Gidday

Apologies for being the newbie in this. I'll admit I am after a quick fix!

Am playing with modelling a 20M single quad loop. Yes I have built one
before with a 75r section and it works fine.

The problem I have is how to represent the 1/4 piece of coax on 4NEC2. A
piece of transmission line seems to have to be connected to a "wire" at
each end and I get an error if I just have one hanging in free space
where the feedpoint is.

I am aware I can do it at the Smith Chart output but I'd prefer to be
able to graph actual Z/SWR.

Thoughts?

Cheers Bob VK2YQA/WA5

The problem I have is how to represent the 1/4 piece of coax on 4NEC2. A
piece of transmission line seems to have to be connected to a "wire" at
each end and I get an error if I just have one hanging in free space where
the feedpoint is.

I am aware I can do it at the Smith Chart output but I'd prefer to be able
to graph actual Z/SWR.

Thoughts?

Cheers Bob VK2YQA/WA5

Bob, I don't know much about quad loops, so not sure what you are doing. Is
the 1/4 wave coax just to transform the single loop input impedance to a
lower value? I think a loop has about 120 ohm input Z, so 1/4 wave 75 ohm
coax transforms it to near 50 ohm.

The transmission lines in NEC are ideal, so you will not get any loss as
with a real transmission line. You do have to have a piece of wire at each
end. The loop end wire is the loop, and the feed point end wire must
contain the NEC source.

Do I have the right idea, or am I completely out to lunch?

Regards,

Frank

Hi Frank

Yes the usage of the 1/4 wave section is simple and well known. I have
been doing it for years. The loop Z varies somewhat depending on shape
and height over ground. I am modelling a triangle and it runs around 145r

When I create the "wire" in free space by istelf I get a warning;
Error: Wire 5, seg 91, Ex-src: Not allowed near open wire-end

Mind you it comes up with seemingly the correct result. If I change the
number of segments in that wire from 1 to 3 the error doesnt occur.

More feeling my way at this...

What I'd really like to do is setup a dual band single quad loop (ie one
loop inside the other) for (say) 20 and 40M with no ATU. ie direct coax
feed...

Cheers Bob


Frank wrote:

Bob, I don't know much about quad loops, so not sure what you are doing. Is
the 1/4 wave coax just to transform the single loop input impedance to a
lower value? I think a loop has about 120 ohm input Z, so 1/4 wave 75 ohm
coax transforms it to near 50 ohm.

The transmission lines in NEC are ideal, so you will not get any loss as
with a real transmission line. You do have to have a piece of wire at each
end. The loop end wire is the loop, and the feed point end wire must
contain the NEC source.

Do I have the right idea, or am I completely out to lunch?

Regards,

Frank

Ok Bob, now I understand what you are doing. I am familiar with using a
quarter wave stub for matching. What I did not understand is the type of
error message you were getting.

I use a different version of NEC 2 (Nittany Scientifics' NEC-Win Pro). It
does not return an error message when attempting to use a single segment
wire, with a source, to feed a transmission line model. I don't think you
need worry about the message, since NEC will return the correct solution.
You can always experiment with the "TL" card by placing a known load at one
end, and observing the input impedance to see that you get the expected
value.

I have attempted modeling a double extended zepp, with a source at one end,
and the results are completely wrong so guess the error message is required
to cover such events. Constructing a physical transmission line model
overcomes the double extended zepp problem.

Typical NEC code dealing with the "TL", as per L.B. Cebik, in his book
"Basic Antenna Modeling, a Hands on Tutorial" is as follows":

..
..
GW 1 21 -3.418 0 15 3.418 0 15 0.000814
GW 2 1 -0.1 0 5 0.1 0 5 0.000814
..
..
TL 1 11 2 1 75 15 0 0 0 0
..
..

73,

Frank


"Bob Bob" wrote in message
news
Hi Frank

Yes the usage of the 1/4 wave section is simple and well known. I have
been doing it for years. The loop Z varies somewhat depending on shape and
height over ground. I am modelling a triangle and it runs around 145r

When I create the "wire" in free space by istelf I get a warning;
Error: Wire 5, seg 91, Ex-src: Not allowed near open wire-end

Mind you it comes up with seemingly the correct result. If I change the
number of segments in that wire from 1 to 3 the error doesnt occur.

More feeling my way at this...

What I'd really like to do is setup a dual band single quad loop (ie one
loop inside the other) for (say) 20 and 40M with no ATU. ie direct coax
feed...

Cheers Bob


Frank wrote:

Bob, I don't know much about quad loops, so not sure what you are doing.
Is
the 1/4 wave coax just to transform the single loop input impedance to a
lower value? I think a loop has about 120 ohm input Z, so 1/4 wave 75
ohm
coax transforms it to near 50 ohm.

The transmission lines in NEC are ideal, so you will not get any loss as
with a real transmission line. You do have to have a piece of wire at
each
end. The loop end wire is the loop, and the feed point end wire must
contain the NEC source.

Do I have the right idea, or am I completely out to lunch?

Regards,

Frank

Tnxs Frank!

Frank wrote:
Ok Bob, now I understand what you are doing. I am familiar with using a
quarter wave stub for matching. What I did not understand is the type of
error message you were getting.

Hello Bob,

the message "Error: Wire 5, seg 91, Ex-src: Not allowed near open
wire-end" is not correct. At least not when a TL is connected to the
wire. May I ask what 4nec2 version you were running, because I thought
I had solved this 'problem' already some time ago.

Thanks in advance,
Arie.

Hi Arie

5.5.2

Will also be emailing you privately about some feedback running the pgm
under Linux Wine

Cheers Bob W5/VK2YQA

Arie wrote:

Hello Bob,

the message "Error: Wire 5, seg 91, Ex-src: Not allowed near open
wire-end" is not correct. At least not when a TL is connected to the
wire. May I ask what 4nec2 version you were running, because I thought
I had solved this 'problem' already some time ago.

Thanks in advance,
Arie.