Carolina Windom revisited: 4 to 1 balun does nothing to choke RF ?
john Wiener
ask a question to which I THINK I already know the answer.
This particular version of the CW OCF dipole uses 300 ohm twin lead
feedline terminated after 33 feet to a 4:1 balun. Richard warned of
significant risk of RF on the outer (inner?) braid of the coax.
I've never read of using a 1:1 unun right after a 4:1 balun to minimize
RF on the outer coax.
My thinking is that the 4:1 balun acts as a voltage type and will do
nothing to ameliorate this. So, I will put some ferrite rings or snap
on ferrites just past the 4:1 balun on the coax.
Does this sound like a reasonable solution?
John
AB8O
John Smith
> ...
> I've never read of using a 1:1 unun right after a 4:1 balun to minimize
> RF on the outer coax.
>
> My thinking is that the 4:1 balun acts as a voltage type and will do
> nothing to ameliorate this. So, I will put some ferrite rings or snap
> on ferrites just past the 4:1 balun on the coax.
>
>
> Does this sound like a reasonable solution?
>
> John
> AB8O
If you use a 4:1 Ruthroff, it would be a voltage balun ... if you use a
4:1 Guanella it would be a current balun ... the 1:1 current balun is
probably more useful behind a Ruthroff ... but hey, once you have tried
all these possible combinations, you can speak from experience! <grin>
If going from 300 ohm to 50 ohm, perhaps you would choose a 6:1. Or,
just go with the 4:1 now and when you need some diversion, later, try
the 6:1 to see what improvments can be had and if the loss in this
design is acceptable to you ... etc.
Anyway, in this document is a 6:1 (actually 6.25:1, resulting in 312ohm
to 50 ohm) made from two 4:1 baluns (I would think Guanella ununs ...
the 4:1 can be made from two 1:1, each wound on the opposite side of
toroid core, reversing coil directions on one side. This could also be
accomplished with 4 ferrite rods ... a 1:1 balun wound on each rod, two
rods combined to make a 4:1 balun, then these "two units" combined to
construct the 6.25:1 balun ...
Regards,
JS
John Smith
[stuff and forgot the URL, as usual :-( ]
The URL for the 6.25:1 ...
http://www.radioelectronicschool.net/files/downloads/ocfdipole.pdf
Regards,
JS
Roy Lewallen
A while back I did some pretty careful measurements of an OCF dipole. I
found that ferrites were required at both the feedpoint and at one or
more places along the feedline. The ferrites at the feedpoint suppress
the conducted common mode current (which is actually forced to exist by
the voltage balun). But the asymmetry of the antenna results in common
mode current being induced onto the feedline by mutual coupling to the
antenna. This isn't a problem in a symmetrical dipole if the feedline is
positioned symmetrically relative to the antenna, since the currents
induced by the two equal halves cancel. But the OCF dipole can result in
quite a lot of induced common mode current.
Ideally, you'd put at least a second bunch of snap on cores about a
quarter wavelength from the feedpoint. But one of the main reasons
people use OCFs is for multi-band operation. So the thing to do is to
place the cores for maximum effectiveness on the band(s) where you have
the most trouble -- the common mode current also depends on the feedline
length and position, and will vary considerably from band to band even
if you do nothing.
My opinion is that users of OCF dipoles are just about always going to
have to deal with some amount of common mode current, and the best you
can do is reduce it to a level you can tolerate.
Roy Lewallen, W7EL
Tam
"Roy Lewallen" <w7...@eznec.com> wrote in message
news:RqudnSt_O9o6eErV...@posted.easystreetonline...
Roy,
What's your opinion on the 4: or 6:1 balun between the 300 Ohm line and the
coax? I see no reason whatever to think that the impedance coming off the
300 Ohm line is anywhere near 300 Ohms. Also, where is it written that a 50
Ohm balun will work at, say, 2000 Ohms. The ferrites as you suggest will
clearly work if you use enough of them.
