4nec2 hidden variables.
 

On 8/13/2011 3:59 PM, Helmut Wabnig wrote:
On Sat, 13 Aug 2011 14:15:22 -0700, Jeff
wrote:

..........

I am making progress, though.
Got the optimizer running and edited a few of the example files
to have the variables selection available.

Works.

Now I watch the optimizer quickly de-optimizing the antennas
until they radiate into the wrong direction or get stuck in a
circular radiation diagram. In one case the optimizer bent
and rotated the dipoles around, I bet no specialist can do what I can.

Real fun, that is :-)


Video of optimizer (25 min):
http://www.youtube.com/watch?v=kYy6Yur127A


Yeah, thats good. Sitting back and watch.
Thanks for the idea of watching the Youtube demos.
Better than endless trial-and-error.
My goal is to find out what makes a direction finding antenna
different from a gain optimized antenna.
F/B ratio is all that matters for ARDF, for example.
Gain is secondary and side-lobes should be absent.
I must get that NEC working by all means.


Then when you set up your optimization criteria in 4nec2, weight the F/B
or F/R with 100% and the others with zero (although sometimes, you may
want to weight forward gain at 10%, just to keep it from driving to zero
gain)

4nec2 hidden variables.
 

On 8/14/2011 11:53 PM, Helmut Wabnig wrote:


When optimizing for F/B also 4nec2 results in a too short driven
element, same as YAGICAD, they obviously use the same formula,
(a long two-liner) from literature.

Unlikely...
4nec2 is just driving the finite element NEC engine underneath. YagiCad
uses analytical approximations.

NEC is *very* sensitive to segmentation, spacing, and wire diameters,
etc. particularly when close together or angled. The NEC4 engine is
quite a bit better, but you're probably using the default free NEC2 engine.

To effectively use this kind of thing, you need to be pretty aware of
the limitations and peculiarities of the underlying FEM codes. While I
wouldn't expect someone to have read the theory manual for NEC2, you
might want to check out some of L.B.Cebik's writeups on using MoM codes,
or the stuff in the ARRL Antenna Compendium from time to time.



I actually built antennas following that prescription,
and they do not work without additional corrections,
e.g. a very long hair-pin match.



Imagine a 3 element 2 meter yagi,

|
| |
| | |
| | |
| | |
| | |
| | |
| |
|

Something is wrong here with the capacitive driven element.

Depends on what you were optimizing for.
If all you said was "drive F/B to maximum", then it won't try to get you
a decent match, all it will do is drive to having element currents that
optimize for F/B, which will almost certainly have an evil feedpoint
impedance.



Now I am having a hard time, because I want to find out in
what differs a gain optimized vs a F/B optimized antenna
and it is questionable whether that can be done with software.
Have not tried other programs yet.

What I do is not try to use the optimizer to design the whole antenna
including matching network. I optimize the basic element design first
(without matching network). Then I add the matching network, hold the
element spacing and length constant, and let that optimize. Then I go
back and allow element length to change, but hold matching network
dimensions/component values (I tend to use lumped element matching
networks) fixed.

It's an iterative thing because it's tough to specify an appropriate
optimization cost function with a simple interface. What you really
want is something like "optimize F/B, but keep the feedpoint impedance
20 ohms and 100 reactive ohms"





A Yagi with fixed element positions can only be used for 1 frequency
(within a very small range) and therefore may be impractical
in many situations. The CEBIK tape measure antenna with their
water-pipe construction could be easily modified with a moveable
center element.

this is just not true (otherwise the SteppIR wouldn't work.. it has
fixed element spacing and works over a huge (1 octave) frequency range)
with fairly good performance.

One can also do similar with fixed spacing, and fixed length, and
reactive loading of the elements at the centers.

4nec2 hidden variables.
 

On Mon, 15 Aug 2011 08:53:35 +0200, Helmut Wabnig [email protected] ---
-.dotat wrote:

http://imageshack.us/photo/my-images/837/snc14096a.jpg/

Cute cat. However the antenna seems a bit odd. The high carbon steel
tape measure elements are not the best for RF conductivity. Good
enough for receive, but I would check for RF heating in transmit.

