Holes in a Faraday Cage

Thread Starter

BackyardBrains

Joined Aug 26, 2010
33
I have been reading that faraday cages need to be built with holes that are sufficiently smaller than the wavelength of these different emissions. We've noticed something strange, and need some help understanding what is going on.

The frequency is related to wavelength by the following equation:
f = c/γ
Where f = frequency in Hz, c is light velocity, and γ represents wavelength in meters.

One rule of thumb often used for Faraday cages to prevent transmission is that the holes need to be no larger than 1/10 of the wavelength of the signal.

So for a 3G cell phone that operates at a frequency of 2.1 GHz (2.1 * 109 Hz), the wavelength = (3 * 108 m/sec) / (2.1 * 109 Hz) = .14 meters. Thus, for a Faraday cage to prevent this noise from entering, the holes in the cage should be smaller than .014 meters (or 1.4 cm).

Also, 60Hz line noise will have a wavelength of (3 * 108 m/sec) / 60 Hz = 5.4 meters or ~17.5 feet. This would mean that 1 foot holes would block 60Hz EM waves, but that is not the case when I build a faraday cage.

When we surround our electronics with a metal screen mesh from Home Depot, as seen below:



This does a wonderful job of blocking 60 Hz noise, but does not effect the 3G cell phones signal when I place it inside the cage! That seems to be the opposite of what that equation is stating. What is my confusion?

I should note that I ground our electronics to the Faraday cage, but I'm not sure where to ground the cell phone to the cage. Is this what is making the difference? I am wondering why?

Thanks!
 

t_n_k

Joined Mar 6, 2009
5,455
Problems can arise with pathways along any join line slots in the structure.
If I understand your comments the enclosure works fine @ 60 Hz but not at 2.1GHz.
BTW the 60 Hz wavelength is 5000km.
 

praondevou

Joined Jul 9, 2011
2,942
1. cell phones are very sensitive (RF receiver)
2. Is there no leakage possibel? No holes bigger than 1/10 of the supposed 2.1GHz wavelength?
3. Are you sure that the phone doesn't switch to another (lower) frequency for GSM?
4. What about the ground connection? What happens if you disconnect the ground?
5. Is the cellphone powered via an power adapter or suspended and not connected via any metallic material to the cage?
 

KJ6EAD

Joined Apr 30, 2011
1,569
The conductivity of the aluminum may not be high enough. Try bonding it to ground at multiple points or weave in some copper wire in a larger mesh pattern to equalize the potential differences across the aluminum mesh.
 
Last edited:

crutschow

Joined Mar 14, 2008
25,125
.................................

This does a wonderful job of blocking 60 Hz noise, but does not effect the 3G cell phones signal when I place it inside the cage! That seems to be the opposite of what that equation is stating. What is my confusion?

..................
Why do you think it's the opposite? The equation is stating that low frequencies are blocked by a given mesh size but high frequencies with a shorter wavelength are not attenuated as much. That's just what you are seeing.

Do you have screen on all six sides of the cube including the bottom?
 

Ron H

Joined Apr 14, 2005
7,014
A little anecdote seems appropriate here.
Back in the mid-90s, I was one of the design engineers on a 128 x 128 digital video routing switcher. It contained 128 PLLs, each with a VCO running at 270MHz. These and all the related circuitry (crosspoint switch ICs, etc.) were contained in a steel cube that was approximately 19" on each side. Of course, it had cooling holes, including fans , and lots of seams sealed with lots of screws. We had paid a lot of attention to EMI containment.
We took it to an EMI test facility, along with 256 RG-6 cables of some specified length (can't remember the length), 256 75Ω terminations, a digital video signal source, and possibly some distribution amplifiers. I can't remember the exact setup, but it was prescribed by the FCC or some such agency.
We set it up, powered it on, and watched the technician run his spectrum analyzer, which was attached to an antenna. His first words were "Planes are gonna start crashing.".
We had rolls of adhesive copper tape, and started taping over suspected leaks. We finally got it within spec. Then we had to go back to work and implement the changes in the sheet metal. It wasn't easy, but we finally made a product that we could ship.
 

Thread Starter

BackyardBrains

Joined Aug 26, 2010
33
A little anecdote seems appropriate here.
Back in the mid-90s, I was one of the design engineers on a 128 x 128 digital video routing switcher. It contained 128 PLLs, each with a VCO running at 270MHz. These and all the related circuitry (crosspoint switch ICs, etc.) were contained in a steel cube that was approximately 19" on each side. Of course, it had cooling holes, including fans , and lots of seams sealed with lots of screws. We had paid a lot of attention to EMI containment.
We took it to an EMI test facility, along with 256 RG-6 cables of some specified length (can't remember the length), 256 75Ω terminations, a digital video signal source, and possibly some distribution amplifiers. I can't remember the exact setup, but it was prescribed by the FCC or some such agency.
We set it up, powered it on, and watched the technician run his spectrum analyzer, which was attached to an antenna. His first words were "Planes are gonna start crashing.".
We had rolls of adhesive copper tape, and started taping over suspected leaks. We finally got it within spec. Then we had to go back to work and implement the changes in the sheet metal. It wasn't easy, but we finally made a product that we could ship.
Thanks for the anecdote. What does "Planes are gonna start crashing." mean? Also, is it safe to assume from this story that trying to analyze faraday cages and EMI from first principles is a fools game?
 

Thread Starter

BackyardBrains

Joined Aug 26, 2010
33
Why do you think it's the opposite? The equation is stating that low frequencies are blocked by a given mesh size but high frequencies with a shorter wavelength are not attenuated as much. That's just what you are seeing.
Yes. You are correct. After reading what I wrote, I'm not sure why I came up with the idea that it was the opposite.

