[nav0239]

Say,

EX 0 1 2 ...

means a voltage source is applied to wire #1 segment #2. However,

EX 1 1 2 ...

based on the user manual, would mean a linear planar incident wave,
coming from 1 degree theta and 2 degree phi etc.

There is nowhere you can specify the feed point any more.

 

[tripelo]

dB, dBi, and dBd

...the "Pattern" window will reveal that the gain(s) are in dbi (db by itself is simply a ratio without a reference base)...

Sometimes these terms are confused. Not implying that you (jbkeh) in particular, have them confused.

............................................................

Realizing that dB, dBi, and dBd have been discussed several times in other threads:
​

Both, dB and dBi are similarly derived from ratios.

The term "dB" is not without a reference base. A reference necessarily exists and for the term to be meaningful must be known by context of the usage.

Yes, NEC correctly states that results are "dBi" for a given antenna. This indicates that antenna gain was calculated as gain over over an isotropic antenna.

Once one knows that 'Antenna A' has 10 dBi gain over an isotropic antenna, and 'Antenna B' has 8 dBi gain over an isotropic antenna, then 'Antenna A' has 2 dB (not dBi) gain over 'Antenna B'.

If one compared an antenna gain measured as dBi to an antenna measured as dBd, the comparison units could not reasonably be either dBi or dBd, they are simply dB. But first, to compare those two antennas, one must either add 2.15 dB to the dBd measurement, or subtract 2.15 dB from the dBi measurement before a valid comparison can be made.

(in common usage, 0 dBd is considered equal to 2.15 dBi)
​

For both nomenclatures: dBi and dBd, the units themselves are dB units.

As someone in a recent post said: "a dB is a dB".

............................................

To reiterate:
​

The "i" in dBi reveals the reference antenna as an isotropic.

The "d" in dBd reveals the reference as a dipole antenna (commonly a resonant or tuned half-wave dipole).

When comparing antenna gains to each other, the units are not dBi or dBd they are simply dB, regardless of whether the reference for each of the measured (or computed) gain values were isotropic or dipole.

In reality, outside of simulations, measurements are made with instruments that measure in units such as milliVolts, microVolts, or milliWatts, with an implied reference of zero (0 milliVolts, 0 microVolts, or 0 milliWatts). In these cases, it is usually not necessary to state the measurement reference (of zero whatever units). These measurements could be converted in the instrument to indicate dBmV, dBuV, or dBm. After scaling the units to a common reference, when one compares measurements to each other, the units are correctly stated as dB to each other (regardless of how they were originally measured).

Conceivably, measurements could be made with reference to any arbitrary value: for example, instead of zero Volts, a Voltage measurement could be made with reference to one Volt. In this case to be clear, the measurement reference would be included with the measurement. Regardless of the reference, the measurement units remain unchanged as Volts.

Outside of antenna work, in acoustics for example. One might determine that a certain sound is louder than another sound by say 6 dB. Add some extra reference; such as saying a sound is 6 dBm louder or 6 dBV louder or 6 dB(anything) louder, then the statement becomes imprecise & unclear. The certain sound is simply 6 dB louder (than the other specific sound).

..........................................

The 'B' in 'dB' is capitalized because the unit bel was named in honor of Alexander Graham Bell, the inventor of the telephone.

A decibel is a tenth of a bel.
.

 

[holl_ands]

RECIPROCITY may help in understanding Total, Horiz. and Vert. Gain terms.
Consider applying power to an antenna, such as the Feedpoint at the bottom of a simple LOOP:
http://imageevent.com/holl_ands/loops/uhfcircular
4nec2 plots are shown for what the antenna would RADIATE with Total, Horizontal and Vertical
polarization. Note that Total is the power sum of H and V.

By RECIPROCITY, if you were to use the same antenna to RECEIVE from an antenna, the 4nec2
plots describe the Gain (re to Isotropic, same in all directions) you would measure if you moved
a theoretical Isotropic transmitting antenna all around the receive antenna.
For 4nec2's usual "EX 0" option, clearly it assumes that this antenna is radiating BOTH Horizontal
and Vertical polarizations, hence the displayed results.....however, since it does NOT
show any difference wrt Left & Right Hand Polarized antennas, I would assume that the H and V
components are IN-PHASE, rather than shifted 90-degrees as found in Polarized signals....for that,
"EX 2" and "EX3" options are available. [Coming up on my "TO DO" list.]