The tape measure is also not a circular rod element. You probably
modeled it using a circular cross section. That's probably a good
first try, but if you want accuracy, the actual element cross section
will be needed. That's what I had to do with this stamped sheet metal
antenna:
http://802.11junk.com/jeffl/antennas/mfj1800/

(Without the hair pin match)
http://img5.imageshack.us/img5/3479/snc14112hairpin.jpg
Therefore I put the driver on a slider
http://img841.imageshack.us/img841/8...14097small.jpg

I presume the hairpin match goes between the wing nuts. Looking at
the photo of the hairpin, it seems small for matching a simple 2m
yagi. It's also a good idea to add a balun and to ground the center
of the hairpin match to the center boom.
http://vk1od.net/antenna/misc/BetaMatch.htm


--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

4nec2 hidden variables.
 

On Mon, 15 Aug 2011 18:30:05 +0100, Jeff wrote:


Firstly it is not 'drag, drop& move'; it is click, drag& drop! Try
reading the Help a little more closely, you will find that it says:

"The next thing to do is add a voltage-source. While still in Add-mode,
click the 'Source button' (right of the 'Wire button'). Next click and
hold down your left mouse-button somewhere in the picture-box. At the
current mouse-pointer position a new source-object is displayed. Drag
the source-object to the middle of the second wire, just between the two
lower wires-ends of the feedline and release the mouse-button. When
properly positioned a new source is now added."

It does not work as described, when inserting a source on a dipole
wire in the center segment, one of the first things to do in Example1.


w.


Utter Rubbish!!! It works EXACTLY as described!!!

Hold down the left mouse button, drag the source to the segment where
you wish it to be, release the button - exactly as described!!!

Activate the source & add button, then click on the middle segment
and expect the source symbol to appear there.
It appears but vanishes in an instant leaving a highlighted segment.
If the sideways motion is not performed, you do not get it.
You have to pull it and that is counterintuitive.
The above description may be correct in that you have to pull
the object. After having worked with dozens of CAD programs
I did not expect such a trap.
Many users claim similar difficulties, the internet is full of them.
Honestly, who reads all the help files, hundreds of pages?


I know there are people who think every bug is a feature.
Those people do not live from their own work,
somebody else must pay for their living.

It took me 3 days to learn 4nec2 because of a long list of "features"
including data loss and program crashes. "Notify the author"
it suddenly says. The user sits there "What did I do wrong?"
which is a nagging question actually when the fault lies elsewhere.

All that counts is WALL CLOCK TIME.
But it's free.

w.

4nec2 hidden variables.
 

Utter Rubbish!!! It works EXACTLY as described!!!

Hold down the left mouse button, drag the source to the segment where
you wish it to be, release the button - exactly as described!!!

Activate the source& add button, then click on the middle segment
and expect the source symbol to appear there.
It appears but vanishes in an instant leaving a highlighted segment.
If the sideways motion is not performed, you do not get it.
You have to pull it and that is counterintuitive.
The above description may be correct in that you have to pull
the object. After having worked with dozens of CAD programs
I did not expect such a trap.
Many users claim similar difficulties, the internet is full of them.
Honestly, who reads all the help files, hundreds of pages?


I know there are people who think every bug is a feature.
Those people do not live from their own work,
somebody else must pay for their living.
..

Well if you don't actually follow the instructions on the Help then what
can you expect! Don't blame the program and then say the Help is wrong
or not there! RTFM.

"click and hold down your left mouse-button somewhere in the
picture-box" then "Drag the source-object to the middle of the second wire".

Jeff

4nec2 hidden variables.
 

On Mon, 15 Aug 2011 15:09:52 -0700, Jim Lux
wrote:

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


A Yagi with fixed element positions can only be used for 1 frequency
(within a very small range) and therefore may be impractical
in many situations. The CEBIK tape measure antenna with their
water-pipe construction could be easily modified with a moveable
center element.

this is just not true (otherwise the SteppIR wouldn't work.. it has
fixed element spacing and works over a huge (1 octave) frequency range)
with fairly good performance.

One can also do similar with fixed spacing, and fixed length, and
reactive loading of the elements at the centers.


In theory ( from the simulations) the Yagi is rather narrowband.
Especially the F/B ratio is very frequency dependend.

The SteppIr does vary the elements, if I understand it correctly?
For practical reasons they do not change element positions on
the boom but sufficiently adapt to each frequency band.