The reason for asking this is that a friend is in the process of putting up
a 75 m dipole, which he only plans to use on 75 m. Everybody is telling him
to feed it with ladder line going to coax through a balun. Why in the world
would you do that?
Tam/WB2TT
john Wiener
Cecil Moore
> My opinion is that users of OCF dipoles are just about always going to
Isn't the section between the voltage balun and the choke
designed to radiate?
--
73, Cecil http://www.w5dxp.com
"According to the general theory of relativity,
space without ether is unthinkable." Albert Einstein
john Wiener
> Roy Lewallen wrote:
>> My opinion is that users of OCF dipoles are just about always going to
>> have to deal with some amount of common mode current, ...
>
> Isn't the section between the voltage balun and the choke
> designed to radiate?
In the version I am putting up, the 4:1 balun is at the bottom of the
twin lead feedline, so the feedline can radiate.
John
AB8O
lt...@yahoo.com
- 'Doc
Cecil Moore
It is alleged to be a multi-band antenna. Assuming that
the standing-wave current is 90 degrees out of phase
with the standing wave voltage, one can actually calculate
an approximation for the feedpoint impedance by dividing
the standing-wave voltage by the standing-wave current at
the feedpoint. With an OCF, the feedpoint impedance will
always be lower than Vmax/Imin and always higher than
Vmin/Imax and thus provide a reasonable SWR for 450 ohm
ladder-line.
The free demo version of EZNEC available from
http://www.eznec.com
will provide a reasonable guess at the feedpoint impedance.
Here are what EZNEC 4.0 says about the
feedpoint impedance of the following:
------------83'----------FP-----47'--------
3.8 MHz, 85.5+j98.5 ohms, SWR(300)= 3.9, SWR(450)= 5.5
7.2 MHz, 177-j181 ohms, SWR(300)= 2.5, SWR(450)= 3
10.125 MHz, 3569-j102 ohms, SWR(300)= 12, SWR(450)= 7.9
14.2 MHz, 96-j357 ohms, SWR(300)= 7.7, SWR(450)= 7.7
18.14 MHz, 797-j911 ohms, SWR(300)= 6.3, SWR(450)= 4.4
21.3 MHz, 933-j1239 ohms, SWR(300)= 8.8, SWR(450)= 6
24.95 MHz, 218-j652 ohms, SWR(300)= 8.5, SWR(450)= 6.7
28.4 MHz, 757+j1019 ohms, SWR(300)= 7.4, SWR(450)= 5.1
As you can see, the myth that this antenna has a 300
ohm feedpoint impedance is just an old wives' tale but
it does have reasonable SWRs when fed with 450 ohm
ladder-line.
lt...@yahoo.com
I sort of thought that this OCF antenna was typically used for a
multiband antenna, that's typically the 'idea' behind using them.
Keeping that 'thought' in mind, how can you 'optimize' this multiband
antenna for one particular band without 'de-optimizing' it for
others? That seems sort of going at it from the wrong direction, why
not use an antenna that works well for the particular band of
interest. Oh, I know that assumes that you can have more than one
antenna, which isn't always the case. So having an 'un-optimized'
antenna is what you'll end up with in most (if not all) bands except
for one. Which is the 'fault' for almost all multiband antennas.
Isn't it?
- 'Do
Cecil Moore
> I sort of thought that this OCF antenna was typically used for a
> multiband antenna, that's typically the 'idea' behind using them.
> Keeping that 'thought' in mind, how can you 'optimize' this multiband
> antenna for one particular band without 'de-optimizing' it for
> others?
Given the SWRs that EZNEC predicts, feeding the OCF
at a current maximum point causes the XMTR (or tuner)
to see the following resistances looking into the
ladder-line.
3.8 MHz, 129 ohms; 7.2 MHz, 120 ohms; 10.125 MHz, 37.5 ohms;
14.2 MHz, 58 ohms; 18.14 MHz, 71 ohms; 21.3 MHz, 51 ohms;
24.95 MHz, 53 ohms; 28.4 MHz, 61 ohms
Let's say you wanted to optimize for 40m. You would
choose a 450 ohm ladder-line length of 67.5 feet and
expect 120 ohms looking into the ladder-line. Use a
tuner or transformer for a perfect match.
