What is the Impact of Sheetmetal on Electromagnetic Signal Propagation?

[chid]

Hi I'm wondering what properties affect the propagation of electromagnetic signals.

I am interested why an aluminium case would adversely affect signal propagation, for example in this case, here


And why is it that a small gap in that casing will allow signals to propagate?
(is the reason related to that the circuit is not complete and thus does not induce a current as a result of the signal)

 

[Simon Bridge]

You've figured that an Al case acts as a shield because Al is a conductor. But I don't get your example - it'a a picture of a phone.

Put simply:

The field inside a conductor is always zero because the charges in a closed conductor are free to rearrange so that they will always cancel out. The applied EM just shifts charges around ... the rest of the charges shift in response sot he total field inside remains zero.

If there is a gap in the shell, then there are some places charges cannot move to, so there is a place where the field won't be totally canceled out (it can't because the charges cannot get to where they need to be to do this). So some of the EM gets through.

It gets a bit trickier than that :) eg. a Faraday cage has quite large holes in it.

 

[chid]

Thanks Simon,

Sorry for not explaining the photo, it's a photo of a phone with an aluminium loop case around it.

I imagine a conductor loop in a different plane, say around the midsection (over the front and back), would have a lesser effect on the EM. Is this justified?

Just curious what type of equations would this involve if I were to determine the proportion of signal that is shielded by this loop?

Thanks!

 

[Simon Bridge]

Hmmm ... iirc the conducting loop around the iPhone is the antenna. Not a shield.
(checks) I'm reading that the antenna actually runs along one side in the iPhone4 ... the antenna is internal so the case should be designed not to block the signal.

What makes you think the loop is a shield?

The QB-EM shield looks different.
Dunno how they work exactly - I imagine that the EMI that causes the problem is directional.

 

[chid]

well it's actually an aluminium case over the top of the iphone.

I wonder what mechanism the EM shield works

 

[CWatters]

Hmmm ... iirc the conducting loop around the iPhone is the antenna.

and it appears to be more than one antenna..

http://www.iphoneactive.com/2011/01/new-antenna-cdma-iphone-4.html

 

[CWatters]

I wonder what mechanism the EM shield works

http://en.wikipedia.org/wiki/Faraday_cage

 

[Simon Bridge]

Yeah - the faraday cage is the one for radio signals.
See this guy demonstrate Faraday cages (scroll down) ... they don't have to completely enclose something to shield it but the shielding is directional.

But this is why I was wondering if the Al was intended as an EM shield - if it is outside the antenna then a sold shield may block the EM that makes the phone work as, you know, a phone. It may just be intended to be pretty.

 

[the_emi_guy]

I get the following question a lot: If I put a floating piece of sheetmetal near a PCB will it:
1 - increase radiated emissions because it will pick up noise and become an antenna?
2 - reduce radiated emissions because it will provide shielding or because "image currents" will be created that will cancel outgoing emissions?

In my experience neither of these wind up being good models for what is happening. Generally the sheetmetal simply introduces parasitic capacitances which, then, may effect emissions.

This distinctive sound (from Simon's "QB-EM shield" link) is what we call "TDMA buzz". The TDMA signal has a strong amplitude modulation at specific audio frequencies. Any non-linear device in the audio equipment will rectify (detect) the RF, then after the inherent low pass filtering we get this buzz. It is an issue for several reasons:
1 - Requires the audio equipment to be designed with 1.9GHz signals in mind. Cheap audio equipment may not have been constructed in a way to permit good filtering at 1.9GHz. There may not even be a ground plane in there. See www.maxim-ic.com/an3880.
2 - CE mark requires audio equipment to have RF immunity to 3V/m 1.9GHz fields with 1KHz AM. But you have to be 2 or 3 meters away from a cellphone before field strength is down to 3V/m. iPhones are usually much closer than this to their speakers. Also, not all speakers are CE marked.

Back to the "shield". It is a misnomer to call this a shield. It is simply a piece of sheetmetal in close proximity to the iPhone chassis and its antenna. It introduces parasitic capacitance between the antenna and the chassis which loads the antenna and disturbs the antenna pattern. The noise is gone because we have weakened, and perhaps altered the directivity, of the interfering RF signal.