EZNEC question - 17m Vertical

[KE5AVC]

I have been playing with the demo version of EZNEC 4.0 to model a 17m vertical that I plan to build.

If the model is built with a single element of uniform diameter, the resulting length for resonance at 18.139 MHz is around 156", close to that obtained with the standard 234/f formula for a quarter-wavelength in feet.

The actual antenna will consist of a 1" diameter copper pipe topped with a 102" whip (0.125" dia.), and I have already worked out the mechanical details of this arrangement.

When this configuration is modelled, the length for resonance at the same frequency is some 20" longer or around 175".

This result seems out of line, and I wonder if is expected when using different diameter sections for the radiator.

Looking for some comments before cutting any pipe.

Charles
KE5AVC

[WIRELESS]

The moral here is never use a software program or formulas to cut exact lengths for antennas.  Always cut longer and trim.

[N3UMH]

I've got to disagree with WIRELESS's statement as a general one.  I've modeled several antennas that work very much like the model.

In this particular case, though, I am going to have to agree with him and say cut the conduit piece long.  Maybe you can trim the bottom of it.

You could probably work the model difficulties out (and I do think you're having model difficulties) but I don't know if you can with the demo version.  

If you want to work on the model for modeling reasons, look at the antenna modeling tips at www.cebik.com, to get an idea of the limitations of the program.

If you're more interested in the final antenna, cut and try is going to be easier and faster in this case.

73,
Dan
N3OX
www.n3ox.net

[KE5AVC]

The antenna has been built, here's some addtional information.

I started with the original 154" length using 51" +/- of 1" copper pipe topped with the 102" (actually 102.75") whip.

The antenna was made with an N-connector at the base, as that was what I had on hand. Before running the feedline to my shack, I attached a temporary line in the form of 15' of 0.5" heliax.

I dragged my TS-480 and 12V supply outside and set them up on a Black & Decker Workmate at the base of the antenna.

I was able to load the antenna on 17m, but the SWR without the built-in tuner was >3:1. It worked better on 15m, somewhere around 2.5:1.

With this in mind, I added a 1"-to-3/4"coupling and an additional 20" of 3/4" pipe. Total length now ~176"

This lowered the SWR to < 1.3:1 on 17m, and I can still load it on 15m with the tuner, as well as the lower part of 6m on its own!

Heard some activity on 17m, and my first contact was a VP5 station in Turks and Caicos Is., running about 50W.

The EZNEC model ended up being darned close -- I'm impressed.

Charles
KE5AVC

[WA7NCL]

The old formula that you used takes the free space wavelength at the frequency and shortens it by 5% to account for a rule of thumb end effect due to size of the conductor.  It is just an old rule of thumb and should not be taken as any more accurate than modeling.

You didn't say what criteria you used to consider the antenna "tuned".  You can get different results if you use minimum SWR or 0 reactance depending on the real part of the drive impedance (radiation resistance).  

Also please note that the number of segments you use for the conductor has a big affect on the accuracy of feed point impedance computation.  That is because accurate modeling of the current distribution is necessary for accurate impedance calculation.  Very few segments will result in an adequate computation of the far field.

Don't let some of the modeling detractors discourage you from using it as a tool to guide your efforts.  Formulas and modeling are guides to reduce the amount of cut and try in antenna work.  They don't eliminate it.