View Single Post
  #21   Report Post  
Old May 24th 05, 05:02 AM
Wes Stewart
 
Posts: n/a
Default

On Sun, 22 May 2005 21:26:49 -0400, "Tam/WB2TT"
wrote:


"Wes Stewart" wrote in message
.. .
On Sun, 22 May 2005 17:41:22 -0400, "Tam/WB2TT"
wrote:
[snip]

CJ,
I ran an EZNEC simulation on a pair of crossed dipoles at 3.7 and 4.5 MHz,
with a 7.2 MHz inverted V below the 3.7. Not good. The 7.2 works OK, but
the
4.5 really messes up the 3.7. Among other things the impedance at 3.7 and
4.5 MHz is about 115 Ohms. I ran the simulation at 50 feet above average
ground. If the 4.5 is for receive only, I would forget about it, and take
the 10:1 SWR that the 80 meter dipole gives you.


I'd be interested to see your Eznec file.


Gladly, but, how do I get an ASCII file out of it. I thought I had opened
EZNEC files in Notepad, but it is garbage. Also, even from within EZNEC when
I click on View File, I get the same garbage as Notepad. The only
questionable thing I did would be the 0.2 foot stub in the middle that the
other wires are connected to; but that works for other configurations.


Tam was kind enough to send me his EZNEC file. I asked for this file
since I went through this myself unsuccessfully (so far) and wanted to
see if someone else had a better idea.

My suspicions were confirmed: there are some NEC guideline violations.
I'm don't want this to come off as picking on Tam publicly, but to
keep others from making the same mistakes, (*I* would never make a
mistake [g]) here's what I see.

Tam uses a single wire at the middle of the dipole to contain the
source. This is necessary because NEC doesn't allow a source at a
wire junction. (EZNEC has a split source that works at a wire
junction, but not a multiple wire junction.)

The EZNEC manual recommends that this wire be at least 0.02
wavelengths long and consist of three segments. I believe the three
segment rule is stipulated to force the segments adjacent to the
source wire to be equal to it. If we are careful about this, we can
go to one segment.

Tam's center wire is only 0.2 feet long and has 11 segments. The
segment length of .018 feet is way too short; the wire needs to be
~16' long at 3.5 MHz to meet the guidelines.

Of course with this gross violation comes another one when the
connecting wires are a more appropriate segment length.

Tam, I suggest that you got to Options--Segmentation Check and make
it "Auto".

I'm still looking at this issue and will report back if I see any
progress.

Wes N7WS