The power (voltage) going into the antenna doesn't know anything about polarization....
But as shown in 4nec2's 3D Pattern charts for a Folded Dipole, the highest Gain lobe is in the
Forward direction with Horizontal Polarization (and equally in Reverse, although hard to see the rear lobe).
Note that the Dipole's "Donut" shape doesn't appear until after summing the H and V components:
http://imageevent.com/holl_ands/loops/folded [See slides 22-24.]

 

[holl_ands]

FYI: In a Circularly Polarized antenna, both antennas must be either Right or Left Hand Polarized.
If a LHP antenna receives a RHP signal, the opposing 90-deg phase shift in the receive antenna
results in CANCELLATION, rather than summing the two signals. This behavior is useful to suppress
Multipath, since Circularly Polarized signals change rotation type when they reflect off flat surfaces.

Further thoughts re dBi vs dBd and dBc:

dBi is Gain relative to a theoretical (non-realizable) Isotropic antenna which would radiate same in all directions.

dBd is Gain relative to a (realizable) Dipole antenna, which has nulls along the rod axis.

However, I've always seen dBd used as a Gain Normalization factor relative to the FORWARD direction,
hence an azimuthal plot marked in "dBd" units does NOT mean relative to the azimuthal plot of a Dipole
at each and every azimuth!!! Can you imagine reading specs for say a simple Omni (e.g. Turnstile)
antenna, seeing the azimuthal Gain plot blow up to infinity, corresponding to the reference Dipole's
nulls along the rod axis.....YIKES!!!!! Clearly, all of the Gain plots we see are relative to Gain in
the FORWARD direction, which may be normalized to ZERO, or more typically dBi (4nec2) or
(as found in W-G and C-M spec sheets) normalized to dBd (= dbi + 2.15dB).

When the azimuthal Gain is measured at an antenna range, an antenna with a known Gain
is aimed at the antenna Unit-Under-Test. The received power level is then plotted as the UUT
is rotated. A Dipole is substituted for the UUT to calibrate the measured Gain (hence origin of "dBd").
To measure Horizontal Gain, the known transmitting antenna would need to be oriented Horizontally
and designed with minimal Vertical Gain (e.g. Dipole, Yagi or LPDA). To measure Vertical Gain,
the known antenna would be rotated around it's axis for Vertical radiation.

I have yet to run across a definitive description for "dBc" other than it is Gain relative to a Circularly
Polarized antenna. Akin to the dBd definition, it appears that this reference is a simple set of
Crossed Dipoles (driven 90-degrees out of phase, such as by Quarter-Wavelength Feedlines,
Quarter-Wavelength spacing of the Dipoles....or electrically separate, offset phase, H & V Transmitters):
http://books.google.com/books?id=v-...DA&sa=X&oi=book_result&ct=result&resnum=4&ved=0CDkQ6AEwAzgK#v=onepage&q&f=false
http://sv1bsx.50webs.com/antenna-pol/polarization.html

BTW: I've more commonly seen "dBc" defined relative to the power in a signal's CARRIER, such
as the transmitted intermodulation noise floor relative to the carrier power.
I've searched all of my antenna books and a couple dozen via Google Books....NONE of them
define dBc for antennas, nor is the reference antenna described....has anyone else seen it????

 

[jbkeh]

My "understanding?" of RECIPROCITY is that an antenna will exhibit the same gain characteristics whether it is receiving or transmitting. Thus, one can evaluate the transmission characteristics (both the horizontally polarized component and the vertically polarized component) and apply the results to the receiving mode.

But we cannot be "blind" to the fact that when receiving, the signal's polarization is determined by some other antenna. If the signal being received is "horizontally" polarized, evaluating the receiving antenna on its "total" gain seems only slightly less illogical than evaluating it solely on its "vertical (polarized)" gain.

 

[tripelo]

...define dBc for antennas, nor is the reference antenna described....has anyone else seen it????

Yes, dBc is commonly a reference to a carrier, frequently used in broadcast and communications applications when referring to an RF carrier.

An isotropic antenna is a physically impossible construct that is usefully defined to radiate the same intensity of electromagnetic waves in all directions. The utility of an isotropic antenna is for convenient comparisons.

As such, it makes sense to define the polarity of an isotropic to match the polarity of the antenna under consideration.

