Shielding; how necessary is it?

[timtam]

It is also possible to over shield. By providing multiple paths to ground from the same point you can introduce ground loops, more of an issue with active electronics but it’s worth making sure that grounded components only have a single path to ground either through your shielding or through a wire, not both.

There is no good evidence that any physical ground loops in guitars are noise-producing. Either by different earth potentials at different points in the loop (obligatory for current to flow to produce noise) or induction. It's easy to test say in a Les Paul - link all four pot body grounds in a loop, listen to the noise floor, then cut one loop and see if it changes the noise floor. It's annoying when supposedly reputable sources like Fralin just say on their website 'don't create ground loops' yet refuse to prove that it matters when challenged.

In jaguars, Fender seems to toss a coin on whether they use pot-body to pot-body wired grounds for the controls on the metal control plate or not - which means some guitars have physical ground loops and some don't. For example the Am Pro jag has wired pot body grounds on the control plate (top below); the Vintera 60s does not (bottom below). The plate of course has the output jack socket, the grounded rim of which grounds the plate when the (grounded) jack is inserted. Thus the pot bodies on the plate are also grounded. So any connecting wire between the pot bodies is essentially redundant. Where Fender does add a ground wire between pot bodies they have effectively created a physical ground loop (potbody1->wire->potbody2->plate->potbody1). But it doesn't matter.

American Pro jag - pot bodies on plate wired together ...


Vintera 60s jag - pot bodies on plate not wired together ...

[adamrobertt]

Tl;Dr - it doesn’t matter that much but it doesn’t hurt anything either. Do what you want and it’ll be fine.

[alexpigment]

Sorry to dig up this thread after a few weeks, but after some recent experiments where I confirmed my long-standing hunch, I wanted to chime in. Shielding basically just gets rid of the hum that occurs when you take your hands off the bridge/strings. When properly shielded, you should have no significant difference in noise floor when your hands are on/off the guitar.

If you have a lot of noise with your hands *on* the strings, it's almost certainly coming in from the pickups (barring an improper ground). It can frequently be loud enough that you don't notice the hands-off buzz, but if you implement other hum cancelling measures (humbuckers, noiseless pickups, dummy coils, Ilitch/Suhr backplates, etc), the difference between your hands on/off will be easy to hear.

For context, I got a new Squier Offset Tele recently, and quickly put a dummy coil in to get rid of the noise that occurs for all single coils in my house. I then noticed the hands on/off thing that I hadn't dealt with in quite a while. Most of my guitars have shielding paint in the cavities, so it's been a non-issue for a long time. I then shielded the guitar, and while the general noise floor with/without the dummy coil did not change, there is now no discernable difference in whether I have my hands on or off the strings.

So to answer the OP's question, shielding is not absolutely critical while playing, though it's nice not have that noise given that adequate shielding is pretty easy and cheap to accomplish. Also, while you'll hear many people say on the internet that the hum that occurs when you take your hands off the strings is "normal", that's a relative term. It's normal for vintage instruments or for certain lines of guitars; it's not normal for instruments that are shielded properly from the factory (which happens to be most inexpensive guitars).

[JackFawkes]

That matches my experiences/tests/observations as well.
While I personally think shielding is a good thing, for the same reasons you listed, I personally don't bother shielding unless the instrument is going to use coated strings...

Jack