Mark, I fully agree.

The amount of unscreened signal picked up by an unscreened multi-turn loop
is negligible compared with what is picked up by the action of the loop
itself and then magnified by the Q of the loop. The unwanted signal cannot
possibly be more than that which would be picked up by a very short vertical
of height equal to the loop diameter and would probably be less.

If there's a problem it is more likely to be picked up on the feedline which
is not influenced by the presence or absence of a screen around the loop.
If something must be screened then screen the feedline.

Loop screening is needed only when precision direction-finding bearings are
being taken with an in-the-clear, precision-constructed, large loop.
----
Reg, G4FGQ

"Tdonaly" wrote in message
...
Mark Keith wrote,

Tdonaly wrote:



Yes, but does your small, inefficient, shielded loop improve the
signal-to-noise
ratio in the directions of its maximum gain over say, a non shielded
loop?

Many claim this, but I didn't see it when I compared them. I found a
shielded coax loop just as susceptible to local, and not so local noise,
as a non shielded loop assuming both are balanced. This is not counting
the feedline, or any common mode currents unbalancing the loops. Both
are capable of very sharp nulls. No difference really, and both are good
at nulling a noise source. But a shielded coax loop quieter than a
regular loop? I don't see it. It's not the loop itself, or having a
shield. It's the keeping of good balance. The shielded loop design and
method of feeding forces a good balance. But if you have a regular loop
that is also just as balanced, I maintain it's just as "quiet". To me,
this "shielded loop being quieter" theory is an old wives tail of sorts.
My two favorite MW loops are both unshielded. Ones a 16 inch dia circle
with 12 turns, and my big one is a diamond with 44 inches per side. "5
turns". Both are on floor stands indoors, and rotate.
I've tried using shielded coax loops, and I saw no reduction of noise.
I've also compared using both shielded and non shielded coupling loops
to feed the loops. Again, no difference in noise levels. MK

--
http://web.wt.net/~nm5k



This pretty much squares with an article on shielded loops written by
Glenn S. Smith of the Georgia Institue of Technology in _The Antenna
Engineering Handbook_. He says the shield enforces symmetry so that
the pattern doesn't suffer, and that's what it's supposed to do. No
mention
of noise at all.
73,
Tom Donaly, KA6RUH


That's what the ARRL antenna book also claims. They talk about shielded
loops in the context of direction finding antennas. The shield is supposed
to make the antenna balanced with respect to ground, and retain
directionality. Also, and I don't recall if anybody mentioned this, but the
shield can not be closed around the circumference, and the maximum wire
length is on the order of lambda/10. The loop is tuned to resonance with a
parallel capacitor.

Tam/WB2TT

Reg Edwards wrote:

If there's a problem it is more likely to be picked up on the feedline
which is not influenced by the presence or absence of a screen around
the loop.

Agreed

If something must be screened then screen the feedline.

It generally is screened already (coax) but it does need some kind of
balun.

It's amazing how many loop designs are paranoid about balancing and
screening the loop itself, but then connect the coax in a totally
unbalanced way. The result is a beautifully balanced loop in parallel
with a vertically polarized random wire.



--
73 from Ian G3SEK 'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.co.uk/g3sek

Hi Ian,

Some old wives even extoll the virtues of screening the coupling loop of a
magloop.

It may be kindly said that screening a loop at least does no harm.

But the screen greatly increases capacitance across the loop and thereby
restricts the tuning range of the proper capacitor. The number of turns has
to be decreased. Or in the case of a single-turn loop its diameter must be
reduced which also rapidly reduces receiving sensitivity.

I've a feeling it also degrades loop Q. It certainly can't improve it.
----
Reg, G4FGQ

========================================

"Ian White, G3SEK" wrote in message
...
Reg Edwards wrote:

If there's a problem it is more likely to be picked up on the feedline
which is not influenced by the presence or absence of a screen around
the loop.

Agreed

If something must be screened then screen the feedline.

It generally is screened already (coax) but it does need some kind of
balun.