Compare the gain of a primarily:

- Horizontally polarized antenna to a isotropic that conceivably radiates horizontally* polarized radiation.

- Vertically polarized antenna to an isotropic that conceivably radiates vertically* polarized radiation.

- Circularly polarized antenna to an isotropic that conceivably radiates circular polarized radiation.​

*Note: A physically realizable antenna that radiates circular polarization, necessarily has both horizontal and vertically polarized components. Not having delved into the strict Electromagnetic definition of an isotropic antenna, cannot be sure whether it is defined to have only circular polarization or not. At the moment, guess this leaves some question about the actual definition of an isotropic.

The term dBic is commonly used for circular polarization to isotropic gain comparisons.
.

 

[300ohm]

There is nowhere you can specify the feed point any more.

Multiple EX cards are allowed, so I believe thats what you would do in this case.

 

[jbkeh]

Multiple EX cards are allowed...

It appears pretty well limited to multiple voltage EX for NEC2:

Only one incident plane-wave or one elementary current source is al-
lowed at a time. Also, plane-wave or current-source excitation is not
allowed with voltage sources. If the excitation types are mixed, the
program will use the last excitation type encountered.

Read in another forum where someone loaded the selected segment with a 300-ohm straight resistance, lit the model with a horizontal incident wave from the +X axis and simply extracted the amp value for that segment. No idea how he made that into something meaningful.

Interestingly, Supernec seems to allow mixed excitation and has a type "6" EX card that seems to address the problem (for anyone with both the financial resources and interest).

 

[300ohm]

If the excitation types are mixed, the program will use the last excitation type encountered.

Darn.

 

[holl_ands]

If you want to know the antenna's response to a transmitted Horiz. Pol. signal, I would choose Horiz.
instead of Total in the 3D Pattern plots. In the 2D Pattern plots, I frequently hit the "M" key
to display Total, Horiz and Total all at once (I also hit the "I" key).

If a "normal" (not Circular Polarized) antenna is receiving from a transmitter broadcasting
equal power in both H and V (e.g. Cir. Pol.) then I would expect the "Total Gain" numbers would
show the improvement wrt say Horiz. Only broadcasts. Many antennas have BETTER Total Gain.

However, I am aware that we need to to something DIFFERENT to make total sense for CP Antennas,
including showing the null response for mis-matched rotations, as Cebik discussed in the fol. two articles
(no registration required for these particular articles):
http://www.antennex.com/w4rnl/col1005/amod92.html
http://www.antennex.com/Sshack/tutorials/book2/chap17.pdf

I read thru them (three times) in the past several days....in addition to when I first came across them,
but I haven't yet figured out what the next step should be....and BTW, when you read Cebik's
Helical Antenna analysis papers, it appears that he is simply plotting Total Gain...which agrees
with all of the Helical Antenna papers I've been looking at as I get closer to releasing some designs.
PS: Some Preliminary Hi-VHF Helical Antenna Prototypes are analyzed here:
[Still optimizing and need to finish writing some descriptive narrative and a "how-to guide".]
http://imageevent.com/holl_ands/circularpolarized
Unanswered question: would 4nec2 show any Gain difference using a CP Antenna to receive
Cir. Pol.'s 90-degree phase shifted signals vs IN-PHASE H and V signals???

We should also bear in mind that antenna Gain numbers don't tell the whole story. Broadcasters
are allowed to add the Vertical component with up to the SAME power as their current Horizontal.
This DOUBLES the available power.....if we can efficiently capture it with a Cir. Pol. Antenna.
Also as discussed earlier, Cir. Pol. Antennas have minimal response to REFLECTED signals.

 

[jbkeh]

I'm definitely way over my head here but are we being a bit too limiting in what constitutes combined polarization?

If (as in your example of crossed dipoles) the two elements are in quadrature, then the signal would be elliptical (and circular if the signal strength of the two elements were equal), but what if the two elements are not in quadrature. As you mentioned, they could be completely in phase (perhaps resulting in a 45-degree tilted linear polarization, if of equal strength) or anywhere between 0 and 90 degrees (resulting in something I can't even envision)?

In the case of signals having both vertical and horizontal polarization components, then antenna designs capable of responding well to both would indeed be advantageous from the view of sensitivity as well as multipath suppression.

Fascinating that a vertical component of equal strength can be added. Must find out if this be the case here in the Great White North. Wonder which stations are doing this (and admitting it).

