30

I'm shielding the pickup cavity of a guitar to minimize interference. I'm doing this with copper tape that has conductive adhesive, and I'm connecting this tape to ground.

Without getting into too much detail, let's say there's a small spot I miss or purposely leave uncovered because it's difficult to reach, so that the Faraday cage is not completely "sealed" all the way around. Would this drastically reduce the effectiveness of the cage, or would it simply reduce it proportionally to the size of the hole?

I ask because I'm fine if it's just slightly less effective, but if it ruins the whole thing, then I'll put in the extra effort.

Glorfindel
  • 1,245
  • 3
  • 15
  • 20
Anthony
  • 541
  • 5
  • 10
  • 1
    I have to wonder, knowing little of music, just how sensitive is this pickup to interference anyway? Or is it the other way around, that it produces interference? – Hearth May 03 '19 at 03:00
  • 3
    @Hearth My understanding is that it depends on the type of pickup. Some pickups (single coils) do nothing to reduce noise. I've heard stories of people picking up radio stations with their guitar. – Anthony May 03 '19 at 03:03
  • 3
    @Toor If you have an answer you need to write it in the _Answers_ section. Your current remark is just a random opinion and can't be vetted by the community or accepted as correct, but will still be the first that everyone will see when the come here. – pipe May 03 '19 at 13:26
  • Look at your microwave. It's a Faraday cage (so you can stand in front of it while it's running without getting cooked), but you can see through the door... – Vikki May 04 '19 at 21:03
  • @Sean: The shielding is also present on the door. – Ben Voigt May 04 '19 at 23:23
  • @BenVoigt: ...but the door shielding has a grid of holes in it which allows you to see through said door, and, nevertheless, prevents the microwaves from getting out and cooking you. – Vikki May 05 '19 at 00:29
  • 2
    @Sean: Absolutely, but while "there are holes in the shielding" is literally true, it's misleading because it implies the shielding is compromised. It would be more accurate to say that the holes are part of the shielding. – Ben Voigt May 05 '19 at 02:58
  • 1
    I see a lot of noise about Faraday cages in right wing survivalist literature. But they never talk about TESTING a Faraday cage. My crude test is to put a cell phone inside and call it. Then put a GMRS radio inside and call it (that is a much stronger signal). I did this with two microwave ovens (unplugged, of course), and they both failed the cell phone test. – richard1941 May 12 '19 at 01:16
  • 1
    It may be that there are better solutions than Fadaray cages for eliminating your interference. Consider bypassing and use of a balanced line with shielded twisted cable. – richard1941 May 12 '19 at 01:17

5 Answers5

34

It's fairly common to make Faraday cages out of mesh rather than sheet copper, so you can imagine that a single small round hole is not going to degrade the effectiveness enormously. But the holes in the mesh must be much smaller than the wavelength you're trying to screen.

In particular, it's the largest dimension of the hole, not its area, that matters. A 1-mm round hole will allow much less leakage than a seam 10 mm long but only 1 um wide.

The Photon
  • 126,425
  • 3
  • 159
  • 304
  • 1
    And considering that we are most concerned about audible interference, we really care about 20-20.000 Hz, such as the 50/60 Hz buzz from mains power. That means a wavelength of at least 15 meters. A "small spot on a guitar cavity" will not matter. – MSalters May 03 '19 at 12:13
  • 2
    @MSalters That makes it a bit weird because it indicates that you can just make a small loop with one wire around it and everything is magically "shielded": the two holes on either side are much smaller than the wavelength. – pipe May 03 '19 at 13:29
  • 1
    @pipe: I don't get what you mean. A faraday cage is a 3D enclosure, with an interior and an exterior. The point of a Faraday cage is that the two are separated, and it works both ways (incoming and outgoing radiation). I don't get what 3D shape you are trying to describe, and what its interior would be. – MSalters May 03 '19 at 13:33
  • 3
    @MSalters, you may also be worried about higher frequency RF signals carrying audio modulation, or combinations of RF signals with beat frequencies in the audio range. – The Photon May 03 '19 at 13:43
  • 2
    @MSalters Enclose the whole device in a perfect sphere, lets's say 20 cm in diameter. Now remove half of that sphere. The hole from what you removed is much smaller than the wavelength so according to your theory the sphere should still shield the device. – pipe May 03 '19 at 14:12
  • 3
    @MSalters In what universe does a 60Hz signal have a wavelength of 15 meters? – Dmitry Grigoryev May 03 '19 at 14:22
  • @pipe: So basically a half a sphere. Topologically a flat disc, a satellite dish, this half-sphere and a full sphere except for a pinhole are all the same shape (flat) and none have interiors. But we don't mean a topological interior here. The current around the edge of the hole generates an EM field with the opposite direction, effectively canceling the incoming wave at this edge. But the half of the sphere you cut out is no longer there, and the EM field from the edge can't cancel the incoming wave there. – MSalters May 03 '19 at 14:23
  • 4
    @DmitryGrigoryev: _at_ least 15 meters, but I can indeed tighten that limit to at least 15 _kilometers_ (for 20 khz waves). Yeah, that's basically the realm where we're better of treating it as electro-static. Ground that cage! – MSalters May 03 '19 at 14:26
  • 4
    @MSalters, also if you're trying to shield something from frequencies below about 50 kHz, remember the shield thickness needs to be more than a couple of skin depths, which might not be practical for such low frequencies. – The Photon May 03 '19 at 15:47
  • 3
    @pipe, remember that microstrip lines don't radiate much (at appropriate frequencies), and they're only shielded on one side. And that radiation is reciprocal (an antenna works equally well receiving as transmitting). So yes, a "shield" that only surrounds one side of a structure could (within limits) reduce the radiation from that structure. – The Photon May 03 '19 at 15:51
  • 3
    @DmitryGrigoryev: One where light travels extremely slowly? – Vikki May 04 '19 at 21:01
15