For the same practical reasons I do it the other way round,
move the driver element in the 3 element yagi, and leave
everything else as is. This is intended for mobile direction finding.
Based on the Cebik pages and improved by making it tuneable
by moving the driver along the boom.

There is a story behind.
ARDF is a sporting endeavour, where people run around in
a restricted area searching for a defined signal, or several
foxes, as they are called. The antennas are optimized for the
fox frequencies. That's child play compared to real world.

One night I got a phone call from a desperate guy in Bosnia.
They had held a model flight competition there, class F1C,
and his model airplane flew ten kilometers away.
They put crash transmitters on the models and they
range in frequency from 140 to 160 Mhz without any approved
frequency raster, just what they can get and not to disturb
the neighbour model transmitter. So they need equipment for
a 20 MHz range. Most of them use amateur radio handhelds.
The transmitters weigh about 5 grams.

The Yugoslavian Balkan war was just over, where the killed 200000
people by shooting and cutting throats, and the mine fields were
still active and most of them still are today. They put fences around
and automatic siren warnings. That desperate guy's model airplane
was in or near such an area and night doomed. He has searched
for the whole day , was exhausted and could not locate it.
"Signals from everywhere and from all directions", he said.
That model was his only one for the competition and very expensive,
they cost a few thousand dollars.

"Signals from everywhere and from all directions."
That is a very interesting statement. Did you ever experience such
a situation? I had to place foxes and go after them to find out what
the problem was which the guy described. I expected an easy job,
because I am such an old fart in radio engineering. It was late summer
and corn standing two meters high, railway routes nearby,
high voltage lines criscrossing the landscape, buildings, hills,
mountains, fences, and now imagine it's getting dark and you are
in a foreign country where you never had been before,
between lakes, rivers, wet ditches and waste water canals.
I never knew that 2 or 3 meter high corn fields act as an antenna
array. The power lines and railways are wave ducts of extraordinary
quality, every metal fence collects and reflects and adds to the
reflections from the rocky mountains. Electromagnetic interference
overlays the fox signal until you only hear noise and crackles.
There you go. I hope the reader can imagine what is the difference
to a well planned amateur sport radio fox hunting event.

Quickly I recognized that a fixed-frequency ARDF antenna is
not suited for the task and one must be able to tune the F/B ratio
for the different frequencies the transmitters use, in combination
with a conventional attenuator. If your antenna receives from the
back, you will hear "signals from all directions".


w.

4nec2 hidden variables.
 

On Mon, 15 Aug 2011 23:14:19 -0700, Jeff Liebermann
wrote:

On Mon, 15 Aug 2011 08:53:35 +0200, Helmut Wabnig [email protected] ---
-.dotat wrote:

http://imageshack.us/photo/my-images/837/snc14096a.jpg/

Cute cat. However the antenna seems a bit odd. The high carbon steel
tape measure elements are not the best for RF conductivity. Good
enough for receive, but I would check for RF heating in transmit.

The tape measure is also not a circular rod element. You probably
modeled it using a circular cross section. That's probably a good
first try, but if you want accuracy, the actual element cross section
will be needed. That's what I had to do with this stamped sheet metal
antenna:
http://802.11junk.com/jeffl/antennas/mfj1800/

(Without the hair pin match)
http://img5.imageshack.us/img5/3479/snc14112hairpin.jpg
Therefore I put the driver on a slider
http://img841.imageshack.us/img841/8...14097small.jpg

I presume the hairpin match goes between the wing nuts. Looking at
the photo of the hairpin, it seems small for matching a simple 2m
yagi. It's also a good idea to add a balun and to ground the center
of the hairpin match to the center boom.
http://vk1od.net/antenna/misc/BetaMatch.htm


Thank you for the links, appreciate your help.


I fear that in the end I will have to cut and tweak using the
antenna analyzer, and with the time perhaps find
"tape measure correction factors".
Want to keep the design as simple as possible,
and hope it works without a balun or additional tricks.
Am not yet sure if it will work that way, or not.

w.

4nec2 hidden variables.
 

On Tue, 16 Aug 2011 09:48:00 +0100, Jeff wrote:


Well if you don't actually follow the instructions on the Help then what
can you expect! Don't blame the program and then say the Help is wrong
or not there! RTFM.