 

[holl_ands]

Yes, some stations are transmitting Elliptical Polarization (unequal H & V) and others Circular (equal H & V).
I restricted discussion to equal H & V since that is the default 4nec2 is using for Total, H & V Gain plots.

A list of U.S. and Canadians broadcasting (or perhaps planning to broadcast ???) Ellip. and Cir. Pol. signals is
found in fol. thread. Esp. look at the E and C lists in Post#4 (note that each webpage is NOT labeled...so see URL):
http://www.digitalhome.ca/forum/showthread.php?t=136180

 

[nav0239]

Hope we can first nail down the discussion if horizontal/vertical gain in 4nec2 is really referring to the gain of horizontally/vertically polarized signal. Or it’s actually referring to horizontal/vertical cross sections in 3D space. My understanding is the same as tripelo described in post #634, i.e., it is referring to the spatial cross sections and have nothing to do the polarization of the signals. Let’s also look at the following examples when doing F7-> frequency sweep: choosing

Gain -> it shows fixed theta and phi. This indicates that it is calculating the gain along a single direction in space.
Ver-> fixed phi angle, theta varies in 3D space (d-theta may be specified if needed). This indicates that it is calculating the gain in a vertical cross section in space.
Hor-> fixed theta angle, phi varies in 3D space (d-phi may be specified if needed). This indicates that it is calculating the gain in a horizontal cross section in space.

Let’s also think about what we (or some of us) mostly wanted to do with 4nec2. We wanted to optimize or evaluate antenna’s performance for receiving TV signals. Currently most of the NA TV signals are horizontally polarized. In the future, the signals may be circular or elliptically polarized, but that should be a topic for future discussion, when we are looking at antenna for receiving those specific signals.

The issue I had at the beginning was that, without getting this correct, I could be wasting time optimizing antennas for an unknown polarization, that’s actually no good or not optimized for receiving the horizontally polarized TV signals.

The antenna reciprocity, as holl_ands described in post #643, is well understood. If the “hor” gain in 4nect2 is actually based on horizontally polarized signal the problem is resolved. I just had another look at the 4nec2 user interface, may be optimizing for E-Phi could be a work around?

The method jbkeh mentioned in post #648 could be a solution. Has anybody actually done this and can help to show how to get the results plotted?

 

[jbkeh]

Let’s also look at the following examples when doing F7-> frequency sweep

Well, perhaps we should first look at what can be done. Think of the possible variables at hand: Number of frequencies (and value of the step), number of directions of transmission in both azimuth and elevation (and again the value of the step(s)), as well as signal polarization.

If we unleashed the entirety, a run would take nearly forever.

So the variables get limited.

Look at the Generate pane (F7). We get to choose among "Far Field Pattern", "Frequency Sweep" and "Near Field Pattern".

If we choose "Far Field Pattern", we are then given a choice among "Full", "Vertical" and "Horizontal". Note that this is in reference to "Pattern", so selecting "Full" allows us only the step size (in degrees) and that step is applied to both the azimuth and the elevation. If we choose "Ver" (or "Hor"), we get to fix the azimuth (or elevation) and vary the elevation (or azimuth) over a step-range. Note that there is NO CHOICE as to polarization.

Similarly, if we choose "Frequency Sweep", we are given a choice of setting for evaluation a single direction (Gain) or a vertical "slice" (Ver) where we set a single azimuth and a step-range of elevations; or a horizontal "slice" (Hor) where we set a single elevation and a step-range of azimuths; or a spherical "area" (Full 3D) where we set a starting point and a step-range of both azimuth and elevation. Again there is NO CHOICE as to polarization.

Why is there no choice for polarization? Because it is ALWAYS considered and calculated.

However, it may not always be shown. The graphs show only Total (composite horizontal and vertical polarizations) values, while the plots, WHETHER THEY DISPLAY A 2D ELEVATION OR A 2D AZIMUTH OR A 3D IMAGE give you the choice of selecting Total, Vertical or Horizontal POLARIZATIONS, or (in the case of the 2d plots) all at once (colour coded), as well as the E-fields, either horizontal or vertical.

If the received signals are horizontally polarized, it would seem rational to look at the antenna's response to them. However, most of the antennae presented here have been so "tuned" to a single polarization plane (and usually aligned with the horizontal) that there is little difference between "Total" and "Horizontal" as the "Vertical" is negligible. If stations elect to transmit a composite polarized signal, then some review of design may be required.

