13
\$\begingroup\$

I know it sounds like a newbie question but I can't wrap my mind around it. An electromagnetic field is electric + magnetic field.

So this means that when shielding an equipment offensively, for example from avoiding causing interference with other electronics, we need to shield the electromagnetic waves, that means both electric and magnetic shielding.

So if we put say a radio inside an aluminium box, aluminium is pretty much the most cost effective material you can find. Some may use copper but aluminum is more cost effective.

Now an aluminum box will shield the electrical field very efficiently given if the box has no holes or seams, or if the cables coming out of holes are properly shielded and grounded.

But what about the magnetic field?

Aluminum has very low permeability. So how can the aluminum box shield nearby equipment from the magnetic field of the radio inside it? It does shield the electrical field, but not the magnetic?

Can somebody explain to me how shielding works with electrical/magnetic waves? Because I can't wrap my head around it how can it shield the electrical part but not the magnetic?

Does magnetic field leakage pose any noise danger to the nearby equipment from this theory perspective?

\$\endgroup\$
  • 3
    \$\begingroup\$ In physics experiments "mu-metal" is often used to (partially) shield magnetic fields. \$\endgroup\$ – nibot Mar 12 '17 at 23:00
15
\$\begingroup\$

You would not be alone in this one. This is an often misunderstood phenomenon.

Static magnetic fields can not be shielded. They can be re-directed using ferrous materials but even those will not block them.

Electric fields on the other hand can be. Since an electric field is basically a voltage in space, they can not pass through a conductive plate that is held at a fixed potential. Space is shorted out as it were.

Alternating magnetic fields of sufficient frequency however, will not pass through a metal plate. The alternating field generates an eddy current in the plate which generates a cancelling magnetic field.

This is all explained in much better detail here.. Wikipedia

\$\endgroup\$
  • 1
    \$\begingroup\$ so if the shield blocks an electric field of X intensity, it should block an alternating magnetic field of similar intensity if it's frequency is sufficiently high? \$\endgroup\$ – user138887 Mar 13 '17 at 0:01
  • 2
    \$\begingroup\$ That is correct.. You can also think of it another way.. An alternating magnetic field NEEDS to have an accompanying alternating electric field. Since it cant have that because space is shorted.. magnetism cant pass either. \$\endgroup\$ – Trevor_G Mar 13 '17 at 2:00
  • 1
    \$\begingroup\$ In near-field ( < 1/2 wavelength?), are not the H and E fields independent? \$\endgroup\$ – analogsystemsrf Apr 12 '17 at 4:51
  • 5
    \$\begingroup\$ To be pedantic, static magnetic fields can be shielded... with liberal application of superconductors. The Meißner effect! \$\endgroup\$ – Hearth Apr 30 '17 at 20:41
2
\$\begingroup\$

In a box, the distance from circuit-to-shield may not be adequate to develop an electro-magnetic-wave. In that case, you can validly consider the Efield separate from the Hfield.

The sea-of-mobile-electrons in metal is very effective for Efield shielding; the electrons roam to where needed on the metal's surface, to oppose the incoming Efield flux lines, coercing that flux to only impinge on the shield metal at exactly 90 degrees.

The ratio of Magnetic Permeability to Electric Permittivity hints at dramatically different effects between Hfield and Efield shielding.

Magnetic shielding varies with frequency. Standard 1 ounce/foot^2 copper foil of 35 micron thickness gives some attenuation (a few dB) at 5MHz. At 50 MHz, that same 35 micron provides sqrt(10) * dB/Neper, or 3.14 * 8.9dB = 28dB attenuation. At 500 MHz, that 35 micron provides 10.0 * dB/Nepers, or 89dB attenuation.

To begin to shield against 60Hz, you need sqrt(5,000,000/60) ~~ sqrt(100,000) = 316X more thickness; thus 35micron * 316, about 10,000 microns, or about 1 cm.

For magnetic fields, aluminum and copper have nearly the same behavior. Mu is same for both; differences appear from their different conductivity. Aluminum instantly tarnishes, so you cannot solder to it. Copper is easily soldered, using a big hot iron.

Regarding your question about noise-danger to nearby equipment, the answer is YES. Signals can interfere with each other. Check out my answer to "Distance between SPI traces....." question.


{edit} High voltage Efields cause lots of charge movement. If the frequency is low, you'll get detectable EXTERNAL movement of charges due to the Efield. In other words, the SkinEffect is your friend but SkinEffect only predicts attenuation; SkinEffect does not prevent external charge movement.

\$\endgroup\$
0
\$\begingroup\$

I do not know the theory that well but I can tell you what I saw being practiced at Qualcomm when I used to work there about 15 years back. So while conducting tests on the phones/chips (such as reference sensitivity tests) we placed the phone in a metal box about 50cm x 35cm x 20cm. From the color of the box it seemed more like copper than aluminium but I guess you can put on artificial colors. There was a wire that carried the signal to and from the outside world. For more sensitive testing the phone along with other test equipment was placed inside a metallic cage, the size of a small room. There were all kind of other precautions that we took so as not to influence the test results. Just to clarify the signals the phones carried were GSM/GPRS/WCDMA signals in the range from about 900MHz to a few GHz.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.