Building a wire antenna from scratch: wire, insulators, chokes, and checking it with a VNA and a model
- Replies 2
- Views 10
- Created
- Last Reply
A better way to browse. Learn more.
A full-screen app on your home screen with push notifications, badges and more.
This thread is for sorting out the decisions in a homebrew wire antenna. A 40 m dipole is a good example, but most of this applies to inverted-Vs, fan dipoles, end-feds and simple verticals. Parts lists get shared a lot, but the reasoning behind each choice gets less attention. Here is a rough order of operations. Please correct or add to it.
Wire. For a permanent span, 14 AWG hard-drawn copper or copper-clad steel is common. Soft-drawn copper stretches over time, so the resonant frequency drifts and the sag changes. Copper-clad steel resists stretching but is stiffer and harder to coil. For portable or lightweight antennas, 18 to 22 AWG stranded wire is fine. Thinner wire only costs you a little in resistive loss at HF, but it has less tensile strength and a slightly narrower bandwidth. Insulated wire has a velocity factor effect, so it resonates lower than bare wire of the same length. The shift is often a few percent, which is why the standard advice is to cut long and trim.
Starting length. The familiar 468/f(MHz) feet formula gives total dipole length. For 7.1 MHz that is about 65.9 ft, or roughly 33 ft per leg. Height above ground, nearby objects, wire type and sag all move the resonance. Treat the formula as a starting point and plan on adjustment. Fold the extra length back on itself at the end insulators instead of cutting it off, so you can still change it.
Insulators and supports. The center insulator carries the mechanical load of the feedline. Give the coax or ladder line some strain relief so the pull does not land on the solder joint or the wire itself. End insulators can be anything non-conductive and UV-stable: ceramic, HDPE, acrylic, or even sections of PVC. Weather and sunlight are the real enemies. Use rope that tolerates UV and does not stretch much. Where a tree can move, add a pulley and counterweight or a bungee section so the wire does not break in the wind. For fiberglass poles, check what the wire does at the joints. Tape or clamp the wire at intervals instead of letting it flap, and keep poles well away from power lines.
Feedline choices. A half-wave dipole fed with 50 ohm coax and a 1:1 current balun or choke is the simplest single-band setup. A multiband dipole is more flexible fed with 450 ohm window line or ladder line into a tuner. With ladder line the feedline is part of the system, so keep it balanced, away from metal, and avoid running it parallel to other conductors. On the coax side, a choke of coiled coax or a stack of type 31 or 43 ferrite is common. The goal is high common-mode impedance, ideally a few thousand ohms, on the bands you use. A choke is not an impedance transformer. It only suppresses current on the outside of the coax shield.
Traps versus fan dipoles. A trap is a parallel LC circuit resonant at the band you want to isolate. Traps work, but they add loss and narrow the bandwidth, and they need to be weatherproofed and tuned individually. Fan dipoles use several wires from one feedpoint and are easier to build and adjust, but the elements interact. In practice you trim the highest band first, then work down.
Verticals and radials. A quarter-wave vertical for 40 m is about 33 ft tall, using 234/f(MHz) as a starting figure. The radial system decides how much power goes into the air and how much into the soil. Elevated resonant radials, a few of them, are efficient. Ground-mounted verticals often need many radials, commonly 16 or more, to bring ground loss down. A low SWR on a vertical with a poor radial field can still be a lossy antenna, so SWR alone is a weak test of efficiency.
Measuring. An antenna analyzer or NanoVNA is more informative than a basic SWR bridge because it shows resistance and reactance as well as SWR. Calibrate at the point where you connect. Keep in mind that common-mode current on the shield can shift readings, and that lossy coax makes the SWR at the radio look better than at the antenna. Measure with the choke installed, and compare results at the antenna end and the shack end.
Modeling. Both EZNEC and MMANA-GAL let you check a design before cutting wire. Use sensible segment counts, a realistic ground type and the actual wire diameter. Expect the model to match reality only as far as the inputs do. Neither program accounts for common-mode current unless you add the feedline as part of the model. Even so, models are good for comparing options such as height, fan spacing, sloper angle or radial count.
Questions for the thread: what is your usual workflow for trimming and verifying a new antenna? Do you model first, or build and measure? And where do you draw the line between a trap design and a simpler fan?
Link to comment
https://www.hamradiobase.com/forums/topic/11780-building-a-wire-antenna-from-scratch-wire-insulators-chokes-and-checking-it-with-a-vna-and-a-model/Share on other sites