For a far shorter answer from a far more knowledgeable source, re-read the first sentence of post 650.

 

[holl_ands]

TOTAL, HORIZONTAL POLARIZATION AND VERTICAL POLARIZATION GAIN FOR A STICK DIPOLE:

Try not to confuse H. and V. Polarization (a property of the RF signal) with the PATTERN plots that
depict 2D "slices" for either Horizontal Plane (X-Y, what we usually post) or Vertical Plane (X-Z)
to show the Elevation Gain. These "slices" are what results when you take a knife and cut
and display the "slice" for the following 3D Pattern plots. The Green, Red & Blue colors in these labels
are the same colors used in the 2D Pattern plots below [Z is up, Y is towards right-rear]:

TOTAL GAIN FOR A STICK DIPOLE:
Total Gain is the SUM of Horizontal and Vertical Polarization Gains. Note the "DONUT" on a pole.
If the Stick Dipole is rotated vertically so that it aligns with the Z-Axis, the Total and Horizontal
Polarization patterns are OMNIDIRECTIONAL. A Vertical Dipole is sometimes used in the FM Band:
http://imageevent.com/holl_ands/loops/folded [See last slide.]

HORIZONTAL POLARIZATION GAIN FOR A STICK DIPOLE:
Note the Gain lobes in the Forward & Rear directions (X-Axis) on the horizon (X-Y plane).
If an airplane dragging a wire antenna (Horiz. Pol.) were to approach and fly directly over,
the Gain would be fairly constant, although off to the sides there is loss.

VERTICAL POLARIZATION GAIN FOR A STICK DIPOLE:
Note the NULL Gain in all directions for signals on the horizon (X-Y plane).
If an airplane with a whip antenna mounted (as usual) on the bottom of the fuselage
(Vertical Pol.) were to approach and fly over, Gain would be minimal until a narrow area
directly over our antenna.

 

[holl_ands]

TOTAL, HORIZ. POLAR. AND VERTICAL POLAR. GAIN FOR A STICK DIPOLE:

4nec2's generates a variety of PATTERN plots, based on the "RP" statement(s), with numerous
options that can be very confusing....which is why I came up with a "standard" set of RP statements
that I copy/paste into every new file I run....I'm not going to try to explain them here....see my RP's.
And good luck trying to manually enter data if you don't choose the "Use Original File" option....

I use fol. with a FIXED FREQ to generate the 3D Pattern plots in the preceding post:
FR 0 0 0 0 590 0 ' Fixed Freq
RP 0 285 73 1510 90 0 5 5 0 0 ' 3D Full Coverage obscures antenna (Fixed Freq)

I use fol. to sweep the (old) UHF Band to generate Raw Gain, SWR & PATTERN plots:
FR 0 39 0 0 470 6 ' Freq Sweep 470-698 every 6 MHz - PREFERRED FOR UHF
RP 0 285 73 1510 90 0 5 5 0 0 ' 3D Full Coverage obscures antenna (Fixed Freq)

Other alternatives are given in my 4nec2 files, including forcing 4nec2 to do FORWARD Gain.

If you hit the "M" key ("Show Multipattern" in Far-Field menu), you will see Total,
Horiz. and Vert. Polarization Gain displays, such as the fol. for a UHF STICK DIPOLE.
But since the "M" key data replaces my filename, I avoid using it when the Vert. Pol. Gain is
negligible, as indicated here.
[I also hit the "I" key to show detailed Gain/F-B/BW information]:

In general, "flat" antennas (e.g. Dipole, Yagi, LPDA) have zero (or very small) Vertical Polarization Gain,
and you'll have to run a 3D Pattern (using multiple Fixed Freqs) to see where the Gain lies for more
complicated antennas.....

In addition to Loops & RabbitEars, the fol. example of V. Pol. sensitivity comes to mind:

HiVHF+UHF DBGH with 28 Rods:
http://imageevent.com/holl_ands/grayhoverman/dbghcombo
Note Vertical Polarization Gain skews to four odd-ball angular quadrants...but NULL FORWARD:

Hi-VHF Simple-Rabbit-Ears at 45-deg Angle with 14.75" Whiskers and 0.75" Feedpoint:
http://imageevent.com/holl_ands/circularpolarized/rabbitears/hivhfrabbitears
Vertical Polarization Gain was very disappointing, dropping below the good Horiz. Pol. Gain by over 6 dB
in the Forward direction. Note that Vert. Pol. Gain increased off to the SIDES of the anenna, but the Horiz.
Pol. Gain is very small in this NULL region. [Total & Horiz. Pol. Gain was similiar to Stick Dipole above.]

