Thursday, September 21, 2017

The shielded faraday coupling loop myth and other considerations

Most common made "shielded" faraday loop
Experimenting with the magnetic loop is still going on. To couple  the loop to the feedline I use a faraday loop inductive coupling in my loop. Now there are several designs, shielded and unshielded. The main advantage of a shielded loop would be canceling out a vertical or horizontal E-field (electro magnetic interference) or both. The question is if this would be really working?

Some explanation of this theory has been given on N4SPP's site:

However, W8JI has written many articles about antennas and has been working as a tech at a commercial radio station. This is what he comments on the internet:

There is no such thing as a "Faraday shield loop". All radiation and reception comes from the shield. Always. Just like it is a single turn conductor.

The idea you can shield the loop is as silly as the idea you can shield a ground lead.

Once you understand that you will also understand that the only way to not have common mode on the feedline is to make the shield perfectly symmetrical about the feedpoint and have both sides of the "shield" symmetrical and identical with the grounding and feedline entry.

My website shows why high frequency fields cannot penetrate the shield, and why the radiation and reception all comes from the outside of the shield.

A shielded loop is electrically no different than a equally balanced unshielded loop.


Some others comment that the so called shielded faraday loop helps to balance the feedpoint. However, it is all about the feedline that radiates or pickup noises. So a simple 1:1 balun or RF choke should do the trick. You see something like that at the Chameleon loops.

Well, technically speaking or writing it all sounds likely. But I don't like likely and want to test and experiment myself. All what counts is practical use, you can only transmit and receive theoretically in your dreams ;-)

I want to know the difference in receiving interference with a shielded and unshielded coupling loop if there is any difference? Besides that I want to know why many use 1/5 diameter faraday loop and others use 1/4 or even 1/3 diameter? Well, first of all I did a test with two faraday loops 1/4 diameter size. One unshielded the other one shielded as in the picture above. See my not so scientific practical test in the video:

What it revealed to me is actually in the video the shielded loop did receive about the same RFI compared to a unshielded loop. I did not move the loop and I did not move the PSU wich caused the RFI.  The RF choke made from about 8 ferrite cores doesn't help much it seems, although it could help for transmit purposes? Myth busted? Well I don't know for shure, but my practical test shows that you can also use a piece of cheap electrical wire and bend it in a round shape then connect it to a coax with just a thermoplastic connector or solder it to the loop.

And well, I did just that to test more faraday coupling loop sizes. A practical experiment. I made 3 faraday loops. 1/5, 1/4 and 1/3 diameter (compared to the main loop diameter). All loops did well and minimum and maximum frequency with reasonable SWR were the same. However if you need a good SWR on a low frequency you need a bigger coupling loop and for a good SWR on a higher frequency you need a smaller one (at least that is what this test reveals). A good average is 1/4 size. I measured the SWR with my MFJ259 analyzer.

Coupling loop diaFreq.min60m40m30m20m17m15mFreq.max
Shielded coupling
loop 1/4

I'm not the only one that discovered this of course. There are so many things tested and written about the magnetic loop including articles about different feed methodes, skin effect, impedances, radiation resistance etc. ect. But I found only one document from a american experimenter that used a variable coupling loop. Great idea....but he used a square loop and a square coupling loop which is easier to adjust.

Just for testing purposes I made a adjustable coupling loop. I can bend/slide it from 1/3-1/5 size of the main radiating loop. That test revealed about the same as above done with 3 different loops. However I was not limited to 1/4-1/5 and 1/3 size but could get anything in between. While testing it became clear to me that a 1/4 or 1/5 size is not always the way to go. But that story takes too long for this post so I write more about that later.


Photon said...

Very useful. I think the noise level was influenced on how much processing the laptop was doing at any one time, as you might expect, and that may explain the very slight difference.

I've built and operated very many loops. Like me, most serious experimenters who are not out to prove their idea, rather than testing an idea, soon abandon the Faraday loop and use a pseudo-gamme match, which is just a wire soldered/connected directly in various ways to the main loop. This is not so practical for portable loops that are 'bendable', but it can be done (using car spade connectors, etc.) Sometimes, a 4:1 balun is used, sometimes not. Always, the results are better than a Faraday loop. See:

PE4BAS, Bas said...

Hello John, thanks for you comment. I assumed you closed your blog but see you start again. I've to catch up reading. As my loop needs to be as portable as possible I stick to the coupling loop although I know a gammamatch would be more efficient. I like the way we experiment, it doesn't prove anything as for anyone else the results could be different. 73, Bas

hamradioal said...

Thought provoking.
I like it!

-DeeT said...

The optimal coupling loop diameter ratio to obtain a 50 ohm match depends on the circumference of the outer loop in wavelengths, among other things, so it makes sense that the size is somewhat frequency dependent.