"click and hold down your left mouse-button somewhere in the
picture-box" then "Drag the source-object to the middle of the second wire".

Jeff
Of course you are right,
but there are thousands of (unread) help pages left.
Should I read them all?

On YOUTUBE another (not me) user said, he falls asleep
after 4 pages of help text ( the getting started stuff).
I printed the "getting started" but fell asleep after page 3.

w.

4nec2 hidden variables.
 

On 16/08/2011 10:25, Helmut Wabnig wrote:
On Tue, 16 Aug 2011 09:48:00 +0100, wrote:


Well if you don't actually follow the instructions on the Help then what
can you expect! Don't blame the program and then say the Help is wrong
or not there! RTFM.

"click and hold down your left mouse-button somewhere in the
picture-box" then "Drag the source-object to the middle of the second wire".

Jeff
Of course you are right,
but there are thousands of (unread) help pages left.
Should I read them all?

On YOUTUBE another (not me) user said, he falls asleep
after 4 pages of help text ( the getting started stuff).
I printed the "getting started" but fell asleep after page 3.

w.

Well since the instructions on how to add a source are on page one, it
would appear that you went to sleep earlier than you thought!!

Just because a program does not work the way that YOU expect, or perhaps
would like, does not warrant the out and out rant that you started this
thread with.

All of the information on how to run the program was actually there if
you had bothered to read it rather than wasting time complaining about
nothing.

Jeff

4nec2 hidden variables.
 

On 8/16/2011 2:25 AM, Helmut Wabnig wrote:
On Tue, 16 Aug 2011 09:48:00 +0100, wrote:


Well if you don't actually follow the instructions on the Help then what
can you expect! Don't blame the program and then say the Help is wrong
or not there! RTFM.

"click and hold down your left mouse-button somewhere in the
picture-box" then "Drag the source-object to the middle of the second wire".

Jeff
Of course you are right,
but there are thousands of (unread) help pages left.
Should I read them all?

On YOUTUBE another (not me) user said, he falls asleep
after 4 pages of help text ( the getting started stuff).
I printed the "getting started" but fell asleep after page 3.

w.


Should you read them?

Depends on if you want to make use of the program or not. for ANY
modeling program (and any drafting program, as well) there's a fairly
big first step in the learning curve.

And, because 4nec2 isn't a "drafting" or "drawing" program, it's UI
doesn't follow the more common conventions. It's designed to make
entering antennas easy for some subset of people. 4nec2 started with
just using notepad to edit the NEC input deck and a formatted editor to
edit cards.

Folks who start with EZNEC often find using raw NEC decks difficult,
because they're used to the tabular model entry of EZNEC.

Folks who start with Visio, pre version 5, find modern versions of Visio
a pain, because the UI changed to conform to PowerPoint.

Autocad is certainly not "intuitive" unless you've been using it a while.


For what it's worth, the 4nec2 help files are a whole lot better than
the documentation for NEC itself.

Beyond the details of entering the geometry and simulation conditions
(which are idiosyncratic for ALL antenna modeling programs.. no two are
alike, nor are the input file formats directly compatible, for the most
part)

There's a fair amount of "art" in effectively using the programs and
understanding the limitations inherent in the modeling technique being
used (method of moments on wires for NEC). It's not just a matter of
creating a bazillion segments and letting your processor grind away,
because there's all sorts of subtle numerical precision issues.

4nec2 (and EZNEC) both provide some amount of help in keeping you away
from egregious model errors (segment lengths that are unreasonable
relative to wavelength and/or wire diameter).

There's also well known techniques for modeling flat surfaces with
meshes that provide results that match actual measurement, even if the
model doesn't "look" quite right if you render it.

NEC is very much a power tool like a chainsaw.. you can do things that
you couldn't do with hand tools, but you can also do things that you
don't really want. It takes some practice and background knowledge to
use it effectively.


The value of purpose built modeling tools like AO is that they have a
very constrained model space tuned to a particular application, so the
program can
a) have a simple model entry (e.g. you can only do Yagi-Uda)
b) have a sophisticated objective function for optimization

NEC is a lot harder: you can enter any geometry from a simple wire to an
entire battleship with 100s of antennas.