John,
The modeling software packages do take into
consideration various ground parameters.
Look carefully and you will see there are
typically several "types" of grounds to
select depending on the software along with
defining ground characteristics.
Also keep in mind that the NEC "engines", as
there are several versions, and the MININEC
"engines" each have their own strengths and
weaknesses for different settings such as
close to ground in terms of wavelength and
radials, tapered elements, and so on.
Software such as EZNEC and 4NEC2 are really
user friendly input/output interfaces to one
of the above "engines".
I tend to use both "engines" to check for any
guidelines that I may have broken in building
the respective models, and compare the results.
Especially if I "discover" a model with results
that are more than what theory says can exist!
QST, the W4RNL site, and others are good places
to learn more and understand what and what NOT
to use models for. Example? Terrain with slopes
is not going to be accurate with most models.
Chrisman is another author who has published
many good articles, especially on verticals based
mainly on models and not real world, so please
keep that in mind when reviewing his data.
Past issues of the NCJ have many of his texts.
The NEC4 "engine" is considered to be the more
accurate than NEC2 as to ground influenced type
of antenna models. Note NEC4, not 4NEC2 as the
latter uses the NEC2 "engine".
NEC4 is also very expensive, though with some
academic settings there can be some savings.
Because we cannot EASILY measure specific ground
data and loss, most models are considered to be
underestimating actual system losses. By how much?
No real agreement, many would suggest 6 - 20 dB!
Anyone looking for your EE PhD dissertation topic?
see the above, start with Brown's work and see if
you can build models that match the real world.
BTW, HFTA, which comes with the ARRL Antenna Book is
good for dipole/yagis once you setup your terrain
files, but of no use for verticals.
So John, you are on the right track here as you are
recognizing that there are MANY variables, so please
stay with this and compare the models and real world.
That's how I built my 1st 80m 4x array as I started
with a 'normal' model which showed me what I could
expect as to pattern/gain/etc. Then I worked with
what I had to do for my backyard's limitations and
after much time, came up with an new idea to build,
then built it and compared the model with real world.
In my case the 4 vertical bases had to be spaced out
much wider than the normal 65' x 65' square. That also
meant the phase lines had to be 3/4w instead of 1/4w
due to the wider square and that the top of the wire
elements could be tilted into toward the tower. The
end result was an "average" spacing of 65' x 65' even
though the base was more like 105' x 105'.
If I had listened to "ham lore" and not learned how a
4x array is supposed to work, then my first attempt
would probably also had been my last attempt as there
is a great deal of unsupported "lore" on arrays today.
CONGRATS on your efforts to LEARN how antennas work!
73 de Billy, AA4NU
-----Original Message-----
>From: John
>Sent: May 21, 2012 9:56 PM
>To: Discussion of Ten-Tec Equipment
>Subject: Re: [TenTec] gas on the fire
>
>My experience is as a self taught student still learning about how the
>NEC models work and I am also trying to move from EXNEC to 4NEC2. What
>I am really trying to say here is that if someone else sees anything
>incorrect, I know I am on somewhat shakey ground, so please feel free to
>correct me.
>
>When the NEC program is opened, one of the first tasks is the wires
>(antenna) definition. This is done on 3 axis's (X, Y, and Z as
>Vertical). The wires definition allows the definition on all 3 axis,
>so an antenna can be defined as lying on (extending for some distance
>along) the X axis and be some amount of height (defined in meters, feet
>or wavelength) above ground on the Z axis. This places the antenna,
>either vertical or horizontal, the defined distance above ground (which
>is also definable.) As the height above ground is increased, changes in
>the pattern diagram will show changes (plus or minus) in the takeoff
>angle. This would seem to indicate that NEC is sensitive to ground
>changes which can include height among others. Although I have not done
>any extensive exercises, I believe that the program will also be
>sensitive to the ground conditions (why make ground definable if there
>is no sensitivity) as well as showing the effect of defined ground
>planes or radial fields.
>
>I hope this helps and provides some insight into your question.
>
>
>John / WA1JG
>
>
>On 5/21/2012 5:00 PM, Richards wrote:
>>
>> On 5/21/2012 3:47 PM, Paul DeWitte wrote:
>>
>>
>>> First, I think that if you modeled a ground mounted vertical and the same
>>> vertical at 330 feet above ground, you might find the radiation pattern
>>> (take off angle) would not be the same for the two installations (just my
>>> guess).
>>
>> This makes sense to me. Can the NEC-based modeling
>> programs do that ? They seem to consider different
>> ground conditions... but what if ground is 350 feet away ?
>>
>> Good question.
>>
>>
>>> Second, so far no one has said anything about grayline DX (or about that
>>> time) when some claim that a dipole works better than their vertical. This
>>> is on the low bands, dont know about higher freq.
>>
>> More of the mysteries of propagation ... I think
>> many hams know very little about the vagaries
>> and effects of propagation -- and I am one who
>> needs to learn more about this aspect of radio.
>> Hams often talk about take-off pattern, yet blithely
>> ignore what might be happening to the signal after
>> it leaves the the back yard.
>>
>>
>>
>>> About S meter accuracy. You either hear them or you dont. If you can hear
>>> them and work them, it dont make any difference what the S meter reads.
>>> I worked a lot of DX on my Omni C that did not move the S meter needle, but
>>> they made it into the log.
>>
>> I thought there was a school of thought that
>> based the signal strength component of signal
>> reports on S-meter readings - in which case it would
>> matter to them whether or not S-meters were
>> 1) accurate, and 2) consistent between rigs.
>> If so, then I think this would be a worthy topic
>> for beating to death. ;-)
>>
>>
>> ---------------------------------- K8JHR ---------------------------------
>>
>>
>> _______________________________________________
>> TenTec mailing list
>> TenTec@contesting.com
>> http://lists.contesting.com/mailman/listinfo/tentec
>>
>_______________________________________________
>TenTec mailing list
>TenTec@contesting.com
>http://lists.contesting.com/mailman/listinfo/tentec
_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec
|