• Quick note - the problem with Youtube videos not embedding on the forum appears to have been fixed, thanks to ZiprHead. If you do still see problems let me know.

Do you need to ground a Faraday cage?

BowlOfRed

Master Poster
BowlOfRed
Joined
Jul 15, 2010
Messages
2,502
Location
Silicon Valley
My memory of college physics is that a Faraday cage works by the fact that all parts of the cage are at the same potential, including all the parts surrounding the interior, so there will be no electric field gradient in the interior. But the actual potential of the cage doesn't matter, so it doesn't need to be grounded to be effective.

I recall studying this for static arrangements, not dynamic. Is grounding necessary for excluding EM waves?

A paper was released in Nature about EM interference for birds navigation.
http://www.nature.com/nature/journal/vaop/ncurrent/full/nature13290.html
The author mentioned that the cage was used both grounded and ungrounded to show differing behaviors. I didn't realize that was relevant.

Does grounding help with excluding a particular range frequencies (possibly due to the capacitance of the cage itself)?
 
For static purposes a faraday cage will protect its content from external charges. However, if it is not at ground potential, you may get a nasty surprise when you try to leave.:eek:

For EM shielding, it is different. As you already guessed, there is a capacitive coupling between the cage and anything inside (this is true even if the inside parts are galvanically connected to the cage). Thus, if the cage is not grounded, an AC signal may be coupled into it.

Hans
 
Inside the Faraday cage there is no way to even experimentally determine whether the cage is grounded or not. Assuming an ideal Faraday cage of course.

The transitioning from inside / outside can be interesting. Aircraft grounding straps have a reason.
 
I am flabbergasted that this could appear in Nature!

Their magnetic orientation capabilities reappeared in electrically grounded, aluminium-screened huts, which attenuated electromagnetic noise in the frequency range from 50 kHz to 5 MHz by approximately two orders of magnitude. When the grounding was removed or when broadband electromagnetic noise was deliberately generated inside the screened and grounded huts, the birds again lost their magnetic orientation capabilities.
Click to expand...

Perhaps the summary is misstated. If not I would consider the highlighted statement as prima facie evidence the study is seriously flawed.
 
BowlOfRed said:
My memory of college physics is that a Faraday cage works by the fact that all parts of the cage are at the same potential, including all the parts surrounding the interior, so there will be no electric field gradient in the interior. But the actual potential of the cage doesn't matter, so it doesn't need to be grounded to be effective.

I recall studying this for static arrangements, not dynamic. Is grounding necessary for excluding EM waves?

A paper was released in Nature about EM interference for birds navigation.
http://www.nature.com/nature/journal/vaop/ncurrent/full/nature13290.html
The author mentioned that the cage was used both grounded and ungrounded to show differing behaviors. I didn't realize that was relevant.

Does grounding help with excluding a particular range frequencies (possibly due to the capacitance of the cage itself)?
Click to expand...


Grounding can affect the frequencies that permeate a Faraday cage, both to enhance or suppress, but only as a result of design inadequacies and is a second order effect at best. I consider the result that the grounding of the Faraday cage resulted in the birds regaining navigational abilities the most significant finding of this study. It would warrant extensive study and comment in the paper. I'm not a subscriber but this tempts me to become one.
 
Last edited:
marting said:
I am flabbergasted that this could appear in Nature!

Perhaps the summary is misstated. If not I would consider the highlighted statement as prima facie evidence the study is seriously flawed.
Click to expand...

I assure you it is not misstated. I looked it up because I heard the author talk about the study on a podcast. He stated that because the cage only works when it is grounded, he used that as a blinding mechanism.

He would ask the graduate students to determine the direction the birds were trying to jump to. But the students would not be able to know if the cage were grounded or not. He said that when the cage was not grounded, the direction was random (presumably due to EM interference), while when the cage was grounded the direction was consistent with the local magnetic field.

I was imagining that very low frequencies would be unlikely to behave differently from truly static potentials, and that very high frequencies would probably just reflect. So that got me wondering about what frequencies might not do that and what would control the coupling.
 
BowlOfRed said:
I assure you it is not misstated. I looked it up because I heard the author talk about the study on a podcast. He stated that because the cage only works when it is grounded, he used that as a blinding mechanism.

He would ask the graduate students to determine the direction the birds were trying to jump to. But the students would not be able to know if the cage were grounded or not. He said that when the cage was not grounded, the direction was random (presumably due to EM interference), while when the cage was grounded the direction was consistent with the local magnetic field.

I was imagining that very low frequencies would be unlikely to behave differently from truly static potentials, and that very high frequencies would probably just reflect. So that got me wondering about what frequencies might not do that and what would control the coupling.
Click to expand...

