Shielded i/o wire Q

[mdc]

Hi folks,

When using RG147 coaxial as for I/O wiring, I've seen it recommended quite often to only connect one end of the shielding to ground, but not the other.

I'm curious as to the theory/science behind this—can anyone explain the reasoning?

So in practice, say I've got a mono input jack, a PCB with a ground pad for the input, and then a ground connection via the footswitch...
I connect the RG147 to the + lug on the jack (with the shielding tied off); then run it to the normal spot on the 3PDT, with the shielding then going to ground on the footswitch.

Do I then run a separate ground wire from the - lug on the input jack to the appropriate ground pad on the board? Or would I leave the input ungrounded completely at the jack?

I've also seen some back and forth about whether the output jack should be connected to the shielding on one or both ends, with no clear conclusion.

Anyway, if someone has a handy MS Paint drawing and some reasonable sounding science to pass along I'd be very grateful.

 

[Big Monk]

Hi folks,

When using RG147 coaxial as for I/O wiring, I've seen it recommended quite often to only connect one end of the shielding to ground, but not the other.

I'm curious as to the theory/science behind this—can anyone explain the reasoning?

So in practice, say I've got a mono input jack, a PCB with a ground pad for the input, and then a ground connection via the footswitch...
I connect the RG147 to the + lug on the jack (with the shielding tied off); then run it to the normal spot on the 3PDT, with the shielding then going to ground on the footswitch.

Do I then run a separate ground wire from the - lug on the input jack to the appropriate ground pad on the board? Or would I leave the input ungrounded completely at the jack?

I've also seen some back and forth about whether the output jack should be connected to the shielding on one or both ends, with no clear conclusion.

Anyway, if someone has a handy MS Paint drawing and some reasonable sounding science to pass along I'd be very grateful.

I believe the thought has always been that in a non-ideal grounding situation, grounding one end of the shielding was a way to prevent ground loops/not add to hum issues.

Personally, I swear by shielded wire in Fuzz pedals. My Universal Fuzz Board has extra pads right near the I/O pads for grounding the shielding.

A better scenario than what you described is to run Input and Shield to the jacks, with the main conductor for the jack run and board run only connecting at the switch, and then grounding the shielded side at the board.

 

[Big Monk]

 

[Feral Feline]

Many thanks for the diagram, Big Monk, I was gettin' confused with all the jacks and runs yet there was no mention of signal — the diagram made everything crystal clear.

Or would I leave the input ungrounded completely at the jack?

You'd need a plastic jack to do that as the enclosure itself is usually connected to ground ... No?

 

[Big Monk]

Many thanks for the diagram, Big Monk, I was gettin' confused with all the jacks and runs yet there was no mention of signal — the diagram made everything crystal clear.


You'd need a plastic jack to do that as the enclosure itself is usually connected to ground ... No?

What I do is use the input jack as the chassis ground, use an isolated output jack, and run all grounds to the board.