Common active magnetic loop antenna expected efield rejection due to shield?

[CopperCone]

So say you got a common loop active antenna like the ones that you use in compliance testing good for like 30MHz.

And you got a efield present.

Naturally even a unshielded loop antenna will reject the efield to some amount.

How many db of rejection could i expect from adding a shield to this antenna? I think i unsderstand that you cant have the existance of a varying hfield without a efield but you can change impedances to make it small with the so called shield.

This is because fabricating a nice shielded loop is a bit of a pain. Currently i just use bent aluminum bar. I suppose to make a nice balanced shielded loop i would need some monster sized hardline.. The idea of a loose bent wire in a nicely formed shield just kind of irritates me and i dont want to do that. But if a shield basically does fuck all i dont want to go through the hassle of finding 1 inch hardline etc. This thing will psychologically irritate me unless the dimensiions/balance is good. I am not gonna put a fucking loose wire into a hoola hoop wrapped in copper tape. Just NO!!

Also other then reducing efield pickup, does a shield cause any other effects like bandwidth, distortion, etc. Keep in mind i am basically connecting this to a impedance transforming differential amplifier, if it makes a difference.

Why do i care? Because these are rather fun to make and if i could justify owning both shielded and unshielded varients of the same size i will, otherwise i am inclined to take the amplifier in the unshielded one i built and eventually transfer it over to a shielded one, when i am able to fabricate it.

To me, from a standpoint of measurement, it seems best to seperate e and h field pickup as much as possible, so building a hybrid unshielded one is kind of useless unless it has some kind of improved parameter that neither varient can provide.

However, is there some kind of communications reason to use an unshielded loop (not from an emi measurement prospective), like for signal hunting or getting a quick intuitive feel (so i guess this could be considered practical/tool use). An analogy that comes to mind is how people still like to use analog vom because its filtering effect and mechanical time constants are handy for gaining quick intuition. I would consider this to be a benefit due to a well known parameter which would in this case be partial efieldbpickup and major h field pickup.

Also like, something that would be handy would be to assume you got a monopole, unshielded loop and shielded loop in a e field and signal power of efield in each of the antennas.. And i guess the same for the same three antennas in a hfield (helmholtz coil?)

[CopperCone]

Also what are the differences between the various possible shield topologies?

This is the common method everyone does, with a gap at the top middle:


But I have also seen stuff shielded like this (i.e. in a fluxgate magnetometer probe):

This method seems to leave a cut out band in the middle of the device.

Can you apply the second method to a magnetic loop antenna?

Also in the first case, grounded on both sides? Have seen both methods.  One side? Second case seems kinda shorted but I don't know the analysis.

[T3sl4co1l]

The second one only works for currents through the center.  Like your Rogowski experiment, that failed because it wasn't correctly shielded in this way.

Tim

[CopperCone]

Am I correct in understanding that the CMRR of this antenna is basically its E-field rejection, (when I look at a loop sideways it looks like a monopole), so I kinda imagine it like two dipoles connected to two amplifiers (though the short at the top is odd)

So, with a perfect CMRR amplifier (lets say 0.005% matched resistors) and perfect geometry of the loop, the rejection should be 'infinite'? Does anything else come into play?

[mikemxyzzy]

Here is some information on loops:

http://www.w8ji.com/magnetic_receiving_loops.htm

Scroll down for shielded loop discussion.

[deBug]

Sorry for the late answer, I'm new to the forum, reading up on old threads.

Anyway to answer your question on the shielding effects on the Magnetic loop and how well it can be shielded it is my understanding that the whole "Magloop don't pick up noise" or "the loop can be shielded from the electrical field only picking up the magnetic field" is incorrect.
It is further my understanding that radio radion always has both the magnetic and electrical components and that they can not be separated. 
Shielded loop antennas always has a gap in the shield, this gap will case electrical current on the outside to be able to flow on the inside of the shield effectively making it transparent.
If the shield is closed the antenna will stop working.

//Harry

[T3sl4co1l]

Quite correct, the shield is ineffective against propagating noise, because propagating noise is indistinguishable from intended signals.

It's the near-field stuff, that doesn't have matched, in-phase complementary fields, that the shielding works against.  In this case, nearby electrostatic fields: 50/60Hz buzz from nearby mains wiring, for example.  (The magnetic fields from mains wiring will still be around, of course!)  Likewise, an electric probe antenna will receive very little magnetic induction, particularly when magnetically shielded.

Tim