Cavity shielding and pickups

[frets]

Hi Guys😸
I hope everyone is well.

I wanted some input. I was watching a video on cavity shielding and the moderator said it is a waste of time because the pickup’s are outside of the shielding. He was proposing that there still are stray signals.

I can tell a definitive difference when I shield.

Is this guy correct with regard to his suggestion that shielding is a worthless endeavor?

[reTrEaD]

Oct 1, 2023 13:10:12 GMT -5 frets said:

Is this guy correct with regard to his suggestion that shielding is a worthless endeavor?

That's a reckless statement. For something to be worthless it must have no effect at all. If it has a tiny effect, it might not be worth the bother, but it's not worthless.

So let's use an analogy and see if we can get a handle on how much effect shielding might have. We'll start with a water hose and you. If I point the hose at your backside, you're gonna get wet. Now let's use something to block the water from hitting you. Shielding never blocks all the hum and noise, even if it completely surrounds the pickup. So we might use a window screen to reduce the amount of water that gets to you, rather than a wall.

If you place a screen between my hose and your backside, do you get less wet? Definitely. Is that the end of the story, nope. Hum and noise can find their way to a pickup in all three axes, but for simplicity's sake we'll only consider two.

Now three of my friends have joined in with their hoses, so you're really getting wet. From the front, from the left, from the right, and just a little from the back. If you remove the screen protection your backside, do you get even wetter? Yeah, just a bit. We've already detrermined that single screen does something. But we're not done yet.

How about if we use three screens, leaving your front unprotected, do the screens protecting you on those three sides cause you to get less wet, even though you are unprotected in the front. Yep. And I think this part of the analogy is fairly similar to a pickup that is mostly within a cavity. For a dogear P-90 that's surface mounted, it's more like the single screen protecting your backside.

Oh, and the wiring itself, especially in a Fender that just uses two unshielded conductors rather than coax, is completely enveloped in the shielding. So there's some benefit gained in that, too.

I guess we're done here, except ...

SPOILER: Click to show

[newey]

This thread started as a long-ago discussion of another topic, but we do have some empirical evidence from old member D2o here: guitarnuts2.proboards.com/thread/3375/benefit-get-shielding

[JohnH]

Hi frets, on the few that I built, I found the same as you, in that the shielding helps, in most cases.

The clearest example that I had was on a fairly nutzy Strat build, I had the guitar plugged in and amp running as I closed the pickguard down over the cavity. This was an Ibanez Roadstar II , with 3x single-coils, where the jack was on the pickguard. So the PG had foil glued behind it, all the pots were grounded and in contact with the foil, all grounded to the jack as so to the amp . In that state, before closing it up, there was expected single-coil hum at 50hz (our mains frequency), plus mains-related buzz due to fluoro lights, computers etc nearby.

My cavity shielding was also foil, but not directly grounded. It was to get its grounding by contact of overlapping edges when the PG was screwed down,

So as the completed PG was offered over the guitar and lowered into position, there was hum & buzz, hum & buzz, then a crackle at contact and then much reduced buzz, with a little hum.

So I reckon good shielding helps with the more annoying buzz component of mains-related interference. Fender seems to agree, and my 2010 Am Special Strat came with foil on the PG and a conductive paint cavity shielding, with its own grounded lug.

I think there's not much gets induced just into the top of the pickups, but that that there is can be reduced by following our mantra of not hanging from hot.

What also helps a lot with the buzz part while playing, is grounding the strings via the bridge, hence grounding the player. All guitars seem to have less noise when the player is fretting at least one string.

But there's one case where I reckon shielding is not needed, and that's on a classically-wired LP or similar. With these, all wire runs are within grounded, braided screens, pickup base-plates and covers are grounded too. Even an old-style capacitor, of the rolled-up foil kind, can have the outer foil on the ground side so protecting it. There's only a few very short solder connections that aren't covered by this arrangement. But all that grounded braiding around wires must add quite a few pF's of capacitance, and that may be part of the classic LP tone.

[frets]

Thanks guys.

[stevewf]

Oct 2, 2023 5:38:51 GMT -5 newey said:

This thread started as a long-ago discussion of another topic, but we do have some empirical evidence from old member D2o here: guitarnuts2.proboards.com/thread/3375/benefit-get-shielding

In that thread, @d2o tested, among other things, a "modified wiring" and found that its recording was less noisy than urthman wiring. While there were uncontrolled factors, it seems that at least the modified wiring was not more noisy. Looking at the wiring, the difference that I noticed is that with urthman wiring, all unselected coils are completely removed from the circuit (no hanging, no shunting) whereas the modified wiring leaves coils hanging from ground.

So does D2o's experiment provide evidence that coils hanging from ground do not increase noise?

[sumgai]

Two problems there, Steve.

The first is that one can't hang from upside-down*, so no, a coil attached to ground can't "generate" a noise, audible or otherwise.

