Shielding boost converter for RF transceiver

[radiolistener]

Well, I did say steel. Ok fine, AISI 1020 steel, as sand-cast. What more do you need to know?

As for hole, let's say 12mm ID, and wire to make 50 ohm coax. Wire is copper, C110 alloy half-hard let's say.  The copper might be held centered with a solid, foam or spider structure thingy of whatever material; if you insist, let's say it's foamed polyethylene.

I will provide you with a reference to a book about electromagnetic shield design for aerospace equipment, where you can find some straightforward formulas for estimation. It explains shield estimation with an example that is very close to your question, with just one exception - without a hole with wire.

Say we attach the receiver at the surface of the cube. Wire does not extend beyond the cube. Is there still a leakage path?

Technically, you described a classic GP antenna and asked whether there will be radiation leakage from it. Yes it has radiation loss.

But now you proposed to connect the ground of the remote receiver with a wire to the ground plane of the transmitter antenna and are asking if it will still be able to radiate. Yes, it will radiate. Why not? 

I just wonder why you decided that the wire between transmitter and receiver ground will block the transmission?   

Ok, that's incorrect.  You require further study on this point.

What exactly incorrect and why?
If you claiming that the cube surface is not involved in radiation process, then this is nonsense.

The surface of the metal cube acts as a counterpoise to the antenna's radiator (the part of wire outside cube).

The hole with wire works here as coax cable to deliver RF energy from oscillator inside cube to the cube surface.
At the cube's surface, it feeds a GP antenna, where the radiator is the part of the wire outside the cube and the counterpoise is the cube's outer surface (ground plane).

Do I understand correctly that you are claiming the GP antenna fed from the oscillator with coax cable will not radiate?

There are no reflections, the material is wholly absorptive (skin effect dominant) and fields decrease exponentially with depth into the material.

This is your mistake. There is no material in the world that absorbs all energy of electromagnetic wave that penetrate into it.

You can find correction for wave reflections inside metal shield plate in literature about electromagnetic shield design for aerospace equipment.

A little bit later I will give you the reference on a book which explains it in details with math.

Attenuation will be exceptional, even at fractional Hz.  It seems further study in this direction is also warranted.

Even if you use some meta-material with exceptional attenuation, it still just attenuate electromagnetic wave, but not cancel it to zero. 
And as I remember my calculations, 0.5 meters metal plate has pretty bad attenuation for magnetic field at low frequencies even above 1 Hz.

By extension, as the block is solid and conductive, near electric field is zero by definition.

by definition? by whose definition? 

Do I understand correctly that you are claiming it's impossible to radiate an electromagnetic wave with an antenna made from conductive material?

Sorry, but in this case, you are saying things that are complete nonsense.
The conductor can attenuate electromagnetic wave very high but not cancel it to zero.

Therefore, for a good shielding, it is necessary to use a sufficiently thick shield plates.

The intensity of the electromagnetic wave decreases as it penetrates deeper into the conductor. However, if you understand the nature of the skin-effect, you know very well that complete cancel of the electromagnetic wave does not occur even if you increase the size of the shielding cube’s wall to enormous dimensions.

The majority part of the current induced in the conductor will flow near the surface, but that doesn’t mean the electromagnetic wave won’t induce currents deeper within the conductor. It will induce currents deeper than skin-layer, but they will be much smaller than close to surface. How much depends on the frequency and conductor properties. The skin depth simply indicates the thickness of the surface layer where the majority of the current flows, but not all of the current induced by the electromagnetic wave. 

If electromagnetic waves could not exist inside a conductor, then electromagnetic waves would not be able to induce currents in the conductor at all, and the concept of skin-effect depth would lose all meaning.

Hmm interesting. If I carve off the corners to make a maximally sized octahedron let's say, what parameters change, and by how much?  Surely you have a quantitative value in mind...

It's hard to predict the near-field configuration for a custom-shaped antenna. I don’t think there is a simple rule for that. So, if you want to know how changes in antenna shape affect its near field configuration, radiation efficiency and radiation pattern, it’s better to use an electromagnetic simulator.

I think for a GP with ground plane shaped as a cube or octahedron, you can use even some NEC-model based simulator. It uses simplified simulation, but I think it will be well enough for your case.

[radiolistener]

I do not assume myself to teach or change anyone in its behaviour. But in my world - call me allergic or whiny-voiced or whatever - it´s neither friendly nor professional to call others ideas or stuff crap, especially not if it´s just because it´s not highly expensive

I said that SMPS is bad for receivers because it produces noise. That's it. Nothing personal.

Do you really think that I need to ask your permission to say something about SMPS? 

If you think I shouldn’t say this because it might hurt someone’s sensitive psyche, well... You could just as easily be offended by anything. Am I supposed to run after you and beg you not to be upset?

And yes, I think that all SMPS are crap due to their noisy nature.
Some are less flawed, and some are more flawed, but they all have noise issues.

I still think the example of the R&S spectrum analyser nicely shows that it is possible to efficiently filter out SMPS noise.

You can filter it, and it don't require to have expensive R&S spectrum analyzer. But if you try to filter it for receiver, you will realize what I'm talking about. As I proposed before, just try to power receiver from SMPS and measure it's noise on receiver. Then try to filter and shield it. And I'm sure, after that you will realize why I'm talking that SMPS is very bad thing for receivers... 

[radiolistener]

As I promised before, here is the book reference: "Electromagnetic Shielding and Corrosion Protection for Aerospace Vehicles", Jan W. Gooch, John K. Daher, Springer, 2007



And this page shows electromagnetic field waves inside conductor on the picture



And, oh horror, this page even talks about the electric and magnetic fields amplitude inside the conductor, which "is zero by definition,", though it’s unclear whose definition it was.   


@T3sl4co1l, how do you think is it worth to send the mail to the authors of these books, I think it will be fair to get them know that their book "is incorrect." and they "require further study on this point."? Isn't it?   
(to be clear, this is sarcasm, I think their book has very good explanation for EM shielding theory and I believe this is one of the best books about the subject)

[Buriedcode]

wow, I come back to the thread and.. what has happened?  I'll try to read every reply because I'm sure theres some good stuff there but geez...

I'm sure you all realise I wasn't suggesting SMPS can't be used with an RF reciever, in the same enclosure, just that my limited experience - and perhaps my less-than-optimal layout  - has given my cause for concern when mixing the two on the same board (or in the same enclosure).


I'm not sure I agree with radiolisteners assertion that it is almost always bad for the reciever ?

[shabaz]

You're just in time. Radiolistener is going to provide a practical demonstration of the shielding ineffectiveness, by sitting with his wifi-enabled laptop inside a metal cube, and sending an e-mail to some book authors. He'll send a message when he wants to be let out.

He's going to start with a 1 mm thick cube, shouldn't be a problem judging by that diagram he posted showing lots of outward arrows and no numbers.