On 10/12/2014 6:54 AM, Lostgallifreyan wrote:
"Ralph Mowery" wrote in
:


"Lostgallifreyan" wrote in message me, and is the
one I need most to solve in my next efforts). What is this NEC program,
and
where can I find it? Sorry, but I have to ask, or Google will likely
flood me
with Nippon Electric Company details.

Start here for the explination. Near the bottom are some places to
download it. There are many versions and variations by slightly
differant names. Some free and some you have to buy.
http://en.wikipedia.org/wiki/Numeric...magnetics_Code


Thankyou. Fortunately this looks like it will be easier than finding certain
versions of the CGG compiler...


Also try EZNEC. They both use the NEC engine but EZNEC provides a
different interface to the modeling.

http://eznec.com/

John S wrote in :

Also try EZNEC. They both use the NEC engine but EZNEC provides a
different interface to the modeling.


Already did. EXNEC v5 demo. I really like the first impression I get, Runs
on all Win32 (very respectable), and I like the way I can grab the 3D plot
and rotate it like a SketchUp image to get a proper look at it. I mentioned a
laser beam tool (called PSST) that graphically models laser cavities, hoping
this EXNEC might be an antenna maker's equivalent. It is. It will take me
a while to understand it though.

On 10/12/2014 7:38 AM, Lostgallifreyan wrote:
John S wrote in :

Also try EZNEC. They both use the NEC engine but EZNEC provides a
different interface to the modeling.


Already did. EXNEC v5 demo. I really like the first impression I get, Runs
on all Win32 (very respectable), and I like the way I can grab the 3D plot
and rotate it like a SketchUp image to get a proper look at it. I mentioned a
laser beam tool (called PSST) that graphically models laser cavities, hoping
this EXNEC might be an antenna maker's equivalent. It is. It will take me
a while to understand it though.


These are complex applications which someone with a mindset not like
yours has produced. There will be a learning curve. The important thing
is to not get frustrated and give up. If you need help, ask.

Cheers.

John S wrote in :

These are complex applications which someone with a mindset not like
yours has produced. There will be a learning curve. The important thing
is to not get frustrated and give up. If you need help, ask.


Understood. Thanks. bear in mind that my mindset did manage to get to grips
with phase modulation synthesis, large scale polyphony and multitimbrality,
and accurate similautions of a few musical instruments, and did do alone with
nothing but expired patents and service manuals, and a very few other
published notes to guide me. I'm not trying to show off, my point is that
while I do not give up easily, I AM alone, so my time is divided. If after
my current break from that I go back to it, I might be listening to radio on
an antenna or two while not thinking of the antennas for months. I've always
been interested in this though, so I won't let it go. I'll look into NEC
because the whole thing about good grounds is a particular fascination now,
especially if improvising in limited or very temporary circumstances.

On 10/12/2014 11:14 AM, Lostgallifreyan wrote:
John S wrote in :

These are complex applications which someone with a mindset not like
yours has produced. There will be a learning curve. The important thing
is to not get frustrated and give up. If you need help, ask.


Understood. Thanks. bear in mind that my mindset did manage to get to grips
with phase modulation synthesis, large scale polyphony and multitimbrality,
and accurate similautions of a few musical instruments, and did do alone with
nothing but expired patents and service manuals, and a very few other
published notes to guide me. I'm not trying to show off, my point is that
while I do not give up easily, I AM alone, so my time is divided. If after
my current break from that I go back to it, I might be listening to radio on
an antenna or two while not thinking of the antennas for months. I've always
been interested in this though, so I won't let it go. I'll look into NEC
because the whole thing about good grounds is a particular fascination now,
especially if improvising in limited or very temporary circumstances.


Excellent! I like to explore and I encourage everyone to do so whether
it be with math tools or getting your hands dirty. Keep it up.

Cheers.

John S wrote in :

Excellent! I like to explore and I encourage everyone to do so whether
it be with math tools or getting your hands dirty. Keep it up.


Indeed. That synth I mentioned, it took me a good chunk of lifetime
waiting for someone to make so I could pay for it. I gave up waiting.