And that brings us to number two, which is to say that ground is another term for common, which then says that we are relating some point in the circuit to that common point. In turn, this means that we literally can't differentiate common from common, i.e. ground from ground. In the most simple terms imaginable, any potential noise generated in a 'grounded' pickup that has no other connection is simply going to ground straight-away, and thus it becomes a non-factor.

And to be sure, there is a way that all of my above ramblings become null and void, and that is if the ground connection itself is bogus in the first place. Bad soldering job, bad mechanical connection, unseen loose screw (or nut), a number of things can contribute to the condition of a bad ground, which then gets blamed as a faulty circuit design, and Holy Moly, all hell lets out for breakfast.

HTH



sumgai

* Unless you're in Australia...

[newey]

Oct 4, 2023 15:29:38 GMT -5 stevewf said:

So does D2o's experiment provide evidence that coils hanging from ground do not increase noise?

sg is right, a coil hanging on the "ground" side won't generate a signal in and of itself. But if other pickups are operating, does the hanging-from-ground coil cause noise to be "picked up" by those other coils? That's the nub of the thing.

I don't think it does, and D2o's experiment does provide some evidence that it doesn't, perhaps. But we're not at the level of experimental validation of anything here, unless someone has a hypothesis of the electrical conditions or states of the circuit which would somehow cause excess noise via a hanging-from-ground pickup.

Hell, if we consider the years of debate over this on this site and elsewhere, no one to my knowledge has ever definitively proven that a hanging-from-hot condition causes extra noise. On that, however, we do have a hypothetical mechanism for how that might occur- and so we try to avoid it if we can do so.

There's a science project for some kid in there somewhere- prove whether a hanging from hot coil causes measurable/audible increased noise.

[ashcatlt]

Oct 4, 2023 21:14:47 GMT -5 newey said:

…no one to my knowledge has ever definitively proven that a hanging-from-hot condition causes extra noise…

Have we not though? Have none of us ever plugged in a new mod and heard nothing but buzz just to find that we missed a ground connection? An unterminated guitar cable makes a noticeable amount of noise, and that’s far less effective an antenna than a big coil of wire. And then if the coil in question has a grounded metal cover but only two wires, touching the cover while hanging from hot would be about like touching the end of a cable (perhaps attenuated a bit, but at least enough to make you go wtf).

Now, relative winding direction should matter in these cases in the same way and for the same reasons as in normal connections, so that leaving one coil of a humbucking pair hanging from hot should be less troublesome than if the were both pushing RFI in the same direction.

Worth noting, too, that any connection in the middle of a series structure is basically a hot connection, and any coil hanging from that middle series connection is hanging from hot no matter which of its wires is making that connection.

And that kind of brings us to hanging from ground. If the pickups and amplifier disagree on where ground actually is - there’s significant resistance between the what the coils call ground and the amp’s signal ground - then hanging from ground matters because it’s essentially hanging from the middle of a series structure, which… You’d kind of have to mess up pretty bad to make that happen inside the guitar, though I guess trying to make signal ground connection via the back of a pot would be a good way to start.

[newey]

Oct 5, 2023 9:51:29 GMT -5 ashcatlt said:

Have we not though?

The examples you cite are instructive (and they all add up to good reasons to avoid a hanging-hot coil), but I was focused on definitive here. Noise from an unplugged guitar cable may be analogous to a hanging coil, but the noise so generated is not coming from a pickup coil. As you note, we would expect the hanging pickup coil to be a worse scenario, but the example is not testing a hanging coil itself, it's a cable plugged into an amp. As for the bad ground connection, the excess noise could be from the hanging coil, but we haven't necessarily eliminated other variables. Sure, likely that it's from the coil as that's the obvious source- but other things could be on the hot side of that circuit besides just the coil.

So, I don't see those examples as definitive. I was thinking of A/B testing- same guiatr, amp etc., controlling for as many variables as we can, with a coil hanging and then with the coil disconnected entirely.

And that kind of brings us to hanging from ground. If the pickups and amplifier disagree on where ground actually is - there’s significant resistance between the what the coils call ground and the amp’s signal ground - then hanging from ground matters because it’s essentially hanging from the middle of a series structure, which… You’d kind of have to mess up pretty bad to make that happen inside the guitar, though I guess trying to make signal ground connection via the back of a pot would be a good way to start.

This is interesting, and I hadn't heard that expressed before. So, if your reasoning is valid (and it sounds good to me, but I'm probably not savvy enough electrically to give this the proverbial Gold Star), then this would supply the missing hypothesis, the lack of which I was lamenting above. If that is indeed a potential mechanism for noise from a hanging-from-ground coil, then some testing of that might also be in order. But it sounds like designing a test for it would be hard because you're essentially saying different amplifiers might display differences. (or so I understood you, anyway).

[ashcatlt]

Oct 5, 2023 12:10:44 GMT -5 newey said:

…you're essentially saying different amplifiers might display differences. (or so I understood you, anyway).

Well, different systems could be different, but what I was kind of saying is that it’s only an issue if something else is broken. I think sumgai said that above.