Jim
I do not question the results of your modeling.
I also appreciate the difficulty trying to find things late at night when
tired. I'm younger than you are and I struggle and make mistakes! (hi)
I am a firm believer in modeling and have used it extensively on these
antennas.
A good portion of my 500 hours were spent just evaluating modeled results.
However modeling is no substitute for field measurements on real antennas.
I have never run 1500w on an OCFD antenna because our power limit is 750w
and I do not run illegal power. I have run full German legal power with
them.
My home brew 80m OCFD has an feedpoint about 20% from one end - similar to
what W8JI recommends - though I determined mine by modeling myself, not from
his web.
I use two Ferroxcube toroids in the home brew balun which are similar to the
FT-140-43 Amidons. Medium sized toroids, NOT big ones. In addition I have
placed a Maxwell 1:1 choke (beads over coax) immediately below the balun.
The antenna has been in the air for 4 years and I run 600w with it on 5
bands during contests. Outside of contests I sometimes run it at 750w.
I've had decent results in CQWW for the past 4 years, sometimes scoring over
1M pts. in about 30 hrs of operation. So the antenna is no dud.
The antenna still measures the same as it originally did. I certainly have
not fried my balun.
I'm also a member of the Yahoo Windom and OCFD group with well over 1000
members. There several people have indeed reported frying baluns in their
Buckmasters and Carolinas', some running just 400w.
This brings me back to my point of balun design. The commercial industry
has been selling these antennas with bad baluns in them for about 50 years.
I contend that this is most of the reason for the shady reputation of the
OCFD antenna.
I have always favored an openwire, center-fed dipole when a single
multi-band antenna was required. I have also usually had a Johnson Viking
or Annecke to match it with.
At this QTH, openwire was not possible (cosmetic reasons).
I was well aware of the reputation of the OCFD and shared that opinion of
them.
Luckily I was also well aware of the lengthy disputes W8JI has had with
other engineers over the single-core 4:1 Guanella not working with hf
antennas. This was reinforced by G3TXQ explaining it to me about 3 times
until it finally sunk into my thick skull.
So I set out to investigate the difference between a single core and dual
core balun in this application. I accomplished that and it was shocking.
I have 500 sets of pairs of measurements, each showing the amount of CMC on
the line, measured at the TX, but with several different lengths of coax -
as well as an associated SWR sweep of the entire band for each CMC
measurement.
You can clearly see when the CMC level goes up, how the skewing of the SWR
curve goes up with it, to the point (with the single core balun, only) that
you get a wonderful double-dip ins SWR, quite like you would expect from a
sleeve dipole.
In order to force common mode current to go up, I moved the coax to one side
or the other, as W7EL had suggested to do in his 1982 paper on "Baluns: What
they do and how they do it."
The preparation for my long test was proceeded by an even longer amount of
time studying your own RFI-ham, and then long discussions with G3TXQ,
GM3SEK, G3UNA, DJ1AT - all very reputable engineers. These 4 helped define
what to measure and how to measure it. Especially GM3SEK coached me in the
calibration of my equipment before beginning the tests, as well as using
Guanella chokes in the A.C. and D.C. leads of the power supply. I built
these chokes according to GM3SEK.
The biggest problem is, there are about 1000 data points waiting to be
evaluated and I have only touched the surface of the evaluation. However I
have began with those points which stood out during the test.
It was easy to find volunteers to help define the test, but nobody has 100
hours to invest in evaluating the results. :-(
Any takers? (hi)
For starters, here are some interesting bits I learned (albeit with 100w):
> The dipole, no balun, but with coax kept exactly 90 degrees from the
antenna had almost no measurable CMC. Note: my instrument is amplified for
under 10mA and inaccurate on that range. Plus the ambient CMC was generally
about 2mA, but varied slightly on different days, so if I measured 3mA, was
it the antenna configuration or just changes in the ambient? No telling.
Moving the coax out of the perpendicular drove the CMC up. Above 10mA my
meter was not amplified and the readings were more accurate.
> ALL OCFD configurations measured (regardless of feedpoint splits)
exhibited the same common characteristics: The CMC was highest (regardless
of how or where measured) on the fundamental frequency and at least a
magnitude higher than it was on any of the even harmonic frequencies!
> Higher even harmonics tended to show no CMC at all unless I dramatically
pulled the coax to one side.
> However, odd harmonics responded differently. A couple of the OCFD
antennas I measured were designed by me and work nicely on 15m, the 3rd
harmonic of the 40m OCFD. The 3rd harmonic had CMC of the same magnitude as
the fundamental. NOTE: I only measured the 3rd harmonic because the 5th
harmonic did not fall within a ham band.
> Understanding the CMC characteristics of the OCFD antenna, and the fact
that it does NOT have to go from d.c. to 30MHz is fundamental in re-defining
the design of the OCFD balun. It must have high CMI over the frequency
range of the fundamental to the 3rd harmonic but may fall off after that.
This led me to using a ferrite mix with a higher permeability and more turns
on the coils.
> The single core 4:1 Guanella balun is about as useful as tits on a bull.
It is PATHETIC in this application. No other way to describe it.
I have lots of other interesting measurements of the effectiveness of RF
chokes, comparing different designs, such as Maxwell, Guanella, and then the
loop version as designed by GM3SEK. I can show measured differences between
these technologies under all kinds of antenna configurations. But in a
nutshell, the Maxwell was a dead last with the Guanella first and the GM3SEK
breathing heavily down its neck.
Jim, I would have loved to have had you out there in the field with me
playing with this stuff!
I wanted to carry on with more measurements during my 2014 vacation, but my
XYL made it clear that that wasn't exactly her idea of a vacation.
I have fried more than my share of commercial baluns in my day, but in
retrospect, all of them were voltage baluns or single-core 4:1 Guanella
baluns. Only one was fried in an OCFD. The others were in openwire-fed
systems. Once I began building my own dual-core baluns, that nonsense
stopped. I have yet to fry one of my own.
73 - Rick, DJ0IP
(Nr. Frankfurt am Main)
-----Original Message-----
From: TenTec [mailto:tentec-bounces@contesting.com] On Behalf Of Jim Brown
Sent: Thursday, April 16, 2015 9:53 AM
To: tentec@contesting.com
Subject: Re: [TenTec] Field day antennas
On Thu,4/16/2015 12:30 AM, Rick - DJ0IP / NJ0IP wrote:
> To all of you guys commenting on the common mode current of an OCFD, I
> have one simple question:
Several years ago, I modeled a 40M OFCD dipole with offsets varying from a
foot increments of a few feet to a lot more, and dissipation in a practical
choke inline with these antennas. The results are not pretty.
Looking late at night (it's 1 AM here), I can't find the pdf showing it, but
the bottom line is that you don't have to go very far off center
(5-10%) to have enough common mode to fry a really beefy choke when running
high power. High power is defined as 1-1.5 kW. You may, of course, be
luckier at 100 W.
Your mileage will vary ONLY to the extent that you run lower power and/or
lower duty cycles and/or a feed that is less off-center. We are stuck with
the laws of physics.
73, Jim K9YC
_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec
_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec
|