[Elecraft] Bal[un] (was: Buddipole vs. tuner and wire)

Tue May 13 13:12:43 EDT 2008

It would really help if we could drop the incorrect "balanced/unbalanced"
nomenclature, but it's ingrained in Ham lore and it causes
misunderstandings. 

A coaxial RF output to an antenna is *balanced*: the currents flowing out
into the coax are as well balanced (perhaps better balanced) than those
flowing in any "open wire" or parallel wire feedline. 

The real difference is that a so-called "balanced" output is isolated from
the rig ground. It does not mean it's inherently any more "balanced" than
the coaxial connector output. 

So what's the big deal? When feeding a balanced load such as a loop or
center fed wire antenna, it's easier to preserve the balance of currents
when using an RF output that is isolated from the rig ground.

Of course, preserving that balance does a couple of good things:

1) When using a parallel wire feedline (so-called "open wire", ribbon or
similar feedline), isolating the RF from the rig ground helps ensure that
all the RF is flowing in the two wires to minimize feedline radiation (or
pickup). Parallel wire lines with balanced currents do not produce an RF
field since the fields produced the wires are opposite in phase and equal in
amplitude, so there is no net RF field produced around the feeder. All the
RF energy is kept in the feed line to be moved to the load, whether it's
energy picked up to be fed to the receiver or energy from a transmitter to
be moved to the antenna. 

2) A parallel wire feedline maintains the intended RF path at the antenna
without complications at the feed point that are found using coax. Consider
the center fed wire, fed with a coaxial line. RF flowing along the center
conductor of a coaxial line flows out along one wire, while the RF flowing
along the inside of the shield flows out along the other wire. But that
shield has an inside and an outside. For RF, the inside and outside of the
shield are *different* conductors since the RF flows only along the very
surface of a conductor, thanks to skin effect. In other words, the inside
and outside surface of the shield on a coax looks like two completely
different and isolated conductors to the RF. When the RF reaches the edge of
the shield where it is connected to one side of a center fed wire, the RF is
free to flow along the antenna wire *and* flow back along the outside of the
coax shield. That means that you have two conductors connected to one side
of your 'center fed wire' - one is the intended antenna wire and the other
is the outside of the coax shield. 

In most cases that current on the outside of the coax shield is not a
problem. However, it can be an issue if the electrical length of the
feedline produces a voltage loop (high impedance point) at the rig. Properly
connected to the rig, a coaxial line shield is unbroken, so the outside of
the coax is contiguous with the outside of the metal enclosure on the rig.
The currents flowing on the inside surface of the rig flow out only onto the
inside surface of the coax. But, if there is RF on the *outside* of the coax
it will flow onto the outside of the rig. If the rig is at a voltage loop
and if the rig doesn't have a good *RF* ground, the rig can appear "hot"
with RF. 

Even if you put a balun at the antenna end to stop RF currents from flowing
back along the outside of the coax, you can end up with RF on the outside of
the shield and so at the rig. That's because the coaxial line is a conductor
very close to the radiator. In spite of a choke balun at the feed point, the
RF field around antenna easily can induce significant RF currents on the
outside of the coax. Those currents can flow down the outside of the coax to
the outside of the rig. In that case, a choke balun at the rig can help.

So what's the big deal with parallel wire feedlines, if coaxial lines offer
balanced currents to the antenna as well? In a word: impedance. Practical
coaxial lines have a fairly low impedance, where practical parallel wire
lines have a fairly high impedance. A coaxial line with an impedance of,
say, 400 ohms, requires a rather large diameter - usually several feet!(The
impedance is based on a ratio between the diameter of the shield and the
center conductor) Similarly, a parallel wire line with an impedance of, say
50 ohms, requires very close spacing of the wires. 

Impedance is important to minimize the SWR extremes, hence the resistive
losses, in a system that might show a wide range of impedances to the
feedline, such as a center fed wire used on several bands. A center fed wire
antenna (doublet) might show something from about 50 ohms where it's 1/2
wavelength long, up to maybe 2,000 ohms or so where it's 1 wavelength long.
A 400 ohm feed line will show an SWR range of about 5:1 over that entire
range. However, a 50 ohm coax line will show an SWR of about 40:1 when
feeding an antenna showing a 2000 ohm impedance at the feed point. That 40:1
means very large RF currents at the current loops along the feed line, hence
large resistive losses. 

You are quite right, Brian, a tuner at the center of the wire that provides
a good 50 ohm match to coax on all bands would be an excellent antenna,
keeping the efficiency and convenience of the coaxial line and help isolate
it from currents flowing out around the end of the coax to avoid "RF in the
shack" issues.

But such tuners are hard to come by, need to be weatherproofed and require
support. So, when a significant length of feed line is needed on a simple
wire antenna used on multiple bands, or any time we're going for absolute
maximum efficiency with such an antenna, higher-impedance open wire line is
usually the choice and the matching system (tuner) is placed in the shack,
or at least in a more convenient location near the ground where it can be
adjusted as needed. 

However, in a portable situation with, typically, very short feed lines,
coaxial line might have acceptably low loss, although it's not likely to
protect the rig from being "hot" with RF on some bands unless a balun is
used to stop currents on the outside of the line. 

If you simply connect parallel wire feeders to the rig, one to the center
pin of the coax RF output and the other to the case, you'll likely find you
have excellent results and even lower losses than you'll get from coax. Such
installations work FB in fixed stations too for many operators. The issue is
that the case of the rig is not isolated from the parallel wire feedline so,
if the feeder is looking into a very high impedance, the outside of the rig
may be *hot* with RF since it's part of one of the feeder wires and some of
the RF that should be flowing in one wire may be finding an alternative path
through the rig, power supply wires, etc. With a small portable battery
powered rig that's often not a problem, but it's why baluns at the rig are
so commonly used to feed parallel wire lines. The balun provides the desired
isolation, but at a cost. Many baluns can show high losses when operated at
high SWRs. It's hard to predict although, typically, choke type baluns are
the best in that regard. 

My basic rule is to avoid any component in the antenna system that doesn't
have a proven benefit, since no component is 100% efficient. That means
there is always some RF lost whenever we put anything in the line. The
question is "how much" and, often, that question has no clear answer.  

Ron AC7AC

-----Original Message-----

OK, I am convinced - keep the tuner and the wire. So, that brings up a  
second question: balanced vs. unbalanced.

Wire antennas that do not require a counterpoise are usually loops or  
dipoles, i.e. inherently balanced even if not resonant. Most tuners  
offer an unbalanced output. SGC says, just connect up the antenna.  
Seems to me that a balun would improve things and keep RF off the coax  
and power leads to the tuner. But how well do baluns handle huge  
mismatch?

Just off the top of my head, it seems to me that the tuner at the top  
of a pole with two legs sloped down, i.e. inverted-V, would make a  
pretty good omni all-band antenna.

More thinking aloud.

Brian Lloyd
Granite Bay Montessori School          9330 Sierra College Bl
brian AT gbmontessori DOT com          Roseville, CA 95661
+1.916.367.2131 (voice)                +1.791.912.8170 (fax)

PGP key ID:          12095C52A32A1B6C
PGP key fingerprint: 3B1D BA11 4913 3254 B6E0  CC09 1209 5C52 A32A 1B6C

_______________________________________________
Elecraft mailing list
Post to: Elecraft at mailman.qth.net
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft    

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com