4
\$\begingroup\$

What properties of metals make them effective at shielding against EMF, and which metals are typically used?

How does Aluminium, Steel, Copper and Lead compare?

\$\endgroup\$
  • 5
    \$\begingroup\$ Conductivity... \$\endgroup\$ – Lelesquiz Jan 2 '17 at 2:04
  • 1
    \$\begingroup\$ Steel is probably the cheapest and thus used most often. All would work. Sometimes you need some kind of RF gasket or conductive spring finger to make contact across a gap at a corner or something like that. In that case you need to consider surface conductivity. Nickel plated steel may be a good choice in that case. Or even stainless steel. I can't see why you would use lead. \$\endgroup\$ – mkeith Jan 2 '17 at 2:10
  • \$\begingroup\$ @Lelesquiz: Sorry for a newb followup question, but being more conductive than steel, why isnt aluminium favoured then? \$\endgroup\$ – Lennart Rolland Jan 2 '17 at 2:19
  • 2
    \$\begingroup\$ a 0.1mm aluminium sheet is soft as ... aluminium foil, wheras a 0.1mm steel sheet is solid enough to form a cap \$\endgroup\$ – Marcus Müller Jan 2 '17 at 2:29
  • 1
    \$\begingroup\$ I think all common metals are more than conductive enough to form an RF shield. So the decision will be based on other properties besides conductivity. If the shield is a box, then steel is a good choice. If you need a flexible braided shield on a cable, then you will probably choose copper (tinned copper). In some cases, aluminum might be used for a box, but it is not very strong unless the wall thickness is made fairly thick. Also, if aluminum flexes back and forth several times, it is likely to break altogether. Copper is expensive. Lead is heavy and soft and hard to dispose of. \$\endgroup\$ – mkeith Jan 2 '17 at 3:46
1
\$\begingroup\$

Besides conductivity, a common property of metals, most other factors will prevail, like weight, price, mounting accesories, available shapes, etc....

\$\endgroup\$
1
\$\begingroup\$

Copper foil (sometimes adhesive backed) is used as it is readily solderable, a (flexible) PCB substate is also often used as the geometry can be tailored by a standard process.

Aluminium and aluminum coatings on plastics are common if no soldering is needed.

Silver is common in radio frequency cavity filter plating to garner the best benefits of the skin effect at microwave frequencies.

Sheet steel is popular for sub-circuit covers and component cans as it is cheap and readily formable. Shapes are made with photo etching, die stamping and laser cutting.

Flexible elements are often made from woven or knitted meshes and sometimes small scale 'expanded metal' or perforated foil is to be found where reduced weight or freeform shape is required.

Surface coatings can be painted metal powder of zinc or silver. Thin coatings on plastic substrates would be more likely vapour deposited, sputtered or electroplated metals like aluminum, nickel, silver or gold.

All of these metals (except aluminium, nickel, silver and gold) will usually be coated with a less reactive metal, tin usually, for corrosion protection.

Lead and stainless steel are at the bottom of the list due to lower conductivity but given that a slightly thicker sheet could be used they are also an option if nothing else is available. They would more likely be selected due to their corrosion resistance and strength or flexibility.

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

Even chewing gum foil will attenuate the magnetic fields of fast current surges.

The last fast&accurate PCB I designed (12 bits, at 6Million Conversions/second, with 4 channels) was quite a new adventure for me, so I sat down with a pulse generator (trise, tfall approv 20 nanosecond) and a scope, and several 50 Ohm resistors (so I'd not short the generator output, nor have lotta confusing reflections) and a 2 meter coax cable/BNC connectors which I promptly cut in half, with the 50 Ohm resistors soldered onto the newly cut ends.

The resistor leads were long enough to form 1" square loops, and with 2 loops I had high-performance Transmitter and Receiver loops.

Remembering back, with 1volt from the generator into the TX loop, I saw 80 milliVolts out of the RX loop. And, sliding chewing gum foil between, the 80 milliVolts dropped at least 10:1.

That insight ----- thin AL or CU is an excellent Hfield attenuator for high frequencies ----- caused me to spend lots of time PLANNING the component placements for my high-data-rate data-acquisition system, particularly in orienting the 4 FiberOptic transmitters to avoid trashing the ADC.

Result? thermal-noise limited performance. No herringbone patterns in the reconstructed video, meaning the injected deterministic energy was 50 or 60 dB down, at least.

\$\endgroup\$
  • \$\begingroup\$ I just arrived here in my fantastical boat! \$\endgroup\$ – Lennart Rolland Feb 5 '17 at 7:08

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.