis on whether where you put which pedal on a daisy chain matters. Of course, it does. You want to put the ground-noisiest pedals closest to the power supply on the daisy chain. This minimizes the foot-ohms and connector-resistances between the noise source and the other pedals which presumably pollute less. It seems to work out on some ad-hoc testing. We maintain a rogues' gallery of bad actors to test with and it kinda works out.

Makes sense.

But we don't build ground-noisy pedals, do we..??

From the thread a week to two ago it seems some commercial pedals still have problems.    There's also different degrees of pollution.   Some charge pump builds are workable but the layouts aren't as good as they could be and maybe in these cases different grounding schemes, power supply connections, pedal ordering can push marginal layouts in or out of the acceptable zone.