[600MRG] Is this snake oil?
Tom W8JI
w8ji at w8ji.com
Sat Mar 25 19:35:21 EDT 2023
Hi Ed,
Common mode is a complicated subject. The devil is in all the details.
I looked at the LFA page just now but it doesn't tell me anything
useful. They didn't compare the same antenna or same antenna pattern
adjustments with an LFA feed to the same antenna with a different feed.
They compared two different antennas, their antenna and another unknown
antenna. We don't know how much of the pattern change was from the feed
method vs. how much from different element tuning and spacing.
I've never measured or modeled an LFA feed as used in their array,
because I don't know the details.
I will say that direct unbalanced coax fed full wave loop antennas will
not drive the coax with hardly any common mode, balun or not. A full
wave loop or quad element can hardly be made to drive any common mode on
a feeder. The reasons for this are the loop itself has a more confined
electric field, and the loop has an extremely high common mode
impedance. Its own common mode impedance is so high that as a source
driving common mode it does a miserable job, plus a normal feed line
exit brings the feed line out where the electric field of the loop is
not fringing down the length of the coax like it does with an open ended
dipole.
We know full wave loops are inherently major common mode suppressors all
on their own without any balun assist.
As a matter of fact it was neglecting this behavior that caused
Kmosko,/W2NLY/and Johnson, W6QKI to erroneously measured 2dB quad gain
over a Yagi. They had a Yagi with a poor feed and compared it to a Quad
with a good feed, but they didn't realize this. This behavior (nobody
understood common mode at feed points back then) created a sub-standard
Yagi, and the quad had a near perfect feed.
It would be interesting to evaluate an LFA and see if the design a
carries this quad feed point advantage.
As to the 630M antenna, the common mode generated depends on the
impedance of the ground system the antenna is pushing against and the
antennas terminal impedance at the element end being fed. If the
antenna element common mode or driving impedance is 1 ohm (the reactive
part is ignored) and the ground common mode resistive part is 20 ohms,
you would probably have considerable common mode driving the feed line.
The feed line would be trying to augment the poor radial system, there
would be enough radial common point voltage top drive the coax shield
pretty well.
How much noise this adds depends on many complex unknown things, but if
I had a poor ground system (like we probably all do) I would look into
isolating the feeder before it gets near noise sources. It might not
help...but it can't hurt.
During the daytime noise on my full size L measures -73dBm more or less
in a 3.1kHz standard noise bandwidth. But this is also a measure of
antenna efficiency. When the band opens at night the noise increases a
few dB, so it is at least partly limiting on propagated noise even on
the most quiet nights. I've been recording levels from a Cuban broadcast
station on 530kHz and comparing those levels to my inverted L noise
levels. 630M has a lot of propagated sky-wave noise here. I'm using a
pretty accurate instrument to measure levels.
Your -110 dBm seems abnormally quiet for 630M even if we move it 20dB or
more for antenna efficiency differences. The -90 dBm is more believable
on a TX antenna.
What bandwidth was that noise measured in?
73 Tom
On 3/25/2023 1:08 PM, Ed Cole wrote:
> Tom,
>
> Interesting topic (addressing the galvanic isolator, I presume).
>
> My LFA feed 6m yagis use a balun that appears to be "just" coax with a
> whole bunch of ferrite beads along its length.
>
> I have used this approach to minimize common mode on other antennas
> (usually placed near the antenna end of the coax run).
>
> I did not use any on my 600m inverted-L that had base loading coil.
>
> Coax tapped into the coil near the ground end (which went to a ground
> rod and had radials on the ground). Coax shield connected to ground
> end of the coil.
>
> My new "T" antenna will have similar configuration. The vertical will
> be 300-ohm open wire feed to two dipoles with 4:1 balun at ground end
> for use on 80 or 40m. The open wire feed will be disconnected from
> the balun and shorted together to act as a shortened vertical with
> dipoles acting as top hats and base connected to the same loading coil
> set up. Coax runs about 100-foot to my 630m transmitter (converted
> 100w NDB) and K3 transceiver in the shack.
>
> See any need for common mode protection? I can acquire ferrite beads
> that will fit over RG-213 (to be placed at the coil connection). In
> the past my noise floor ran between -110 dBm and -90 dBm (as measured
> by my SDR-IQ).
>
> 73, Ed - KL7UW
>
> On 3/25/2023 6:53 AM, Tom W8JI wrote:
>> It is sad how little our community knows and understand the way
>> common mode is created, and how that common mode enters a system
>> signal path. It leave us open for all sort of odd devices and theories.
>>
>> With a coaxial line, common mode can only enter the signal path at a
>> flaw or break in the shield. It works this way in everything from
>> loop antennas to our feed lines. A "shielded magnetic loop" does not
>> receive signals from the wire inside the shield, for example. The
>> wire inside the loop shield is just a transformer winding that
>> couples to the shield inside layer, and that layer couples to the
>> shield outside at the shield gap where the necessary electric field
>> appears to drive the outside of the shield. The outside layer of the
>> shield is the only actual antenna if the system is built properly. If
>> it is improperly built, then the feed line shield and everything
>> connected to the feed line shield it is involved.
>>
>> This is even why a small e-probe antenna gets more sensitive when we
>> mount it higher, it has more feed line shield to pick up signal.