It's amazing how many loop designs are paranoid about balancing and
screening the loop itself, but then connect the coax in a totally
unbalanced way. The result is a beautifully balanced loop in parallel
with a vertically polarized random wire.



--
73 from Ian G3SEK 'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.co.uk/g3sek

"William Mutch" schreef in bericht
the shielded loop performs this way is that near field noise is
cancelled while far field signal is only attenuated by some factor
relating to capture area. In my temporary rooftop mount I was unable to
easily check out the effect of broadside null.

William,
I have done some work on local QRM reduction during the last
few years. Summarized on:
http://home.plex.nl/~jmsi/
Most important is avoiding any coupling with the coax/feedline.
With small magnetic loops this is easy to accomplish and my
guess is that this is why loops are less susceptible for local QRM.
That is why I choose small loops instead of small dipoles.

73 de Jan PA0SIM

Reg Edwards wrote:
Depends upon the source of the noise. I remember a small shielded loop
being effective against localized electrical noise in my college dorm.

Yes, but did it make any difference when you removed the shielding?

All I know is that it was extremely superior to a 5' telescoping vertical
receiving antenna. I was amazed and delighted at the difference. I could
copy Radio Moscow on my SX-99 which I couldn't even detect on the 5' vertical.
--
73, Cecil http://www.qsl.net/w5dxp



-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----

Reg Edwards wrote:
"Yes, but did it make any difference when you removed the shielding?"

I think Mark Keith`s testimony is accurate. Signal grabbing depends on
the area enclosed by the loop.

Small loops discriminate against noise due to their directional
response. So, if the actual antenna is the shield or the contained
conductor makes little difference except we have Terman`s testimony that
the shield can equalize electrostatic response and in some cases reduce
noise.

Best regards, Richard Harrison, KB5WZI

On Tue, 13 Jul 2004 17:57:21 -0500, "Crazy George"
wrote:

OH, for Pete's sake. Loops are sensitive to the H vector. Wires receive
the E vector. Most near field noise tends to be predominantly E field.
But, that seems to only be effective up to 3 or 4 MHz, due to the wavelength
factor, i. e. the near field shrinks as you go higher in frequency. Fully
formed far field wavefronts of noise sources will be just like wanted
signals, and unless some polarization difference is available, then
directivity is the only way to improve S/N. Only in special circumstances
can you see much improvement above 5 MHz due to near field/far field
differentiation.

In the _far_ field both the E and H fields are inversely proportional
to distance and have the 120 pi (377 ohm) relation (impedance) between
the fields. However, in the _near_field_ ( 1 lambda) the 377 ohm
relationship is no longer valid and the magnetic field is inversely
proportional to the square of the distance, while the electric field
is inversely proportional to the cube of distance.

Summarising the graph from an article by Lloyd Butler VK5BR in Amateur
Radio, August 1990: The output voltages from both E and H field
antenna system are calibrated to the same value at 1 lambda (i.e. in
the far field). The antennas are moved closer, when the E and H
antennas are moved to 0.05 lambda, the E antenna delivers 50 dB and
the H antenna 40 dB (relative to 1 lambda) i.e. the H-field is 10 dB
quieter. At 0.005 lambda, the E field antenna output is 110 dB and the
H-field 80 dB, i.e. the H field antenna is 30 dB is quieter.

Thus, with same far field sensitivity, the sensitivity to very local
interference is attenuated considerably when _only_ the H field is
used. However, at 3.5 MHz and 80 m wavelength, 0.05 lambda corresponds
to 4 m and 0.005 lambda to 40 cm, so we are talking about really close
noise sources. At even higher frequencies the number of potential
interference sources is dropping within the 0.05 (or even 0.1) lambda
radius from the receiving antenna, in which the H antenna has an
advantage.

However, on the 135 kHz LF band (lambda 2.2 km), the distances would
be 110 m resp. 11 m, thus much more unwanted interface sources could
be eliminated.

Shielding the H-loop simply prevents the stronger E field from
entering the loop and thus destroying part of the advantage of using
the H-antenna.

Paul OH3LWR