Jeff just posted about the perils of antenna towers and hillsides vs NEC2's
flat earth, and many earlier things (liek modellign with Sketchup prior to
engineering in metals and plastics, have taught me caution. measure thrice,
cut once. THis is how and why I want to use NEC, to save me from falling into
expensive and time consuming traps, which I think it will do very well.

On 10/12/2014 12:09 PM, Lostgallifreyan wrote:
John S wrote in :

Excellent! I like to explore and I encourage everyone to do so whether
it be with math tools or getting your hands dirty. Keep it up.


Indeed. That synth I mentioned, it took me a good chunk of lifetime
waiting for someone to make so I could pay for it. I gave up waiting.

Jeff just posted about the perils of antenna towers and hillsides vs NEC2's
flat earth, and many earlier things (liek modellign with Sketchup prior to
engineering in metals and plastics, have taught me caution. measure thrice,
cut once. THis is how and why I want to use NEC, to save me from falling into
expensive and time consuming traps, which I think it will do very well.


Jeff seems to be a very experience man with RF. I don't doubt his
comments about what is possible with a practical situation. I, however,
am playing with some fundamentals (free space, antenna characteristics,
etc) so that I understand how to catch up to the gurus here. My toy is
the modeling program. I do like to compare theory to practical
situations, however.

On Sun, 12 Oct 2014 12:03:43 -0500, John S
wrote:

Excellent! I like to explore and I encourage everyone to do so whether
it be with math tools or getting your hands dirty. Keep it up.

This might be useful:
http://www.amanogawa.com
Java required. Among the "Linear Antenna" animations is a common
dipole model, which shows the effects of different element lengths:
http://www.amanogawa.com/archive/DipoleAnt/DipoleAnt-2.html
I think (not sure and too lazy to RTFM) that "directivity" means gain
linear gain.

More NEC modeling tools:
http://wb0dgf.com/nec-mininec.htm

I'm partial to 4NEC2.
http://www.qsl.net/4nec2/
It comes with a huge collection of sample antennas suitable for
analysis and plagiarizing. Most of my stuff was done with 4NEC2:
http://802.11junk.com/jeffl/antennas/
(Not everything shown is my work). There are plenty of tutorials on
the web and videos on YouTube showing how to get started with 4NEC2.
https://www.youtube.com/results?search_query=4nec2
There is also a multi-threaded engine for speeding up 4NEC2. Highly
recommended:
http://users.otenet.gr/~jmsp/

If you are into reverse engineering commercial and ham antennas to see
how they work, there are a number of collections online.
http://www.hdtvprimer.com/ANTENNAS/comparing.html
http://www.hdtvprimer.com/SIMS/
http://www.qsl.net/kp4md/modeling.htm
http://www.arrl.org/antenna-modeling-files
If you want to go Googling for .NEC files, try something like this:
https://www.google.com/#q=filetype:nec+antenna
https://www.google.com/#q=filetype:ez+antenna
Some of the commercial antennas I've modeled turned out to be not very
good such as this 2.4GHz yagi:
http://802.11junk.com/jeffl/antennas/mfj1800/
(Hint: It's a 200 ohm feed, not 50 ohms).

One important "trick" is to always run an average gain test on your
design. This is a sanity check to make sure your model somewhat
resembles reality. There are at least two examples of antennas on my
web pile that do NOT pass this test and are therefore impressive
looking garbage. The results should be 1.0 for most antennas.
http://fornectoo.freeforums.org/run-average-gain-test-t355.html
http://www.arrl.org/files/file/Antenna%20Modeling%20for%20Beginners%20Supplementa l%20Files/EZNEC%20Modeling%20Tutorial%20by%20W8WWV.pdf

Please do NOT get hung up on using just one modeling program. While
the others have different interfaces and file formats, they are often
better at doing specific things. For example, I use several yagi
design programs to generate the dimensions and an NEC2 deck. I then
use EZNEC or 4NEC2 to more closely look at the result, and to add
support structures, real grounds, tapered elements, etc. While you
can buy programs that will do it all, you don't want to know the
price.

