Hi Gedas,
If I have read you correctly......
First, set your ground descriptions to "very poor" or "extremely poor". In
reality, ground losses on 630m will be huge and unavoidable, so one might
as well take advantage of it in the model. You are only three HUNDREDTHS of
a wavelength above ground, roughly the same as your 40 meter dipole was 4
feet off the ground. How well-performing, and how decoupled with ground
would you expect that to be.
For this one, you MUST include the feedline as a conductor the same
diameter as the feedline stripped of jacket, plus insulation same type and
thickness as coax jacket.
Then you will need to do something serious in reality to block the feedline
shield, which otherwise will become the principal radiator, radiating 90%
or more of the power. If you don't then you have to include in your model
every conductor connected to the coax shield, transceiver chassis, table
ground buss, including the ground itself, only done reasonably in NEC4. You
will most likely need a double precision engine to get it to calculate
right.
I can have very short segments that the program calls "too short" for the
feed. That is done with the long wire either side of center then with THREE
very short wires (1 foot is good) between the two long ones. Complete a
square box by putting three more wires the same length as the center short
wire. You can think of it as a box with two short wings coming off the top
of the box.
Make wires 2,3,4,6,7,8 all three segments. Set segments in 1 and 5 long
enough to turn off the warning.
wire 1 2 3 4 5
===================+==+==+==+======================
6 | |7
+==+
| 8
W |
i |
r |
e |
|
9 |
|
Put the source in wire 8 center segment, and a part of the coil as a load
in wires 2,3,4. Adjust the size and ratios of coils in 2,3,4 for resonance
at 50 ohms at the source in wire 8.
Wire 9 is the coax shield. Segment 1 gets a load that is the coax block.
Wire 9 last segment, set a load in it to 25 ohms and connect it to ground.
You can detect the model's sensitivity to grounding if varying the R in the
last segment load significantly changes the source reading from wire 8.
You screw around with the R and L in wire 2,3,4 turns to match reality for
the R in a coil of X turns producing a certain uH. and there's nothing that
says that L in wire 2 load has to be the same as wire 4.
As to reality, a good common mode choke will be REQUIRED for your actual
results to make any sense at all. Read and reread
http://k9yc.com/630MTXChokes.pdf until you know it cold. Make it using his
specs. Don't screw around with his specifications. He's very thorough and
has already done the screwing around for you in advance and measured the
results.
His graphs will allow you to create additions to the model which use his
MEASUREMENTS on the choke, which you can put in a load in the first (top)
segment of wire 9.
A note, based on tough experience with fringe problems where ground is
important, use of NEC2 (especially single precision engines) can have
scattered killer issues that only go away when you pay the price, and get
EZNEC Pro4 and the NEC4 license. It's less than you would pay for a third
tier transceiver, and W7EL will take your questions because he knows you're
serious. BTW, Roy's help doc in the Pro4 version has gotten really good
over the years and is always the first place to go. If it ain't working
like you think it should, FIRST thing go dig in his help doc. I may have
preserved an entire 24/7 month or two of my life doing that.
Hope this has been of some help.
73, Guy K2AV
On Mon, Mar 4, 2019 at 4:01 PM Gedas <w8bya@mchsi.com> wrote:
> I was wondering if I could get some help developing then modeling a
> short dipole for 630m use.
>
> I consider myself to have intermediate modeling skills and have modeled
> a lot of antennas in EZNEC etc.
>
> Here is what I would like to do. Take an existing 160m dipole (flat top)
> at 70' and use a pair of inductors, one on each leg about 1-2 feet away
> from the center insulator. I have already modeled this and so far things
> seem normal. My model was simplistic and did not include any coax. For
> the loads I simply increased the number of segments to 35 in each leg
> such that the load ended up being within several % away from the
> feedpoint. I had specified that the loads be at the ends (0%) but in
> reality EZNEC put them several % out.
>
> Now we all know given this tiny antenna for this band the feedpoint
> impedance is going to be low.....in the ball-park of 12 - J30 is about
> the best I could achieve for the lowest SWR at 474 kHz. Not quite
> resonant but close enough.
>
> Now, I seem to remember an old trick that I saw many years ago which I
> think used to be called a gamma coil at the center of the dipole,
> basically across the coax right at the feedpoint that will help bring
> the feedpoint impedance up to 50 ohms. Similar to what many mobile whip
> antennas have at their feedpoint.
>
> So I guess I have several questions that I need help with. One is help
> figuring out what this center inductor looks like in reality, what value
> it should be, _and how is it to be connected to the dipole and the
> coax_. And two, how to model such a beast along with, say, 150' of RG-8
> coax.
>
> In my basic model I simply used one segment for the center and is where
> the feedpoint is within EZNEC, then I have two additional wires
> connected off this short center wire, one on each side of that center
> segment that makes up the actual dipole.
>
> Because of the frequency each segment has to be nearly 2' long ! This
> too is where I need help and that is in attaching not only my coax feed
> but this center inductor in the model. I hope the above makes sense. TU
>
> Gedas, W8BYA
>
> Gallery at http://w8bya.com
> Light travels faster than sound....
> This is why some people appear bright until you hear them speak.
>
>
> _______________________________________________
> Antennaware mailing list
> Antennaware@contesting.com
> http://lists.contesting.com/mailman/listinfo/antennaware
>
_______________________________________________
Antennaware mailing list
Antennaware@contesting.com
http://lists.contesting.com/mailman/listinfo/antennaware
|