If the cage was grounded at, say, the edges along the bottom it would effectively change the Faraday cage structure and would significantly reduce the longer wavelength levels which may not be attenuated much by the Al cage. Much of their work was at these wavelengths but the cage attenuation was measured at generally higher frequencies.

This is why I say it warrants an extended discussion of the way the Faraday cage was "grounded."

If that was included in the study then great, otherwise I can't understand how Nature would publish it let alone place it in the lead as a cover article.

If "grounding" the Faraday cage made the sort of change they saw then it was because they changed the physical structure of the Faraday cage and had nothing to do with the grounding but rather with the relatively low penetration of B fields perpendicular to the floor. One could achieve similar effects by running a copper pipe around the outer base edge to complete a loop. Grounding that pipe or not would then have no significant effect. This is why it warrants an extended discussion and should not be referred to as simply grounding the Faraday cage in the summary.
 
Last edited:
MRC_Hans said:
For static purposes a faraday cage will protect its content from external charges. However, if it is not at ground potential, you may get a nasty surprise when you try to leave.:eek:

For EM shielding, it is different. As you already guessed, there is a capacitive coupling between the cage and anything inside (this is true even if the inside parts are galvanically connected to the cage). Thus, if the cage is not grounded, an AC signal may be coupled into it.

Hans
Click to expand...

This is only true to the extent the Faraday cage is not ideal and even there it is a second order effect due entirely to the Faraday cage deficiencies.
 
marting said:
If the cage was grounded at, say, the edges along the bottom it would effectively change the Faraday cage structure and would significantly reduce the longer wavelength levels which may not be attenuated much by the Al cage. Much of their work was at these wavelengths but the cage attenuation was measured at generally higher frequencies.

This is why I say it warrants an extended discussion of the way the Faraday cage was "grounded."

If that was included in the study then great, otherwise I can't understand how Nature would publish it let alone place it in the lead as a cover article.

If "grounding" the Faraday cage made the sort of change they saw then it was because they changed the physical structure of the Faraday cage and had nothing to do with the grounding but rather with the relatively low penetration of B fields perpendicular to the floor. One could achieve similar effects by running a copper pipe around the outer base edge to complete a loop. Grounding that pipe or not would then have no significant effect. This is why it warrants an extended discussion and should not be referred to as simply grounding the Faraday cage in the summary.
Click to expand...

I have access to the full article--

I don't see a problem with their experiment. Grounding certainly makes a difference.

Their Faraday cage is 5-sided (no floor) solid aluminum plate, connected (or not) to a single grounding rod. They also tested completing the cage with a floor but found it had little effect. They don't explicitly say how well the floor edges were connected to the walls in that test, though they do describe how well the other panels are attached to each other, so you'd expect them to have done the same here.

This is really beside the point of the test though: the EM measurements they took next to the birds correlated with the bird's orienting performance, so their cage was clearly effective enough.
 
Pulvinar said:
I have access to the full article--

I don't see a problem with their experiment. Grounding certainly makes a difference.

Their Faraday cage is 5-sided (no floor) solid aluminum plate, connected (or not) to a single grounding rod. They also tested completing the cage with a floor but found it had little effect. They don't explicitly say how well the floor edges were connected to the walls in that test, though they do describe how well the other panels are attached to each other, so you'd expect them to have done the same here.

This is really beside the point of the test though: the EM measurements they took next to the birds correlated with the bird's orienting performance, so their cage was clearly effective enough.
Click to expand...


Interesting. No floor and solid Al on the top and 4 sides does not a Faraday cage make. Since the cage is either grounded or not at one point the major difference is the presence of E Fields at the freqs they are measuring. The box would attenuate them but there would be a much larger attenuation when grounded. There should be no impact on B fields from the single point grounding. The amount of attenuation depends largely on the capacitance of the wall edges to the ground when ungrounded.

Is there an analysis of the way the "Faraday cage" modifies the EM fields when grounded?
 
marting said:
Interesting. No floor and solid Al on the top and 4 sides does not a Faraday cage make. Since the cage is either grounded or not at one point the major difference is the presence of E Fields at the freqs they are measuring. The box would attenuate them but there would be a much larger attenuation when grounded. There should be no impact on B fields from the single point grounding. The amount of attenuation depends largely on the capacitance of the wall edges to the ground when ungrounded.

Is there an analysis of the way the "Faraday cage" modifies the EM fields when grounded?
Click to expand...

Except that they measured both E and B fields, and both were attenuated by grounding the cage (or "cage", if you prefer). The article includes graphs for both, under various conditions.

Their measuring equipment was a Rohde & Schwarz FSV3 Signal and Spectrum Analyzer with Schwarzbeck EFS 9218 E field antenna and an ETS Lindgren Model 6511 B field antenna.
 
