Question about ground wiring for true bypass looper

[dave999z]

Hey all, I know there are a million threads about this, but I could not find the answer to my particular questions.

I'm building a multi effect switcher. 

First question...  In most of the diagrams I see, the sleeves of all the send/return jacks are connected together in a line, which then connects to the one true ground point (usually either the sleeve of the input jack or the negative terminal of the power jack).  My question is if I am using Switchcraft jacks that are not insulated, the jacks sleeves will all be connected together via the enclosure itself.  So why is it necessary to also wire the sleeves together?  Doesn't that create the potential for ground loops?

Second question...  Most wiring schemes ground out the send (input) to each effect while bypassed, which I know is a good thing.  But I've seen it done two different ways.

Some diagrams just connect the bypassed effect input to a ground bus that is used for the LEDs, like this...

http://1.bp.blogspot.com/-SzM32SDZGdM/UX_DmZFl5aI/AAAAAAAAFhg/Bp3b9omf8Ds/s1600/TB+Looper+5+with+buffer+and+tuner.png

But others connect each bypassed effect input, as well as the LED negative, back to the sleeve of the effect input jack, like this...

https://www.premierguitar.com/articles/the-diy-true-bypass-lesson-1

Is one way preferred over the other?

Would there be any advantage to doing both -- in other words, connect the LED negatives to an LED ground bus, but ground each bypassed effect input jack by sending it back to its own sleeve.  That would keep the LED circuit completely separate from the audio, except for at the one true ground point.

Thanks,
Dave

[iainpunk]

Euhm, there are a few things id like to point out.

1) you are makeing a lot of loops, with the grounds, i would personally have only one of the send/return pair grounded. This prevents additional ground loops.

2) chassis ground is not always the most reliable, so i would personally use jacks that are isolated and use a wire to make a "†rve" ground (sorry for the black metal joke). Remember point one in this comment while doing so.

3) keep the LED ground away from the signals ground path as long as possible, it there is a battery only for the LEDs, dont ground it, this could only introduce noise.

If you use the battery for both the LEDs and a buffer, try to split the ground as close to the battery as possible, and make the LEDs sit in their own loop.

I hope this helped
Iain

[PRR]

The loop-area matters.

A wire running parallel (and very close) to chassis has very small area.

_Me_, I would screw jack to chassis AND run a jumper wire. For anything as small as a pedal or even pedal-board, this will not be a problem. (There may/will be problems, but not from that.)

[dave999z]

Thanks for the reply.

Unfortunately, isolating the jacks from the enclosure is not an option (enclosure is already built and holes drilled).  So I can't take suggestions 1 and 2.

Still wondering whether I'm just asking for ground loops if jacks sleeves are touching enclosure AND jack sleeves are also wired together.

[R.G.]

Unfortunately, isolating the jacks from the enclosure is not an option (enclosure is already built and holes drilled).  So I can't take suggestions 1 and 2.

Hmm. Both Amphenol and Switchcraft make nylon-body jacks which mount in the same 3/8" hole as metal bushing jacks. If you haven't bought jacks yet, sounds like you're at the perfect place to do this.

I'll insert my standard notes about ground.
Wires are not short circuits with zero resistance. Every wire, including all ground wires, is a low-value resistor. As such, it creates a voltage across it equal to the wire's resistance times the current. So the only way to keep two circuits from interfering through their grounds is to make them have two ground wires, neither of which carries the other circuit's "sewer ground" current.

Everything else is just a question of how much ground interference you can stand, not whether it's zero.

[iainpunk]

Thanks for the reply.

Unfortunately, isolating the jacks from the enclosure is not an option (enclosure is already built and holes drilled).  So I can't take suggestions 1 and 2.

Still wondering whether I'm just asking for ground loops if jacks sleeves are touching enclosure AND jack sleeves are also wired together.

Hmm. Both Amphenol and Switchcraft make nylon-body jacks which mount in the same 3/8" hole as metal bushing jacks. If you haven't bought jacks yet, sounds like you're at the perfect place to do this.

