Building a wire antenna from scratch: where do you spend effort first, and what do you verify before trusting the SWR?
- Replies 2
- Views 54
- Created
- Last Reply
Top Posters In This Topic
-
Admin_Terry 1 post
-
Mod_Dan 1 post
This is a discussion starter for anyone building HF wire antennas, dipoles, inverted-Vs, or quarter-wave verticals from raw materials. I'd like to compare notes on which parts of the build matter most and which are habit. Here is a framework, and then some questions.
Wire and length. A half-wave dipole in free space is roughly 468/f(MHz) feet overall, which is about 66 ft for 7.1 MHz. Real installations come out a few percent different because of height, nearby objects, wire diameter, and insulation. Insulated wire usually resonates lower than bare wire of the same length, so cut long and trim. Common choices are 14 AWG hard-drawn copper, 12 AWG for longer spans, and copper-clad steel where stretch is a problem. Soft-drawn copper stretches over time and drifts your resonance and your sag. Stranded wire is easier to handle and tolerates flexing better at the feedpoint.
Insulators and supports. End insulators only need to be UV-stable, mechanically strong, and not hygroscopic. At 100 W the voltage at a dipole tip is not what stresses cheap plastic. Weather and tension are what kill it. At the center, the most common failure is mechanical: coax weight pulling on a soldered joint. A center insulator with strain relief, a drip loop, and a sealed coax end all prevent trouble later. Rope should be UV-resistant, and a pulley or counterweight at one end lets the antenna move in wind instead of breaking.
Balun or choke. A dipole fed with coax wants a current balun or choke at the feedpoint. Without one, the coax shield carries common-mode current, which distorts the pattern, brings RF into the shack, and changes the SWR reading depending on how the cable is routed. Choke options include a coil of coax, ferrite beads or toroids on the coax, or a purpose-built 1:1 current balun. What matters is choke impedance on the bands you use, not whether the choke is called a balun. A choke that works on 20 m can be nearly useless on 80 m, or can heat up on the wrong mix.
Ladder line. Open-wire or window line feeding a dipole through a tuner is an old and effective way to cover several bands with one wire. Losses stay low even when SWR on the line is high. Keep the line away from metal, bring it away from the antenna as perpendicular as practical, and use a tuner with a good balanced output or a quality 4:1 or 1:1 balun at the tuner. Some line lengths on some bands produce very high voltage or current at the tuner, so it helps to model the actual combination or try a few lengths.
Verticals and radials. A quarter-wave vertical element is about 234/f(MHz) feet. The ground system matters as much as the element. Elevated resonant radials, a few of them, behave very differently from many short radials laid on the ground. A vertical with sloping radials tends toward a feedpoint impedance nearer 50 ohms, while radials at right angles give a lower figure. Buried or ground-laid radials work through quantity, so more helps up to a point of diminishing returns. A low SWR on a vertical tells you little, because lossy ground can produce a decent match and weak signals together.
Fiberglass poles. Telescoping fiberglass poles are popular for vertical wires and inverted-Vs. Check that the pole is actually fiberglass, since carbon fiber conducts and will detune a wire taped to it. Tape or tie the wire so it is not carrying the pole's bending load. Wind loading on a lightweight pole grows quickly with height, so guy it if it will stand unattended.
Traps and tuned multiband designs. A parallel LC trap isolates the outer part of a dipole at its resonant frequency and acts as loading on lower bands. The tradeoffs are narrower bandwidth, some loss, mechanical weight, and traps that must be tuned and weatherproofed. The usual alternatives are fan dipoles, which interact with each other, and a single wire on ladder line with a tuner. None of them is automatically better.
Measuring. An antenna analyzer or NanoVNA is much more informative than a single-frequency SWR reading, because it shows where resonance really is and how wide the usable range is. Measure at the feedpoint if you can. Long, lossy coax makes SWR look better than the antenna deserves. Short test cables and calibration matter on a VNA. If readings change when you touch or move the coax, you probably have common-mode current.
Modeling. EZNEC (NEC-2 engine) and MMANA-GAL (MININEC-based) are both useful before you cut wire. Model height, ground type, wire diameter, and segmentation carefully, since modeling errors around junctions and low heights can mislead. Use models to compare options and see trends, and expect to trim the real antenna.
Questions to start the thread: when you build, do you prototype with long wire and trim, or model first? What choke have you found acceptable for multiband coax-fed dipoles? And at what point do you stop adding radials on a vertical and decide the extra work isn't worth it?
Link to comment
https://www.hamradiobase.com/forums/topic/11074-building-a-wire-antenna-from-scratch-where-do-you-spend-effort-first-and-what-do-you-verify-before-trusting-the-swr/Share on other sites