Pulvinar said:
Except that they measured both E and B fields, and both were attenuated by grounding the cage (or "cage", if you prefer). The article includes graphs for both, under various conditions.

Their measuring equipment was a Rohde & Schwarz FSV3 Signal and Spectrum Analyzer with Schwarzbeck EFS 9218 E field antenna and an ETS Lindgren Model 6511 B field antenna.
Click to expand...

For some reason the Abstract no longer shows up on the OP's link. I've signed up for Nature but can't access the program. I get the same link w/o abstract and no link to read the paper.

I just watched the video and have a few thoughts.

1. The basic problem is injecting EM. The quasi "Faraday cage" converts whatever is inside to near field effects, the nature of which I can't discern from the video. Did they experiment with B v E field injections?

2. It seems extremely likely that the "ground" they attach is providing capacitive shielding and, depending on the way they have wired up the RF injection, significantly changing the actual injection qtys.

Still, it's a highly interesting experiment with effects that are fascinating. I don't think this is fabricated. Absent getting the paper it is hard to conclude more other than something is going on re the low end of the RF spectrum.
 
marting said:
For some reason the Abstract no longer shows up on the OP's link. I've signed up for Nature but can't access the program. I get the same link w/o abstract and no link to read the paper.

I just watched the video and have a few thoughts.

1. The basic problem is injecting EM. The quasi "Faraday cage" converts whatever is inside to near field effects, the nature of which I can't discern from the video. Did they experiment with B v E field injections?
Click to expand...
They used the same model B field loop antenna to inject noise. The graphs for that case show both B and E intensities at least an order of magnitude above the baseline that didn't affect the birds, over the 2kHz-5MHz range.

2. It seems extremely likely that the "ground" they attach is providing capacitive shielding and, depending on the way they have wired up the RF injection, significantly changing the actual injection qtys.

Still, it's a highly interesting experiment with effects that are fascinating. I don't think this is fabricated. Absent getting the paper it is hard to conclude more other than something is going on re the low end of the RF spectrum.
Click to expand...

They do have a theory as to why the birds might be effected this way: the magnetic sense is an orientation method of last resort, and the presence of this range of RF noise tells them that there's a solar storm making it unreliable. Red light also turns it off, presumedly because it tells them at sunrise/sunset to use the sun instead.

They're at a loss though as to how the birds are detecting it.
 
Pulvinar said:
They used the same model B field loop antenna to inject noise. The graphs for that case show both B and E intensities at least an order of magnitude above the baseline that didn't affect the birds, over the 2kHz-5MHz range.

They do have a theory as to why the birds might be effected this way: the magnetic sense is an orientation method of last resort, and the presence of this range of RF noise tells them that there's a solar storm making it unreliable. Red light also turns it off, presumedly because it tells them at sunrise/sunset to use the sun instead.

They're at a loss though as to how the birds are detecting it.
Click to expand...

I must say I don't see how the single grounding point could have a significant effect on the B fields inside the cage. I'm looking forward to reading the paper. I don't know why I can't seem to get it online after subscribing nor why the Abstract isn't showing up. Just the authors and title.
 
I was able to get a link to the paper.

There isn't enough info for me to see how grounding the 5 Al sided frame at one point could produce the signal changes shown.

I suspect an instrumentation issue. Were the instruments self contained (and powered) inside a 6 sided, shielded, structure grounding the outside or not should make no discernible difference in the internal fields at AM frequencies with the relatively good shielding they are using.

Apparently they briefly tested a 6 sided shield but didn't find it productive. Unfortunately, earth for a bottom and using one deep earth ground makes it difficult to identify what is going on.

It is a fascinating investigation. Something is obviously affecting the bird's B field navigation. I'm sure they will receive plenty of feedback on improving the setup. I look forward to further developments.
 
Q: Do you need to ground a Faraday cage?


A: Not as long as it is behaving properly.
 
Perhaps one of the reasons a 6 sided, Al sheet Faraday cage wasn't used is that when they tried it they got confusing readings. An emitted RF B field from a loop will create all sorts of standing waves as a high percentage of the changing B field will be reflected from the walls. No doubt the lack of a bottom sheet helped that a lot.

It's a fascinating experimental result on the margins of accepted science and the approach they are using seems pretty clever. I don't see an obvious problem. However, the grounding of the open bottom "Faraday Cage" is sloppy and the abstract verbiage re grounding the Faraday cage is highly misleading when it has an open bottom.

This really calls for a more careful experiment with an actual Faraday cage. They could also use ferrite material positioned at points on the inside to control the B field buildup from reflections and make a more controlled B field.
 
You must log in or register to reply here.

Back
Top Bottom