I'll insert my standard notes about ground.
Wires are not short circuits with zero resistance. Every wire, including all ground wires, is a low-value resistor. As such, it creates a voltage across it equal to the wire's resistance times the current. So the only way to keep two circuits from interfering through their grounds is to make them have two ground wires, neither of which carries the other circuit's "sewer ground" current.

Everything else is just a question of how much ground interference you can stand, not whether it's zero.

That grounding note is genius!!
Furthermore, If i want to isolate a jack from a hole without buying fancy nylons, i often tape off the hole's edge with electrical tape, screw in the jack and cut away the excess tape thats sticking out from underneath the washer. Its verry easy. I hope this helped.

[dave999z]

Here is a diagram of how I'd like to wire it...

--The jack sleeves will be grounded together via a wire (signal ground) and also via the enclosure itself.

--The LED circuits will be completed through a different ground path (LED sewer ground).

--The two ground paths will meet at the input jack sleeve.

--While an effect is bypassed, its input will be connected to the LED sewer ground.

Anyone see any problems with this wiring scheme?  Will the signal ground and enclosure ground create ground loops?

[dave999z]

Alternatively, would there be any reason/advantage to doing it like this?

The only difference is that while an effect is bypassed, its input will be connected to the signal ground (i.e., effect send jack's tip will connect to its sleeve) instead of connecting it to the LED sewer ground.

[R.G.]

Here's some food for thought: why is it that the jacks and the power supply for LEDs and such have to be shared? What happens if the LED power supply is completely independent of the signal grounds?

[dave999z]

Here's some food for thought: why is it that the jacks and the power supply for LEDs and such have to be shared? What happens if the LED power supply is completely independent of the signal grounds?

I actually don't see a problem with that?  If the LED sewer ground were connected directly to the negative terminal on the power jack (as opposed to going through the input jack sleeve), and there were no connection at all between the LED power jack negative terminal and the input jack sleeve, it would seem the LED circuit would be completely separate from the signal circuit.  In other words, this:



I don't think that would create a ground loop issue, because there would be no part of the signal path that has two routes to ground?

But none of the diagrams out there is like that.   

[zbt]

As suggested, break the loops ?





Sorry, not tested yet.

[iainpunk]

As suggested, break the loops ?





Sorry, not tested yet.

only one of the send/return jacks should be lifted, and ground should run through from in to out

cheers

[Processaurus]

With unbalanced signals (like guitar pedal world) the ground connection of the input jack is worth special consideration and routing, as any current/voltage drop through that wire can end up as noise in the signal, since it is what the first electronics stage is measuring the input signal against.  The less current those wires carry, the better a ground reference it is. It's good to do the things that take power (LED is an easy one) and take their sewer grounds back to the power supply input, or at least away from the first high impedance stage.  Agree with others it is fine to ground the effect input, in bypass, with the LED sewer ground- you aren't using the effect and don't hear it.

Sadly/ironically we are stuck with a tradition of using a stereo input jack to connect the battery ground- it's a worst case scenario, all of the current flows back through the sensitive input ground wire, for maximum noise on the ground connection there.

Also it is good to wire the jacks sleeve connection to ground for reliability- I've seen meltdowns on stage from malfunctioning pedals where the ungrounded jack got loose and would cut out randomly during songs, then start working again; the ground connection was flaky because the chassis connection was flaky.  Sure, could use insulated jacks, they're kind of cool, and the subtle theory of grounding in pro audio stuff is cool. Insulated input is nice for the reasons above. But I couldn't imagine the tiny distances and connection to the chassis being a source of noise in practice.  It's more like something you worry about with studio wiring and connecting AC powered things from different rooms where you get buzz and ground loops.

[zbt]

Have good answer from Gord T., pipthepilot, and others here

https://www.diystompboxes.com/smfforum/index.php?topic=128552.0

As suggested by Iain for pop resistor
and Ben Milner for sewer, carry forward to output.




Any correction?

Or should all ground connect to Return Sleeve, for one box only.

[iainpunk]

Any correction?

Or should all ground connect to Return Sleeve, for one box only.

mainly one thing, the dotted line, i would place a connection there, but with a switch, so you can ground or lift if you need to.

the rest looks great!

cheers

[zbt]

also add mute when input is unplug



Need some time to test it