Shielded magnetic loop probe / antennas

[Swamp Thing]

From a purely physics point of view, what would be the performance tradeoffs between the following types of shielded H-field pickups or antennas?

Sources:

• https://owenduffy.net/blog/?p=18180
• https://emcesd.com/tt120100.htm

(A)

(B)

https://owenduffy.net/blog/wp-content/uploads/2020/05/SimplerLoop-300x214.png

 

[berkeman]

I have experience with this type of probe, and have built several for my work. But you need to go first -- what are your thoughts on this type of B-Field RF probe?

 

[Swamp Thing]

The type "B" looks a bit sus to me. The outer is used as half of the loop, so it is exposed to all the electrostatic noise. But there may be some other big advantage to that design, idk.

 

[Baluncore]

Screened loops are magnetic field antennas, hence the term "B-field RF probe". If the external electrostatic shield is asymmetrical, or carries signal current, the output from the probe will also be influenced by the electric field. Screened loops must be significantly smaller in diameter than the operating wavelength, so they have a simple dipole pattern.

The type "B" looks a bit sus to me.

It is more than suspicious, it is bad.

 

[DaveE]

It isn't common for shielding something "half way" to make sense. That's why the others have a tiny gap in the shield. Although I'm sure someone will give us a counter example.

 

[Baluncore]

It isn't common for shielding something "half way" to make sense.

The shield cannot cover the full turn, or the shield will become a shorted turn on the pickup loop.

The gap in the shield is there to prevent a shield current flowing. The shield must be symmetric, so it cannot be open on one side, or it will detect the electric field.

Metal detectors also often use a multi-turn coil, inside a shield that has a narrow insulated gap.

 

[tech99]

I accept that shielded loops do work, for example in direction finders, but it is unclear to me why the alternating B-field inside the loop does not have an associated electric field.

 

[Baluncore]

... why the alternating B-field inside the loop does not have an associated electric field.

Because the loop is small in wavelengths and the local E-field is 'shorted' by the external conductive shield.

 

[Swamp Thing]

Found this paper
https://ieeexplore.ieee.org/document/1284886

also accessible here without login:
https://www.researchgate.net/public...neration_of_Magnetic_Fields_Using_Loop_Probes

If they are right, then there can be situations where Fig (B) in my question could actually make sense. A key point they assert is that if the diameter is not negligible compared to wavelength, then the outer conductor can't be regarded as a short, nor is it really working as a shield -- it is actually part of the pickup structure. So if your aim is to minimize common mode noise then symmetry is the key. The loop has to be designed to work as a balun as well as a pickup loop.

I realized also that there are different scenarios with different goals, for example:

(1) The probe signal is well above our amplifier's noise floor, the instrumentation itself doesn't produce much common mode noise, and we want to isolate the magnetic field from electrical fields near the probe. Example: probing currents and magnetics around a switched power supply circuit.

(2) The probe signal is quite weak, and our equipment (e.g. computer) emits common mode interference into the feed line. Example: a shortwave loop antenna connected to an SDR receiver which is connected to a computer via USB.

My interest happens to be focused on scenario (2) where you can't really have diameter <<<< wavelength because then the signal would be below the amplifier/SDR's noise floor. In this case symmetry is important, and Fig (B) can have some merit.