Why does my phone get a signal in a metal box?

[H_man]

Why does my phone get a signal in a metal box??

Hi guys,

Earlier today I was telling someone about how EM radiation of low wavelength would have trouble passing through a cold plasma. I gave the example of the metal mesh in front of a microwave (as metal can be treated like a plasma).

Well... to demonstrate I popped my mobile in a metal box and tried calling it. The damn thing rang. So then I tried earthing the box and the signal still got through.. Do I need a thicker box or I have I misunderstood something fundamental.

Cheers,

Harry

 

[DaveC426913]

Hit the start button on the microwave. That'll stop your phone from ringing.

 

[masudr]

In a conducting medium, the wave equations simplify to a damped oscillating equation. The coefficient of the damping term gives us the skin-depth: the depth at which the EM wave will be $e^{-1}$ of it's value.

If the thickness of the material is about one order of magnitude greater than the order of the skin-depth of the material, and if the material is truly a conductive media, then EM waves will be severely attenuated. I doubt at that thickness the mobile phone could pick anything up. If the material was about as thick as the skin depth, we're only looking at a factor of about 0.37 (from memory)

We would also need to know what the signal strength is outside the box, and the minimum signal strength required to be able to send/receive calls which will depend on your phone model, and the proximity of phone masts. I'm afraid my knowledge of the power of mobile phone receivers/transmitters is poor. Perhaps someone else knows or there is a specification for your phone model on the internet that might tell you.

 

[H_man]

I always thought about the skin effect in terms of waves traveling inside the wire and thus not quite the same thing as waves trying to enter the medium from outside.

In the skin effect the resistance increases with freq. whilst in the ionosphere the low freq. ones get bounced back and the high travel through.

Thanks for your though I am still a little confused..

 

[masudr]

Yes, I suppose there's also the issue of angle of incidence and such-like, as that, combined with the geometry of the object, will affect the signal strength inside the box.

 

[berkeman]

Most likely the box top did not make a good RF seal, so the box was not a good shield. You can't just put a simple friction-fit metal box top on a metal box and expect a good shield. When we design RF shielding for either emissions or immunity, we have to pay a lot of attention to the seams.

If you welded the top (or door or whatever) on, and made sure all the other box seams were welded, then you probably wouldn't be able to call it. Poor phone.

 

[cesiumfrog]

Wow.. I can call my mobile even when it's sitting inside my microwave oven! However, from past experiment, the desired demonstration can be heard by shutting a pocket AM/FM radio inside the microwave.

 

[DaveC426913]

Most likely the box top did not make a good RF seal, so the box was not a good shield. You can't just put a simple friction-fit metal box top on a metal box and expect a good shield.

Really? Wow. Sneaky buggers, those RF waves.

So, how is it that a tiny crack will let radio waves in, but the grid on a microwave is effective enough to stop microwaves?

 

[Gokul43201]

Really? Wow. Sneaky buggers, those RF waves.

So, how is it that a tiny crack will let radio waves in, but the grid on a microwave is effective enough to stop microwaves?

The skin depth goes inversely like the square root of the frequency. Microwave frequencies are what, 4 to 10 orders of magnitude higher than radiowaves (but I think cell phones operate in the microwave range)?

 

[H_man]

In reply to Dave...

I actually remember seeing not long ago someone with a detector trying to eliminate any leaks from the microwave he was working on (as part of some research) and after almost several weeks he gave up only to discover that it seems most microwaves do leak.

I assume its not enough to be harmful though

Although Gokul... now that you mention the skin effect again I re-state something I am a wee bit confused about. Why doesn't the ionosphere have a skin effect, it seems to have the opposite. If the propagating wave has a frequency above the electron plasma frequency it can get through.

Cheers for the insight Berkeman

 

[berkeman]

Really? Wow. Sneaky buggers, those RF waves.

So, how is it that a tiny crack will let radio waves in, but the grid on a microwave is effective enough to stop microwaves?