> Oh, I know that assumes that you can have more than one
> antenna, which isn't always the case. So having an 'un-optimized'
> antenna is what you'll end up with in most (if not all) bands except
> for one. Which is the 'fault' for almost all multiband antennas.
> Isn't it?
Not for this one: http://www.w5dxp.com/notuner.htm
That antenna system is efficient on all HF bands.
It is fed with 1/2WL of ladder-line on 80m and 3/4WL
of ladder-line on 40m. Also appropriate lengths of
ladder-line on all other HF bands. The ladder-line
length averages around 100 feet.
JB
<lt...@yahoo.com> wrote in message
news:f562e1ab-870f-4d41...@r66g2000hsg.googlegroups.com...
Seems to me it's not "optimized" for any band, but is just a fudged
radiator fed by ladder line, matched with a tuner or maybe a pi-net without
too much line loss. You can fiddle with it to your hearts content, but in
the end will it load up where you want it to and will you get out? It would
be nice to have an antenna analyzer to have some indication of YOUR
installation as you fiddle. Obviously you will have trouble using any of
the modern internal ATUs.
Years ago, I threw one up but fed it with coax. It didn't work so I rebuilt
it as parallel fed Dipoles for 80 and 40m.
Thanks for the NEC output, Cecil.
Roy Lewallen
>
> Roy,
> What's your opinion on the 4: or 6:1 balun between the 300 Ohm line and
> the coax? I see no reason whatever to think that the impedance coming
> off the 300 Ohm line is anywhere near 300 Ohms. Also, where is it
> written that a 50 Ohm balun will work at, say, 2000 Ohms. The ferrites
> as you suggest will clearly work if you use enough of them.
>
> The reason for asking this is that a friend is in the process of putting
> up a 75 m dipole, which he only plans to use on 75 m. Everybody is
> telling him to feed it with ladder line going to coax through a balun.
> Why in the world would you do that?
As you suspect, the impedances encountered by the transformer on some
bands are wildly different than its nominal design impedances. In the
one which I carefully measured, the result was no surprise. When the
antenna impedance was substantially different from 300 + j0, the
transformation ratio wasn't 6:1, and the transformer added series and/or
shunt reactance, sometimes a pretty large amount. And this was the case
on most bands.
This isn't to say that an OCF dipole can't be fiddled until, radiating
feedline and all, it manages to present an acceptable SWR on several
bands. But when it does, it's not working at all like predicted by a
simplified analysis which ignores the strong feedline coupling and very
non-ideal transformer effects.
Roy Lewallen, W7EL
John Smith
> ...
> This isn't to say that an OCF dipole can't be fiddled until, radiating
> feedline and all, it manages to present an acceptable SWR on several
> bands. But when it does, it's not working at all like predicted by a
> simplified analysis which ignores the strong feedline coupling and very
> non-ideal transformer effects.
>
> Roy Lewallen, W7EL
I have switched my balun designs to the one in this URL, Figure 4 -
Improved 4:1 Current Balun, page 3:
http://home.earthlink.net/~christrask/Trask4to1Balun.pdf
Can be used balanced, or forced to unun fashion ... bandwidth becomes
increased (and, it "just works better for me!")
But, what works for you is the most important, always ...
Regards,
JS
Cecil Moore
> Thanks for the NEC output, Cecil.
I showed how to optimize an OCF for a single band, e.g.
40m. If the 50/120 ohm transformer is link coupled, it will
cause a common-mode current node at the transformer on 40m.
Any antenna system can be optimized if one knows what they
are doing.
For instance, a shunt 1000 pf cap at the twinlead to coax
junction will optimize a G5RV for 75m.
Cecil Moore
> This isn't to say that an OCF dipole can't be fiddled until, radiating
> feedline and all, it manages to present an acceptable SWR on several
> bands.
