It's disappointing to see that people have gotten so many different
results modeling such a simple antenna with EZNEC.

Jerry's results are correct. You'll get slight variations with differing
segmentation and ground conductivity, by they won't be large differences.

I haven't followed through the article's math, but there's either a
computational error or something fundamentally wrong with the equations
to produce a value of j1776 for the antenna's reactance. The fact that
the author's reactance is just about twice what it should be points to a
likely error in computation.

Jerry, what the EZNEC result means is that you'd need a *total* of j881
ohms at the feedpoint to resonate the antenna. You could do this by
adding half the amount to each leg, or the total to one leg. Or, you
could move the coils out toward the end, but you'd then need more
inductance. Of course, you've still got a feedpoint resistance of about
11 ohms, plus coil resistance, to transform into something close enough
to 50 ohms to make your rig happy.

Roy Lewallen, W7EL

Jerry wrote:
October's QST has an article "Designing a Shortened Antenna" pp 28-32. It
gives an example of a shortened dipole for 40 meters at 7007 khz. Dipole
length of the half-sized dipole [p30 "a second example"] is 10.61 meters.
This is 20 feet off ground using #12 wire. The formulas give a solution of
XL= +j1776 ohms or an inductance of 40 microhenries at 30 degrees from each
leg.


I tried to simulate this antenna on EZNEC. A 10.61 meter antenna at 7007 khz
gives impedance =11.33 - j 881.2 ohms.

If EZNEC is correct wouldn t I need an inductance of +j881.2 on each leg of
the dipole, rather than +j1776?

By coincidence, one of the tutorial examples for the MultiNEC program
deals with the subject of loading coils on a shorty dipole. With
MultiNEC you can define the antenna model using variables for both the
amount of reactance in the coils and for the placement of the coils
along the wire. Each time you change the variable that controls the
placement of the coil you can use the MultiNEC "Resonate" function
which will automatically invoke EZNEC in an iterative manner and
adjust the reactance of the coils until the antenna is resonant.

The example was written to illustrate and explain various capabilities
of MultiNEC, of course, and not to address the particular scenario of
this thread. It's available in standalone form at
www.qsl.net/ac6la/mnex3.html.

Dan

For given location of loading coil along a short antenna wire, of given
overall length, of given wire, rod or mast diameter, the following programs
will calculate the inductance needed to resonate it to 1/4-wavelength.


Two loaded wires connected back-to-back form a 1/2-wave resonant dipole.


Program ADDALOAD will tell just that. It will also tell you the value and
position of a loading capacitance if antenna length requires a resonating
capacitor.


Program LOADCOIL will do likewise and also tell you how many turns of what
gauge wire to wind on a coil of given former length and diameter.


Program VERTLOAD does a similar job insofar as dipoles are concerned.


Program MIDLOAD is dedicated to very short centre-loaded dipoles. All coil
and wire dimensions and number of turns are taken into account. Such
antennas can be fed via either coax or balanced-pair feedlines of any
impedance.


Program SOLNOID2 will design any coil of any proportions for any given
dimensions and required inductance, plus a variety of other performance
characteristics including temperature rise for given applied RF volts. Nice
for tank coils, loading coils and antenna traps.


All these programs are used in a "What-if " user-friendly mode. USA
citizens may be at a slight disadvantage - metric measurement units are
used although in some cases a translation is provided. Once one of these
simple loaded anennas has been designed, provided you have a BIG,
UNOBSTRUCTED back-yard, EZNEC may be used to estimate its receiving and
radiating pattern. Otherwise your guess is as good as mine.


However, it should be borne in mind, a coil-loaded short antenna is
essentially a single-band device and a tuner is nearly always needed even
for the single band. With a tuner it may be goaded into doing something
useful on one or two higher frequency bands.


For optimum efficiency the location of a loading coil is about 2/3 or 3/4 of
the way along the wire length. The bigger the physical size of the loading
coil, the higher the Q, the smaller the coil loss and the nearer it can be
placed towards the end of the wire. But location is VERY non-critical.


If the wire is not very much shorter than self-resonance then efficiency can
be maximised by NOT inserting a coil in the wire and depending entirely on a
high-impedance balanced transmission line and a tuner. Without a
high-in-the-air loading coil the antenna reverts advantageously to multiband
operation.


In general, good efficiency is possible with a 450 or 600-ohm feed-line
provided the overall length of an unloaded dipole is not less than 0.3
wavelengths at the lowest frequency of interest.


One antenna with its feedline, from coax to open wire, or both types in
tandem, can be compared efficiency-wise with another by using program
DIPOLE3.


All the above small programs, one self-contained file only, can be
downloaded in a few seconds from the following website and run immediately.
No training course needed. Free to USA citizens. ;o)

----
=======================
Regards from Reg, G4FGQ
For Free Radio Design Software
made in the original Birmingham,
go to http://www.g4fgq.com
=======================