Yeah, the microwave oven thing has always bothered me. I've had to stand on my head at RF frequencies (like 30MHz to 1GHz or so) to get good RF gasketing and shielding to work, and the dang 2.4GHz microwave ovens have a rubber gasket on the door! But then I learned how lossy microwave ovens are, and it made a little more sense. Kind of worrisome how leaky they are, even fresh from the factory.

One way to think about the metal box and metal top thing, is that if the top is not gasketed well (like with good electrical contact every tenth of a wavelength or so), then the box top can "ring" with respect to the rest of the box, and that let's some EM field into the box.

I actually had this very problem with a product that I was helping to design many years ago. Well, except the source of the RF energy was inside the box, and we were trying to keep it from getting out. Pesky FCC rules and all. We were able to do it without more gasketing, though, by spoiling the box top resonance with a couple carefully placed ground contacts placed near the middle of the top. That spoiled the primary half-wave resonance, and the higher order resonances that resulted were low enough in amplitude that we passed the radiated EM test limits.

 

[Gokul43201]

Although Gokul... now that you mention the skin effect again I re-state something I am a wee bit confused about. Why doesn't the ionosphere have a skin effect, it seems to have the opposite. If the propagating wave has a frequency above the electron plasma frequency it can get through.

This is actually NOT the opposite of the skin effect in metals - it is the same thing!

In metals too, you have a plasma frequency BELOW which, the wave vector for transmitted waves is purely imaginary, and hence the wave decays inside the metal. Above the plasma frequency (typically in the high UV range), metals too are transparent to radiation. The skin effect, therefore, exists below the plasma frequency.

 

[Meir Achuz]

I guess I have a cheap cell phone or a good filing cabinet.
My cell won't pick up a call when in the untransparent filing cabinet.
If I could get in there, I would say "Can you hear me now?"

 

[orion141]

I was told in one of my physics classes that one of the reasons that cell phones actually work inside of a conductor (metal box) is because they utilize AC.

 

[berkeman]

I was told in one of my physics classes that one of the reasons that cell phones actually work inside of a conductor (metal box) is because they utilize AC.

Welcome to PF, orion. Unfortunately, you either misunderstood the instructor, or the instructor was mistaken, or they were pulling your leg. Air conditioning will have no effect. (see, now I'm pulling your leg)

 

[gilp456]

Is the microwave a very good faraday cage if lower frequency devices are able to TX/RX while inside (larger wavelength)? It seems backwards to me that it should be able to attenuate 2.45GHz well and not lower freq like 1.9GHz or even a radio tuned to ~100MHz (much larger wavelength).

Can anyone elaborate a little more? Thanks.

 

[DaveC426913]

I recall either a thread or a f2f discussion about this many moons ago.

Apparently, the effort required to insulate something from radio noise is quite substantial. You've got to check every seam and material and do all sorts of other tweaks, or it will leak like a seive.

 

[vanesch]

With conductor shielding, what is very important is the maximum linear dimension that is still an aperture. A very narrow but long slit let's through quite some radiation (from the moment that the slit length is not much smaller than the wavelength). There will of course be some attenuation, but you won't get a genuine "faraday cage" effect if there is a long slit (almost no matter how narrow) somewhere. So yes, the side of the door of a microwave oven is essentially such a slit. In order to do it right, one would actually need to have some zig zag covering with two metal pieces incrusting one into the other. That's why tuner boxes usually have such kind of covers.

 

[Panda]

Whilst shielding you need to take into account skin depth, apertures and surface currents. All of these things have been touched on in previous threads. What I would say is that without testing your Box through either EMC testing or theoretical modelling it is impossible to say what the dominant leakage mode is as the whole EMC field is a black art.
I've been in certified EMC chambers and had my mobile ring, mobile phones are just little ***%%^^^$$'s that will ring no matter how hard you try.