I bought the 300 ohm feedpoint myth when I was at Texas A&M
in 1958 when I didn't know any better. I had an OCF fed with
300 ohm twinlead fed through a 6:1 air-core Heathkit balun
driven by a DX-40. The results were amazing to me at the
time. Now I know the pi-net output of the DX-40 would achieve
a match to almost anything and I should have been using a 1:1
balun. The Heathkit balun didn't have much loss and the 300
ohm twinlead didn't have much loss. I have no idea what the
actual impedances were, but losses were minimized and the
system successfully worked the world at the height of the
most active sunspot cycle in recorded history.
JB
"Cecil Moore" <nos...@w5dxp.com> wrote in message
news:DlcCk.872$Ei...@flpi143.ffdc.sbc.com...
> JB wrote:
> > Thanks for the NEC output, Cecil.
>
> I showed how to optimize an OCF for a single band, e.g.
> 40m. If the 50/120 ohm transformer is link coupled, it will
> cause a common-mode current node at the transformer on 40m.
> Any antenna system can be optimized if one knows what they
> are doing.
>
> For instance, a shunt 1000 pf cap at the twinlead to coax
> junction will optimize a G5RV for 75m.
Of course but I'm too lazy to run the numbers if it isn't my project. My
favorite single band wire antenna is the Dipole. How is OCF an improvement?
I also liked my Inverted L made out of a homebrew 9' bug catcher mobile
antenna with 25' horizontal clip on extension for 40/80 and clip on shunt
caps at the mobile mount. Works well in the campground.
Keeping the antenna matched to the coax DOES make a difference, considering
coax losses.
John Smith
> ...
>
> Of course but I'm too lazy to run the numbers if it isn't my project. My
> favorite single band wire antenna is the Dipole. How is OCF an improvement?
> I also liked my Inverted L made out of a homebrew 9' bug catcher mobile
> antenna with 25' horizontal clip on extension for 40/80 and clip on shunt
> caps at the mobile mount. Works well in the campground.
>
> Keeping the antenna matched to the coax DOES make a difference, considering
> coax losses.
>
A "one size fits all" antenna is still VERY MUCH a pipe dream ... in my
humble opinion (or, IMHO ...)
Regards,
JS
Tam
"John Smith" <assembl...@gmail.com> wrote in message
news:gbceob$4o1$1...@news.albasani.net...
At least the "one size" will be very large. I once saw a truly monumental
commercial or military log periodic.
Tam/WB2TT
JB
> > I also liked my Inverted L made out of a homebrew 9' bug catcher mobile
> > antenna with 25' horizontal clip on extension for 40/80 and clip on
shunt
> > caps at the mobile mount. Works well in the campground.
> >
> > Keeping the antenna matched to the coax DOES make a difference,
considering
> > coax losses.
> >
>
> A "one size fits all" antenna is still VERY MUCH a pipe dream ... in my
> humble opinion (or, IMHO ...)
>
> Regards,
> JS
Then there are the discones and LPDAs where size matters.
The bug catcher worked well and still does despite many repairs, 15-40 m by
itself. 40 and 80 with the extension. It's a modified hamstick with 4"
dia. by 8" hand wound #14 coil in the center with scrap plastic forms.
There must be an unwritten law that says only the ugliest projects will
work. It was a pipe dream of flea clips, but the goal was to do all
matching on the antenna.
I used it mobile with a 365 pf broadcast variable on the mount and a remote
operator into the cab. Add a little series XsubL with a flea clip, matched
out with the cap to raise the impedance at the mount. The coil would self
resonate so 12 and 17 wouldn't work (perhaps a longer stepped pitch later).
Shorting around the coil got me 10m back. I would tune for min SWR
(actually ALC) and the difference between not heard with an OK SWR and heard
well after a fine tuning was repeatable over and over. Even with as little
as 15' of RG8x. Worked a lot of DX on the commute back and forth to
school. Had no time to do it any where else.