2

Is it possible to make a solid Faraday cage with a hole for a wire (to transmit information inside)?

What is needed to prevent electromagnetic waves from going through the hole?

nvq
  • 31
  • 1
  • Normally you use a filter and/or feedthrough capacitor. – Kartman Aug 30 '21 at 05:59
  • 1
    You could use fiber-optic cable through a small hole. If you want to use copper wire, you can use shielded cable. There are bulkhead mount RF connectors that could be installed so the shield is continuous with the cage. – user57037 Aug 30 '21 at 06:08
  • nvq, it's just a matter of allowing the conduction band electrons to move around the hole fast enough for whatever you are trying to block. This just means making the hole "small enough." What kind of EM waves do you intend on blocking? – jonk Aug 30 '21 at 06:46
  • @jonk EM waves up to 1mm wavelength – nvq Aug 30 '21 at 08:11
  • @nvq You need a hole smaller than about half of the shortest wavelength you want to exclude, in rough terms. – jonk Aug 30 '21 at 08:24
  • Putting a plain wire through the hole will defeat the faraday cage, EMI will piggyback onto the wire. You need to use a non-conductive form of transmission, or filter the signal with a feedthru filter. – Mattman944 Aug 30 '21 at 10:46
  • @Mattman944 I haven't done the mathematical analysis (lots more code to write I don't have time for), but placing a low pass filter in series with the *'antenna'* where it leaves the shield should be sufficiently useful. Though I don't think one would want any of it connected to the shield itself, just as you would not want any circuitry within the shield connecting to the shield. But I'd be interested in seeing detailed analysis. I may yet spend some time developing the PDEs and some kind of mesh to test all this when I get the inclination some day. – jonk Aug 30 '21 at 18:15

2 Answers2

4

Keith Armstrong over at https://www.emcstandards.co.uk/ wrote together a blogpost on this topic last year, see here: https://www.emcstandards.co.uk/constructing-io-panels-for-shielded-rooms

Citing from the blog post:

There are four ways to get power, data, signals, etc., into a radio-frequency (RF) shielded chamber through an I/O panel without destroying the shielding effectiveness of its external surfaces (i.e. its walls, floor, or ceiling):

  1. Conducted via filters, which must be 360° shield-bonded to the surface of the I/O panel (either on its inside or outside). See Figure 4.
  2. Conducted via shielded cables, which must be 360° shield-bonded to the surface of the I/O panel (either on its inside or outside). See Figure 4.
  3. Radiated via RF in waveguides, which must be 360° shield-bonded to the surface of the I/O panel (either on its inside or outside)
  4. RF-modulated light in free space or guided by optical fibres, passing through a waveguide-below-cutoff, which is a metal tube that must be 360° shield-bonded to the surface of the I/O panel but can be protrude from the panel entirely on its outside, entirely on it inside, or anywhere between those two extremes.

What's important to remember is that a Faraday cage is a theoretical concept that you will never fully achieve, meaning that you will always have leakage and losses. I've built a DIY shielded chamber for EMC testing purposes (i.e. a Faraday cage), and it works ok. You wont have any cell reception when you're inside the chamber, but there is definitely signals that can penetrate the chamber.

When you try to get a signal in or out from a Faraday cage using a wire, you WILL have leakage in and out from the cage. The question is how you minimize it or keep it below an acceptable level.

SamGibson
  • 17,231
  • 5
  • 37
  • 58
Lars Petersson
  • 567
  • 2
  • 12
  • This is really good. But I think you should add one or two highlights from the document and the official title of the document. The reason for this is that "link only" answers become totally useless if the link goes bad, which happens all the time. Adding the title of the document may allow future readers to find it even if this link goes bad (if it is available for download somewhere else). – user57037 Aug 31 '21 at 05:47
  • 1
    Good point, thank you! Added something on my own experience on the matter as well. – Lars Petersson Sep 01 '21 at 07:01
1

A small hole will not affect the functioning of a faraday cage for wavelengths larger than the largest dimension of the hole.

However, a conductive wire from inside the cage, to the outside of the cage which is not electrically bonded to the cage can defeat isolation. It is a pathway for both static electric fields and electromagnetic waves to pass through the cage.

To pass a signal between the inside and outside of the cage without compromising the shielding requires special care. Optical coupling is one simple method of transmitting signals while maintaining isolation. If you know the frequency range of the signals you are trying to block, and your communication signal is outside of that range, you could use passive filters to pass your signal, but block the unwanted noise. (That is essentially what optical coupling does, but with frequencies other than optical).

Math Keeps Me Busy
  • 18,947
  • 3
  • 19
  • 65