Do you have screen on all six sides of the cube including the bottom?
I have tried wrapping my cellphone in that same mesh. All 6 sides. Still rang when I called it.
 

Ron H

Joined Apr 14, 2005
7,014
Thanks for the anecdote. What does "Planes are gonna start crashing." mean?
The technician was facetiously implying that our box was emitting so much EMI that it would interfere with aircraft communications.
Also, is it safe to assume from this story that trying to analyze faraday cages and EMI from first principles is a fools game?
No, I don't think so. We were engineers, but none of us were experienced in handling EMI. We understood the principles, but,as always, the devil is in the details. We eventually got it right.
 

Thread Starter

BackyardBrains

Joined Aug 26, 2010
33
Sometimes you have to go to considerable effort to properly RF seal an electronic equipment enclosure.

http://www.metexcorp.com/emirfi_theory.htm

The same rules would presumably apply to a good Faraday Cage design.
Thanks. Will read tonight. Are the principles of a Faraday cage the same as those of EMI shielding? I was just curious about the "holes" in a faraday cage.

Asked another way. If I were to make my Faraday Cage out of metal (no holes) in the exact same configuration, do you think it would still not work? Meaning that it's not the "holes", but some other EM shielding principle that is allowing EM signals getting through...
 

Thread Starter

BackyardBrains

Joined Aug 26, 2010
33
The technician was facetiously implying that our box was emitting so much EMI that it would interfere with aircraft communications.
Ohh :) Those kind of planes. lol. I was thinking it was a 2-d field! Thanks for your comments. I am learning that stuff is more of an art then a science.
 

crutschow

Joined Mar 14, 2008
25,125
Thanks. Will read tonight. Are the principles of a Faraday cage the same as those of EMI shielding? I was just curious about the "holes" in a faraday cage.

Asked another way. If I were to make my Faraday Cage out of metal (no holes) in the exact same configuration, do you think it would still not work? Meaning that it's not the "holes", but some other EM shielding principle that is allowing EM signals getting through...
If the box is of a good conducting metal and reasonably thick with no holes or seams, then it will make a very good EMI shield.

When we sealed military electronics boxes for EMI we used conductive gaskets between the shielded connectors and the housing to minimize any EMI leakage in the gap. And the lids were often recessed in a groove so that the gap would have a right angle between the inside and outside to make it impossible for anything to get straight through the gap, it would have to take a right angle turn which helped the attenuation.
 

#12

Joined Nov 30, 2010
18,210
I posted this in another thread, but here I go again.

I rolled up a cell phone in aluminum foil, 4 layers thick, and folded the ends over by about 3 inches on each end. The cell phone did not ring.

After removing the aluminum foil, I put the cell phone in an aluminum project box and it still rang. I put that aluminum box in another aluminum box and that was enough to stop it ringing. None of these solutions had a ground attached.

The moral of the story is that RF sneaks in through any seam, even if it has to bounce. More bounces = more attenuation. Grounding the shield is not imperative.

That's all I have.
 

Mark_T

Joined Feb 7, 2012
47
I think your problem is the conductivity of the mesh, it should have thick conductors between gaps. Look at microwave oven and its lots of round holes and very thick conductor. At these frequencies current is carried only in the skin of the mesh. bulk round wires become resistive. Flat wires are required, like copper ribbon.
 

crutschow

Joined Mar 14, 2008
25,125
A true Faraday cage requires Mu metal, not aluminum.
Perhaps true if you are trying to provide the best shielding for low frequency magnetic fields such as from 60Hz power. But a thick piece of aluminum will still provide reasonable magnetic shielding due to generated eddy currents that counter the applied field. Mu metal has limited added effectiveness for higher frequency EM fields.
 

dataman19

Joined Dec 26, 2009
135
Faraday cages are of several styles:
Passive and active being the most common.\
** Passive:
Shielded with more than one layer of mesh - "Which is grounded at more than one point".
Doorways also have to have EMI Shielding Gaskets, that electrically seal the doors when closed. The mesh in the door gaskets effectively become part of the solid shielding.
..
The fact that your cell phone is "ringing" means you have not built a solidly sealed faraday cage. "Yes" a faraday cage is most often made of a MuMetal mesh - but multiple layers of varying screen of almost any metal will give satisfactory results (in other words keep the cell phone from ringing).
..
Power leads have to also be shilded and must pass through ferrite beaded filters. In fact, a majority of all faraday issues are with power cables and doorways.
..
As for the metal project box - "Did you solder all the seams? (you will have to solder all the seams in the faraday screens in your cage as well. Was you cage and or project box grounded using a ground that exhibits no more than 1/10th of an ohm in resistance?
...
** The "active" Farady Cage.
Thes3 are also extremely complicated. Especially since they have a negative voltage applied to one mesh layer, a grounded layer, and a posititive voltage applied to another layer. In all there are at least two active (charged layers) and three grounded layers.
..
What I have always found interesting is that "you have to punch holes in the outer grouded layer and at least the middle grounded layer to get wiring through. So essentially you do a majority of the filtering and use the final inner grounded layer as insurance.
..
But the doors - those damned doors - Ya still have to deal with the doors.
...
By the way, are you referring to the screen mesh as the "holes"? Faraday Shields do not generally have "holes". But if you are only filtering "specific" EMI, then you can design with holes.
...
My best advice is to ground the heck out of everything and seal the door cracks as well. Not sure if you have good grounding, use a megger (a micro ohmmeter). It only takes one little minute crack to literally crack your faraday cage.
...
Dave
 
Top