Hence Rabbit Ears aren't that great for Elliptical or Cir. Pol. reception.
The SWR was also much more sensitive to Whisker Length than I thought...just 1/2-inch will severely
degrade SWR, although there is only a minor impact on SWR. Note that this is for some SIMPLE Rabbit Ears,
which I have in my "mobile test kit" and does NOT include any impedance matching networks. Actually,
since they collapse, the elements taper to a much smaller radius than modeled...making SWR even worse:

 

[Elvis Gump]

Hi Folks,

Experiment #1

Purpose of Experiment:

To observe what happens to the Total Horizontal Forward Gain vs. Channel Number curves as the feedpoint gap distance is changed from 0.5 inches to 3.0 inches in steps of 0.5 inches for several families of the Gray Hoverman Driver Array Assembly.

Figure 1: The Gray Hoverman Driver Array Assembly
http://img54.imageshack.us/img54/3411/grayhovermandriverarrayzv9.jpg

In Figure 1, the length of wire number 17 corresponds to the feedpoint gap distance; wire numbers 2 through 7 and 10 through 15 represent the zigzag elements; and wire numbers 1, 8, 9 and 16 make up the stub elements.


Methods and Materials:

At http://www.digitalhome.ca/forum/showthread.php?t=81982&page=19, Post 284, I found a drawing by Elvis Gump of the First Generation Gray Hoverman Antenna developed by Autofils and others. This seemed like a great place to start since it is a real working and well optimized antenna. The feedpoint gap is 1.77 inches. The length of each zigzag element is 7.07 inches. The length of each stub is 5.59 inches.

In this experiment, I planned to let the gap distances vary from 0.5 inches to 3.0 inches in steps of 0.5 inches and the stub lengths vary from 2.0 inches to 6.0 inches in steps of 1.0 inch. I decided to fix the zigzag element lengths at 7.07 inches in order to keep the total number of antenna modeling files down to a modest count. The next job was to construct all of these models and to save the resulting NEC input files.

The 4nec2 software provides access to four different NEC input file editors. Each has its uses. The most primitive is “Notepad Edit” (Ctrl+F1). Using Notepad to prepare an NEC input file is equivalent to sitting down in front of an old IBM Hollerith Card Keypunch Machine with a deck of blank cards.

The original Numerical Electromagnetics Code (NEC) program was developed in the 1970s for the US Navy at the Lawrence Livermore Laboratory. The software evolved over the next few years. In January 1981, a three part manual was published detailing the NEC2 program description and theory, program listing, and user’s guide. The187 page User’s Guide speaks mostly of preparing an NEC input card deck. Each card type and its associated fields are thoroughly described.

The second editor, called “NEC Editor” (Ctrl+F2), is a much more user friendly interface than Notepad and the 187 page User’s Guide. It knows about each card type and it fields. The user can insert a card into the deck by selecting an appropriate card type and filling in the well labeled fields.

The third editor is called “Geometry edit” (Ctrl+F3). An in depth knowledge of card types and their fields is not required to model an antenna. Instead, one can focus on laying out an antenna’s elements in a three dimensional coordinate system, XYZ space. The geometric editor has buttons to switch from 3D space to the two dimensional XZ, YZ and XY planes. For flat, two dimensional, designs like in this experiment, one can work exclusively in the YZ plane. The geometric editor features a “drag and drop” drawing style for creating straight wire elements. A graph paper like grid is provided in the two dimensional planes for drawing individual straight wire elements. In the “Select object mode”, the properties of any wire can be displayed and/or edited by left clicking on the object of interest. The ends of wires can be easily connected by taking advantage of the “Snap to grid” and “Snap to wire” functions. Beware of “Keep connected” function. When adjusting a wire, there are times to turn it on and times to turn it off. The “Edit/Undo Move” function found on the standard menubar is critical, especially after accidentally making a mess of things.