Testing antennas is also somewhat of an art. A pile of test equipment
is always nice:
http://802.11junk.com/jeffl/pics/home/slides/test-equip-mess.html
For the price of an HF radio, you can buy various "antenna analyzers"
and "network analyzers[1]" that will do a good job of emptying your
bank account. Being cheap, I just use a "return loss bridge", RF
sweep generator, and an oscilloscope. The RLB can be easily built.
These are nice:
http://www.wb.commufa.jp/ja2djh/html/e_rlb.html
Mine are ugly so no photos. There's not much to it. Just plan on
blowing out a few diodes:
http://www.qsl.net/n9zia/wireless/appendixF.html#11
http://www.qsl.net/n9zia/wireless/pics/rtrn_loss_bridge.png
What limits an RLB is that it only shows VSWR or return loss. There's
no indication if the antenna is capacitive or inductive. That's not a
real limitation as your radio could care less if the 50 ohms it sees
is resistive, capacitive, inductive, or a combination of these, as
long as the conglomeration works out to 50 ohms.

The bad news is that the RLB has to be near the antenna feed point in
order to get useful results. That usually means climbing the tower.
You can try using it at the xmitter end of the feed line, but I don't
recommend it. I recommended to one former friend how to install a
coax switch at the feed point, with one port going to an RLB. That
worked splendidly until just after he applied 500 watts to the
antenna, and fried the RLB diode. Make sure the switch has sufficient
isolation and that the RLB input is shorted when not in use.

The sweep generator is probably the most expensive part of the puzzle.
The frequency range you want to cover is always important. The HF
range is covered nicely by cheap DDS (direct digital synthesizer)
modules found all over eBay. For example:
www.ebay.com/sch/i.html?_nkw=dds+module
Add a PIC controller and output amplifier, and you have a sweeper. Or,
just buy one:
http://www.ebay.com/itm/121362216469
One problem with these is that they don't have frequency markers, but
that can be added later. For VHF and UHF, commercial equipment is
probably best. If you can find cable TV test equipment, it will
usually get you to about 400 MHz. That's three HP8620 sweepers in the
pile on the left, none of which work perfectly:
http://802.11junk.com/jeffl/pics/home/slides/BL-shop5.html

For the oscilloscope, you'll need something that is DC coupled. That
eliminates most PC sound card based software. For running an RLB, a
cheap 2.0 MHz DSO (digital storage scope) works just fine. The tiny
pocket size scopes found all over eBay will work, but the screens are
too small to see any detail. That might work if you can unload screen
captures to a computah and/or printer.

That leaves the question of why bother modeling when you can just
cut-n-try an antenna until it works? If you want to make one of
something, cut-n-try is a very good and efficient way of getting one
of anything to work. However, if you're planning on making more than
one, or publishing your design for others to copy (or steal), then
some clues as to the theoretical possibilities of the design are going
to be needed. A few years ago, I worked on a UHF antenna that was
sufficiently high in gain, and therefore narrow band, that customers
had to order it by the frequency of operation. I beat the NEC2 model
to death until everything possible was thrown into the model. U-bolts,
mounting hardware, pipes, yard arms, screws, nearby power wires, other
antennas, vehicles, buildings, etc. It was massive overkill, but very
necessary. When field tests were run on the first few installs, the
measured radiation patterns and return loss graphs were perfect. You
can't do that without a computah model.



[1] Incidentally, I've seen this problem a bit too often with MFJ
antenna analyzers and some VNA's.
http://802.11junk.com/jeffl/MFJ-269-repair/
If you own an MFJ analyzer, I suggest you stock up on matched diodes
for the inevitable rebuild.

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

Jeff Liebermann wrote in
:

This might be useful:...

It it. I saved the whole post, to be sure I can find it fast any time. I
do have a habit of staying with one program so I'll be careful about that.
I'm already aware that NEC2 and NEC4 and MiniNEC all have things they are
particularly good at, so I'll not limit my choices the way I usually might.

I need to risk asking something that maybe should be obvious to me, just to
be sure I'm not starting on a very wrong assumption.. As I'm likely to only
be listening, when I see a Smith chart or other diagram indicating relative
signal transmission strength at some distance and direction, does this follow
the rule of many 'simple' transducers, in that the same chart exactly models
the sensitivity of the same antenna for reception? I'm hoping it does,
otherwise life might get complicated.