[Antennas] " G5RV " TYPE ANTENNA / "CUSHCRAFT MA5V"ANTENNA QUESTIONS

2002年7月01日 11:48:50 -0500

Pat is on the mark with his comment that the best coax length is "zero."
But, I have to take exception to his closing sentence which I may not be
interpreting the way he meant when he wrote it.
The G5RV or any center-fed wire antenna of any length will in general
possess a complex driving-point impedance at various frequencies in various
bands. The G5RV was designed specifically for 20-meter operation only, with
a little gain, to fit into Varney's back yard in the space available, and
to allow the use of some coax in the feedline but not too much as it was
expensive and hard to get post-WW2 in England. 
But why did Varney want to use coax at all? Probably because of the
convenience of bringing it into the shack and because he determined that on
20 meters (only) he could use a particular length of coax that presented an
input Z that his transmitter could live with.
So, he devised a feed system in which he first "remoted" the resonant
(approximately) impedance of the antenna on 20 meters (only) away from the
actual feedpoint by the length of a half-wave of balanced line feeder -
about 30 feet or so. This brought the feedpoint for the coax (a) nearer the
ground and (b) took the weight of the coax largely off the flat-top and its
supports. It, of course, shortened the coax required by some 30 feet or so.
But it did introduce the problem of feeding balanced line to a balanced
load with an unbalanced feedline.
And what is that resonant impedance of the flat-top on 20 meters (only)?
And why did he use balanced line? And why did he use the particular line -
twinlead or whatever - that he did?
About the same as for a half-wave dipole (50-75 ohms). Because of the
balanced load presented by the antenna. And because the line impedance
played no role in the operation of the system and it was cheap and easy to
obtain. A half-wave section intended to be an impedance repeater can have
any Zo - its only important attributes are length and Vp which determine
its electrical length.
It is important to note carefully that this half-wave section of line
serves no other purpose in Varney's design. It is *not* part of the
radiating system "folded back" on the lower bands where the flat-top is too
short. This is a widely held incorrect notion that is quickly dispelled if
one considers the direction of the currents in the line and notes that they
are out of phase as line currents have to be in order to avoid line
radiation.
The half-wave section does not have magical impedance transforming
properties on any band except 20 meters and there it serves only as an
impedance "repeater" or 1:1 transformer at that frequency only. On any
other band, it acts like any line of a given Zo, Vp and length: it is an
impedance transformer in addition to being an r-f extension cord.
Now, the question is how long should the feedline be if coax is not used?
And the answer, as almost always is the case, is "exactly long enough to go
from the antenna feedpoint to the tuner." And a tuner is necessary if there
is a requirement to provide the transmitter with very nearly a 50-ohm
resistive load. For some installations and feedline lengths, an input Z may
result which is beyond the range of the tuner. In that event a slight
change in line length usually brings Zin to a value the tuner can handle.
In that sense and that sense only is there any significance to the
ladderline length used with a G5RV.
The G5RV with its half-wave repeater section of line connected to coax of
the "proper length" presents an overall transmission line system that
manages to take the 50-75 ohm input Z of the antenna on 20 meters and
produce something in that neighborhood at the input to the coax for the
transmitter. In some cases, the Zin of the coax is appropriate for the
transmitter and everything works well.
But, away from 20 meters - or even at the band edges - the input Z of the
coax can take on values that some transmitters may not accept and a
transmatch or tuner is required, *or* the coax length must be trimmed until
a "magic length" is found for that particular installation and that band.
The easy way out of all this and the ability to operate on bands other than
20 comes with following Varney's advice: use ladderline or twinlead or
whatever all the way from the antenna to a tuner and leave off the coax.
This not only avoids the added loss in the coax on bands other than 20
meters due to a possibly high SWR but it also removes any feedline length
consideration since the tuner can usually accommodate a reasonable range of
line input Z values over the various bands. And this is Pat's point, very
well stated.
The only problem I have with his second sentence is that if coax *is* used,
for whatever reason, there may be some lengths on some bands that produce
an input Z that the transmitter will not accept and either a tuner must be
used or the coax length altered. Thus, one may find that there really is a
magic length on some band(s). But almost always the use of a tuner gets
around that issue and allows a convenient coax length to be used.
Pat also makes the statement that can never be repeated enough: a low SWR
is meaningless as an indicator of either how well an antenna radiates or
how efficient it is. An efficient antenna that radiates well can have a
high SWR on the feed system or can equally well have a low SWR. SWR alone
is not an indicator of either efficiency or of attaining desired radiation
performance.
So, kudos to Pat for his posting and I ask his forebearance for my muddling
a bit in the details.
I wish that it were possible to dispel the mythology that has grown around
the G5RV antenna. The facts: it is nothing more than a 3/2-wavelength
flat-top wire on 20 meters that shows resonance within that band and has a
small gain over a dipole due to its increased length. It will operate on
bands other than 20 meters, as will almost any other length of wire in the
same location. But its length has no magical property other than on 20
meters. Its complex feedline structure of twinlead or ladderline plus coax
has no magical properties that make the antenna radiate any better or with
greater efficiency than any other 102 ft piece of wire fed the same amount
of power. The complexity of its feedline system is basically unwarranted
either in terms of efficiency or convenience.
While I have no personal agenda against the G5RV, as one interested in
antennas and transmission lines for a lifetime, I find it regrettable to
see the same half-truths and misstatements continue to be repeated year
after year. A similar cult effect attached itself to the Coax Dipole,
although that one has pretty well surrendered to the analysis of Walt
Maxwell, W2DU, and others who have demonstrated that its apparent (and
minimal) bandwidth improvement is largely due to the added loss of the
coax.
Incidently, there is now available an excellent transmission line analysis
program available for the downloading called TLDetails. It can be used to
analyze the classic G5RV system - along with the EZNEC program - to arrive
at quantitative results that show just what is happening in the G5RV feed
system. Go to
			http://www.qsl.net/ac6la
for a free download. This is clearly one of the better programs in the this
area and I highly recommend it for simplicity of use and clarity of
results.
73/72/oo, George W5YR - the Yellow Rose of Texas 
Fairview, TX 30 mi NE of Dallas in Collin county EM13qe 
Amateur Radio W5YR, in the 56th year and it just keeps getting better!
QRP-L 1373 NETXQRP 6 SOC 262 COG 8 FPQRP 404 TEN-X 11771 I-LINK 11735
Icom IC-756PRO #02121 Kachina 505 DSP #91900556 Icom IC-765 #02437
Pat W wrote:
>> The G5RV optimum performance occurs when the coax
> length is zero!
>> Contrary to some misleading opinions, there is NO
> magic coax length, other than zero. Beware that the
> longer the coax, the greater is the loss and, the lower
> the is SWR!(If the coax is long enough, the SWR will be
> 1:1 from DC to daylight). SO, some think that because
> the SWR is reasonable on some bands the system is
> efficient when it's not. The only way to determine the
> tuning range is to use multiples of 1/2 wavelength of
> ladder line(or other open wire line).