“NEC editor (new)” (Ctrl+4) is the forth editor. It is a logical extension of the second “NEC editor”. The user interface has subject tabs for “Symbols”, “Geometry”, “Source/Load”, “Frequency/Ground”, “Other” and “Comment”. Most of the tabs provide a spreadsheet like grid for data entry and editing. In some future experiment, I hope to become familiar with using the elegant “Symbols” method for modeling an antenna”. But for now, I will move ahead using the “Brute Force” technique.

I used 4nec2’s geometric editor to build the NEC model files. The stubs were very easy to draw. I was a little worried about the 7.07 inch zigzag elements. The snap to grid and the snap to wire functions made drawing the zigzag elements much easier than I first thought. The strange 7.07 inch length turns out to be the hypotenuse of a right triangle with two 5.0 inch sides on either side of the 90 degree angle. The two 5.0 inch sides neatly correspond to the graph paper like horizontal and vertical grid lines. In the add wire mode, it became a matter of left clicking the mouse to anchor one end of a new wire then dragging 5.0 inches horizontally and 5.0 inches vertically (or the equivalent diagonally in just one step) and releasing the mouse button.

Adjusting the feed point gaps was difficult in the beginning. Then, I read the help file and discovered a drag and drop technique for moving and rotating groups of wires. It goes something like this. While in the geometric editor’s 2D select object mode, left click a wire that will shortly become the “dragging handle”. Move the cursor to an empty spot on the grid. Next, left click and immediately drag and drop diagonally over the group of wires to become a temporary unit. A rectangle will appear showing you what you are about to include. When you let up on the mouse button all of the wires with both ends inside the rectangle will be selected and will turn red. One of the wires will appear thicker than the others. This is the "dragging handle". You can left click and then drag and drop the unit to its new location. It takes a little time to get used to it, but it is very effective. Try it, you will like it.

Altogether, thirty NEC files were created and separated into the following five family groupings:

H01 - 7.07in ZigZags - 2in Stubs - 0.5in Gap.nec
H02 - 7.07in ZigZags - 2in Stubs - 1.0in Gap.nec
H03 - 7.07in ZigZags - 2in Stubs - 1.5in Gap.nec
H04 - 7.07in ZigZags - 2in Stubs - 2.0in Gap.nec
H05 - 7.07in ZigZags - 2in Stubs - 2.5in Gap.nec
H06 - 7.07in ZigZags - 2in Stubs - 3.0in Gap.nec

H07 - 7.07in ZigZags - 3in Stubs - 0.5in Gap.nec
H08 - 7.07in ZigZags - 3in Stubs - 1.0in Gap.nec
H09 - 7.07in ZigZags - 3in Stubs - 1.5in Gap.nec
H10 - 7.07in ZigZags - 3in Stubs - 2.0in Gap.nec
H11 - 7.07in ZigZags - 3in Stubs - 2.5in Gap.nec
H12 - 7.07in ZigZags - 3in Stubs - 3.0in Gap.nec

H13 - 7.07in ZigZags - 4in Stubs - 0.5in Gap.nec
H14 - 7.07in ZigZags - 4in Stubs - 1.0in Gap.nec
H15 - 7.07in ZigZags - 4in Stubs - 1.5in Gap.nec
H16 - 7.07in ZigZags - 4in Stubs - 2.0in Gap.nec
H17 - 7.07in ZigZags - 4in Stubs - 2.5in Gap.nec
H18 - 7.07in ZigZags - 4in Stubs - 3.0in Gap.nec

H19 - 7.07in ZigZags - 5in Stubs - 0.5in Gap.nec
H20 - 7.07in ZigZags - 5in Stubs - 1.0in Gap.nec
H21 - 7.07in ZigZags - 5in Stubs - 1.5in Gap.nec
H22 - 7.07in ZigZags - 5in Stubs - 2.0in Gap.nec
H23 - 7.07in ZigZags - 5in Stubs - 2.5in Gap.nec
H24 - 7.07in ZigZags - 5in Stubs - 3.0in Gap.nec

H25 - 7.07in ZigZags - 6in Stubs - 0.5in Gap.nec
H26 - 7.07in ZigZags - 6in Stubs - 1.0in Gap.nec
H27 - 7.07in ZigZags - 6in Stubs - 1.5in Gap.nec
H28 - 7.07in ZigZags - 6in Stubs - 2.0in Gap.nec
H29 - 7.07in ZigZags - 6in Stubs - 2.5in Gap.nec
H30 - 7.07in ZigZags - 6in Stubs - 3.0in Gap.nec

These Gray Hoverman Driver Array files can be found in the Yahoo Group named AntennaModelingExperiments under Hobbies and Crafts.