Regarding the actual use case in question:

Having built musical instruments and experimented with this tape in the past, I can say it's a complete waste of time if you're using humbuckers, since those pickups are designed to cancel hum anyway, and almost a complete waste of time if you've got single-coil pickups, since most pickups nowadays come with shielded leads. If you do a good job grounding the pots, bridge, jack, etc., you'll be fine with no extra shielding. (How's your soldering?)

If you absolutely must shield the electronics cavity, use conductive paint, since you can paint it into every nook and cranny. With paint, there are no tape overlap regions that may or may not be in good electrical contact, and you don't have to worry about the tape adhesive losing its grip when you leave your instrument in a case in a hot car, causing the tape to fall off the cavity wall and short out your wiring (unbeknownst to you!).

If you're using vintage pickups with unshielded leads, you may consider just sleeving your leads between the pickup and the electronics cavity in a tube of conductive tape (just make this on your own from a sufficiently long piece of tape), and grounding that tape to the pot body to mimic shielded modern pickup leads.

Bort
  • 1,084
  • 9
  • 18
9

The effectiveness of the shield (with and without holes) will depend on the frequencies you're concerned about, since the maximum size of holes in Faraday cage is supposed to be 1/10 of the wavelength or less. Reality check: a domestic microwave operates at 2.4 GHz (12.2 cm wavelength) and has a shielded window with holes of 5 mm or less.

If we're talking about audio frequencies, your biggest concern will be the skin depth of copper which is about 8mm at 60 Hz, so a copper tape (which is often 35μm) thick is essentially transparent to such waves.

At 1 MHz the skin depth will be about 60μm so several layers of copper tape may have an effect. A wavelength at that frequency is still around 300m, so small holes will not matter. Note that if you're in an environment where an object less than 1 meter in size picks up significant audio interference at 1 MHz, nearby objects about a quarter-wave length (75m) should noticeably resonate (as in, long metal cables would "sing" loud enough for you to hear).

At 100 MHz the copper foil is really effective (with skin depth of only 6μm). The wavelengh is around 3m, so holes of reasonable size will not be of your concern.

Only if you're expecting radiation in GHz range interfere with your guitar, holes in your shield can become problematic.

Dmitry Grigoryev
  • 25,576
  • 5
  • 45
  • 106
  • 2
    Hm, sounds like I now need to ask a question about all this skin depth of copper and how that affects blocking.. – pipe May 03 '19 at 15:19
5

As long as the size of the hole is less than one wavelength of the frequency you are concerned about, AND your circuit is at least a wavelength inside the shield, you will be fine (6.28 nepers of attenuation will be yours).

analogsystemsrf
  • 33,703
  • 2
  • 18
  • 46
  • 2
    Just less than 1 wavelength? A half-wavelength slot antenna radiates pretty well, doesn't it? – The Photon May 03 '19 at 03:29
  • 2
    The rule of thumb more commonly used is that the largest dimension of the hole must be less than 1/10 the wavelength of the radiation you're concerned about, and depending on application even that can be insufficient – llama May 03 '19 at 23:54
  • 1
    Notice I included "And your circuit is at least a wavelength inside the shield". This is from Feynman Lectures series. – analogsystemsrf May 04 '19 at 10:02
1

Faraday cages aren't entirely closed boxes. Like cages, they have gaps. The size of the gap mainly influences what wavelengths will be able to penetrate, and not so much the amount.

For example: a microwave oven always has a mesh in front of the window. The holes in it are small enough to prevent the microwaves from escaping, but large enough for light to pass through. Another example: your car can be considered a Faraday cage in the event of lightning. It will protect you from the strikes, because the wavelength is way too large. However... because of the huge gaps in the cage (glass windows) we can still receive cell phone signals through it.

I'm not sure what kind of signals you're trying to block, but since we're talking about audio I'm guessing fairly low frequencies (large wavelengths). So as long as the gap isn't too big, I don't think it'll be too much of an issue.

Opifex
  • 505
  • 3
  • 11