>>
>> In ALL of these cases , the point of common mode ingress into the
>> receiver signal path is always at a discontinuity or break in the
>> shield. In a shielded loop it is at the shield gap. In a e-probe it
>> is at the probe element, where the shield center is continued out
>> past the shield by the "probe". This causes people to miss what is
>> really going on, and they make up a well intentioned but seriously
>> flawed theory.
>>
>> If the shield has good integrity to the receiver input any ingress at
>> the receiver would be negligible compared to ingress at the antenna
>> or an upstream gap or break. The meaningful leak-in point, unless
>> there is a flaw in the cable shielding, would be at the antenna.
>>
>> Another thing often missed is the antenna (which by definition needs
>> common mode on it to function as an antenna) receives signals
>> directly from noise sources through local induction fields or long
>> distance electromagnetic radius fields. Even if the feed line shield
>> has unwanted common mode noise, the noise level on the shield has to
>> exceed the desired antenna's pickup of noise.
>>
>> One of the very worse antennas for this is an E-field probe, because
>> the lack of a stable infinite ground means the shield is as much the
>> antenna as the probe is!! There is no way around this without an RF
>> ground plane (electrical mass) at the probe! This is why elevated
>> probes become more sensitive with more height...because the shield is
>> the bulk of the antenna.
>>
>> All of this is important to get our heads around to build better and
>> cleaner systems. To me, while this isolator has a possible function
>> (I designed one for DX Engineering 20 years ago so I obviously
>> thought they have a place) it would not be when it is placed at the
>> receiver. It would be when placed somewhere outside where the shield
>> path should be broken before the reaching the ingress point, which is
>> usually at the antenna. I sure wouldn't put one at my receiver or at
>> my receiver multiplexer panels. A single proper bead would be way
>> better in my shack, if I need anything.
>>
>> 73 Tom
>>
>>
>> On 3/25/2023 12:26 AM, D.J.J. Ring, Jr. wrote:
>>> https://www.bonito.net/hamradio/en/galvanic-antenna-isolator-gi1000/
>>>
>>> The Galvanic Antenna Isolator GI1000 suppresses noise reaching the
>>> ground connections on the receiver. The GI1000 covers a wide-band
>>> range of 50 kHz – 1 GHz (typ. 1,5dB insertion loss). This unit can
>>> be used in many receiving applications, It will also work with
>>> frequencies up to 1,25 GHz but above 1000 MHz insertion loss may
>>> increase by up to 3db. The unit also has integrated double
>>> over-voltage protection and input and output are blocked for DC
>>> voltage.
>>>
>>> The galvanic isolator GI1000 separates the path of the direct
>>> current between the outer shield of the coax cable and the shielding
>>> of the antenna feeding line in order to suppress interference caused
>>> by potential differences. This is achieved by using a small toroidal
>>> transformer. The inner conductor of the coax cable is insulated with
>>> capacitors providing coarse and fine voltage surge protection.
>>>
>>> Close connection please
>>> The GI1000 should be connected to the receiver as closely as
>>> possible to avoid other interference. We offer short very adapter
>>> cables with different connectors.
>>>
>>> Galvanischer Antennen Isolator GI1000
>>>
>>> GI1000 & CCMC30 a great team
>>> Even a cascaded use of the GI1000 directly at the receiver input and
>>> the CCMC30 following it directly after the coaxial power inserter
>>> can in some cases lead to an increased efficiency. Especially in the
>>> lower frequency range, surprising improvements can occur:
>>>
>>> CCMC30 mit G1000 kaskadiert
>>>
>>> Galvanic Antenna Isolator GI1000 Specifications:
>>> Excellent protection against sheath waves and voltage surges.Double
>>> voltage surge protection. On the antenna side Gas Discharge tube
>>> with an ignition voltage of 90v. On the receiver side you will find
>>> an ESD-diode (30KV; max.pulse power 350W (8/20μs) Input and output
>>> has been blocked for direct currents of max. 50V.
>>>
>>> Tested by Clint Gouveia
>>> The well known DXer Clint Gouveia (Oxford Shortwave Log) is using
>>> the GI300 (previous model) on nearly all his DX-Expeditions. He
>>> wrote a Test including Videos. Please ready his tests here: read the
>>> test
>>>
>>> 73
>>>
>>> DR
>>> N1EA
>>>
>>> ______________________________________________________________
>>> 600MRG mailing list
>>> Home: http://mailman.qth.net/mailman/listinfo/600mrg
>>> Help: http://mailman.qth.net/mmfaq.htm
>>> Post: mailto:600MRG at mailman.qth.net
>>>
>>> This list hosted by: https://www.qsl.net
>>> Please help support this email list: https://www.qsl.net/donate.html
>>
> ______________________________________________________________
> 600MRG mailing list
> Home: http://mailman.qth.net/mailman/listinfo/600mrg
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:600MRG at mailman.qth.net
>
> This list hosted by: https://www.qsl.net
> Please help support this email list: https://www.qsl.net/donate.html
--
This email has been checked for viruses by AVG antivirus software.
www.avg.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.qth.net/pipermail/600mrg/attachments/20230325/1cbee343/attachment.html>
More information about the 600MRG
mailing list