A horizontal frequency sweep was produced for each of the above NEC files using 4nec2’s Generate (F7) function. The following parameters were used: Resolution: 5 degrees; Frequency Start: 455 MHz; Frequency Stop: 715 MHz; Frequency Step: 6 MHz; Forward Theta: 90 degrees; Forward Phi: 0 degrees; Forward delta-Phi: 5 degrees; Backward Theta: 90 degrees; Backward Phi: 180 degrees; and Backward delta-Phi: 5 degrees. A 455 to 715 MHz range with steps of 6 MHz was selected to produce points which fell in the center frequencies of each UHF channel from 14 through 51 with three extra points on either side of this channel range.

As the Total Horizontal Forward Gain data were generated for each NEC file, it was immediately transferred to an Excel spreadsheet for later plotting. See http://www.digitalhome.ca/forum/showthread.php?t=85987&page=14, Post 208 for 300ohm’s method of transferring output data from 4nec2 to a spreadsheet.


Results and Conclusions:

The Total Horizontal Forward Gain vs. Channel plots are shown below. The first family of gain curves shows the Gray Hoverman Driver Array which is most similar to Elvis Gump’s drawing without a reflector.

For this family, 7.07 inch zigzag elements and 6.0 inch stub elements, there appears to be very little practical difference between the Total Horizontal Forward Gain curves with feedpoint gaps distances ranging from 0.5 inches to 3.0 inches.

As the stub element lengths become shorter and shorter, the effect of the feedpoint gap distance slowly begins to appear.

**********************************************************
Experiment #1 Continued in GH Driver Array - Modeling Experiment - Part 3
Due to an Image Limit of 4 per Post
**********************************************************

Wow. I think all that is pretty amazing. I came here today and searched my name just to find that old drawing and found this post as well. But I have to confess I don't understand a jot of it. The drawing referred to as 'mine' was merely created trying to work out how to convert the original measurements of the Gray-Hooverman from MM to INCHES, and tidy it up a bit in Adobe Illustrator. It was such a mathematical nightmare which became such a headache I finally had to seek out a MM ruler to build my antenna. The final result was a effort made of scrap & recycled bits of odds and ends on the cheap.

It was never intended to be some ideal design of any sort, just a drawing I could use as while drilling and hammering my thumb and that sort of thing to close as I could get in the ballpark of good enough. Anything approaching technical know-how goes to the originators of the design in the GH thread itself, I'm just a monkeysee-monkeydo amateur of the most amateurish sort.

Still the old antenna has held up rather well through high winds and thunderstorms and the baking hot summers of Mississippi I must say. It never experiences rain fade either, which is more than I can say for my DirecTV dish. I worked for those thieves in the interim in what they laughingly call Technical Support for about a year and a half since I built my antenna before I quit in total disgust about how badly they wanted 'agents' to lie to customers, deny service and generally be unhelpful. It's hard enough to walk someone's grandma through how to work her remote over the phone without managers roaming the floor browbeating you about how the call is taking too long.

If the drawing was of any help to anyone in the meantime I'm very happy, but please consult Autofils and those other wonderful guys who helped me because I had very little idea of what I was doing at the time.

Elvis Gump

 

[dakrtan]

In 4nec2 how do you measure the space in between the 2 reflector rod or distance X. In Nikiml website is easy I can just click or highlight 2 things I can see DZ or DY or DX, but I have no idea how to figure that out in 4nec2. I try to draw a line from one point to another to find the distance in geometry editor, but the result is different from Nikiml website. I hope I ask question in the right place. Sorry, I’m still learning. Thanks for all your help.

 

[nikiml]

In 4nec2 when you select a wire you see the x,y and z of the two ends.
All you have to do is figure which end has which coords. Write them down, do the same for the next wire and the subtract to get dx,xy or dz.

dx = x1-x2
dy = y1-y2 etc.

 

[300ohm]

When you click on a wire in 4nec2 Geometry Edit :

The green circled area shows you the length of the wire.

The red circled area shows you the X, Y, Z coordinates, in that order, XYZ.

The blue circled area allows you to change viewing angles.