HF Wire Antenna Comparison Shootout: Real-World Testing

Every ham forum has the same endless debate: dipole vs. end-fed vs. G5RV vs. OCF vs. random wire. Everyone has opinions. Few have data. I spent a winter testing five common wire antennas at my station — same height, same feedpoint location, same radios, same propagation conditions — switching between them during QSOs and logging signal reports.

This isn't a definitive scientific study. It's one station, one location, one set of conditions. But it's real-world data that I wish someone had given me before I strung my first wire.

The Contenders

1. Center-fed dipole (66 feet total, fed with coax, for 40m fundamental)
2. End-fed half-wave (EFHW) (66 feet, 49:1 transformer, for 40m + harmonics)
3. Off-center-fed dipole (OCF/Windom) (66 feet, fed at 1/3 point, 4:1 balun)
4. G5RV (102 feet top, 31 feet ladder line, coax to shack)
5. Random wire (84 feet, 9:1 unun, tuner required)

All were hung at approximately 35 feet at the center, with ends dropping to 15-20 feet (inverted-V or sloper configuration depending on available trees). Same antenna supports, same general area of my property.

Test Methodology

Signal Comparison Protocol

• WSPR beacons: Ran WSPR on each antenna for 24-hour periods on 40m, 20m, and 15m. WSPR gives you SNR data from hundreds of receiving stations simultaneously. This is the most objective comparison possible for a single amateur station.

• A/B switching during QSOs: Installed a remote antenna switch. During conversations, I'd switch antennas and ask for comparative signal reports. Subjective but useful.

• Reverse Beacon Network (RBN): Sent CQ on CW on each antenna back-to-back (within 2-3 minutes to keep propagation constant). RBN spots include SNR figures from multiple receivers.

• FT8 PSKReporter: Ran each antenna for equal time periods on FT8 and compared geographic reach and SNR reports on PSKReporter maps.

Limitations

• Propagation varies minute-to-minute. I can't guarantee identical conditions even with quick switching.
• Ground characteristics at my specific QTH affect each antenna differently.
• All antennas were in the same general space, so mutual coupling exists (I ensured only one was connected at a time).
• Height and orientation limitations of my trees determined exact geometry.

Take this as directional data, not laboratory-grade measurement.

Results: 40 Meters (7 MHz)

This is where the dipole is fundamentally resonant and the EFHW is a half-wave. Fair fight.

Antenna Avg WSPR SNR (EU) Avg WSPR SNR (JA) RBN Avg (NA) Tuner Needed? Dipole -14 dB -22 dB 18 dB No EFHW -15 dB -21 dB 17 dB No OCF -16 dB -23 dB 16 dB No G5RV -15 dB -22 dB 17 dB Slight (1.5:1) Random Wire -17 dB -24 dB 15 dB Yes

Analysis: On their fundamental frequency (40m), the dipole and EFHW are essentially equal — within 1 dB, which is measurement noise. The OCF is slightly worse (possibly due to higher-angle radiation from the feed point offset). The G5RV is close to the dipole. The random wire loses 2-3 dB, likely due to tuner losses and less optimal current distribution.

Verdict on 40m: Everything works. Pick based on convenience, not performance anxiety.

Results: 20 Meters (14 MHz)

Now it gets interesting. The dipole is NOT resonant here (it's designed for 40m). The EFHW works as a full-wave antenna (second harmonic). The G5RV was originally designed for 20m.

Antenna Avg WSPR SNR (EU) Avg WSPR SNR (JA) RBN Avg (NA) Tuner Needed? Dipole -18 dB -25 dB 14 dB Yes EFHW -12 dB -19 dB 20 dB No OCF -14 dB -21 dB 18 dB Slight G5RV -13 dB -20 dB 19 dB No Random Wire -16 dB -23 dB 16 dB Yes

Analysis: The EFHW wins on 20m by a clear margin (4-6 dB over the dipole!). Why? At 14 MHz, the 66-foot EFHW is a full wavelength — it has gain. The current distribution creates a pattern with more low-angle radiation than the 40m dipole forced to work on 20m through a tuner.

The G5RV also does well — it's designed for 20m originally. The dipole, working through a tuner as a random length on 20m, shows the loss of non-resonant operation.

Verdict on 20m: EFHW has a genuine advantage over a 40m dipole on this band. If you want a single wire for both 40m and 20m, the EFHW is the better choice.

Results: 15 Meters (21 MHz)

15m is the third harmonic of 40m. The EFHW should work as a 1.5-wavelength antenna.

Antenna Avg WSPR SNR (EU) Avg WSPR SNR (JA) RBN Avg (NA) Tuner Needed? Dipole -16 dB -21 dB 16 dB Yes EFHW -11 dB -17 dB 21 dB No OCF -14 dB -20 dB 18 dB Slight G5RV -15 dB -21 dB 17 dB Yes Random Wire -17 dB -23 dB 15 dB Yes

Analysis: EFHW dominates again on 15m. At 1.5 wavelengths, it develops gain lobes at lower elevation angles — perfect for DX. The dipole working through a tuner on a non-harmonic frequency loses several dB.

The G5RV loses its advantage here — it needs a tuner on 15m and the ladder line section doesn't present optimal impedance.

Verdict on 15m: EFHW is clearly the best multiband performer.

Qualitative Observations

Noise Performance

The EFHW had noticeably MORE received noise on all bands compared to the dipole. This is a known issue — the single-ended feed picks up common-mode noise from the feedline and nearby sources. A good choke at the transformer helps but doesn't eliminate it entirely.

The dipole, being balanced and fed with coax + a 1:1 current balun, was consistently 1-2 S-units quieter on receive.

Net result: The EFHW's gain advantage on TX was partially offset by higher noise floor on RX. In quiet locations, the EFHW wins. In noisy locations (urban/suburban), the dipole's lower noise floor might make it effectively equal.

Bandwidth

The dipole had the widest SWR bandwidth on 40m (about 250 kHz below 2:1). The EFHW was second (about 200 kHz). The OCF claimed multiband operation but needed a tuner on most bands for SWR below 2:1.

Ease of Installation

• Dipole: Needs center support + two end supports. Coax to center is straightforward.
• EFHW: One high end, one low end, feedpoint at one end (convenient if a tree is at the property edge). Simpler to install in many situations.
• OCF: Like a dipole but with offset feed — needs center support at the 1/3 point. Slightly more complex rigging.
• G5RV: Needs center support high enough to accommodate the ladder line drop. The ladder line must hang freely and not contact metal. More fussy installation.
• Random wire: Easiest. Run it anywhere, any direction, any length. Just need a 9:1 unun and a tuner.

Reliability

Over one year:

• Dipole: zero failures
• EFHW: transformer had water ingress after 8 months (replaced with sealed unit)
• OCF: balun heated during high-power operation on non-resonant bands (derated to 500W)
• G5RV: ladder line broke in ice storm (replaced)
• Random wire: unun worked perfectly all year

The Verdict

If You Can Only Have ONE Wire Antenna for HF

End-Fed Half-Wave (EFHW) for 40m (66 feet). It gives you:

• Resonant operation on 40m, 20m, 15m, and 10m without a tuner
• Genuine gain on the harmonic bands (20m, 15m)
• Single feedpoint at one end (easier installation for many setups)
• Adequate performance on other bands with a tuner

The tradeoff: higher receive noise (use a choke) and the transformer is a potential failure point.

If Noise Is Your Primary Problem

Center-fed dipole with quality 1:1 current balun. Lower noise pickup, balanced feed rejects common-mode interference. You'll need a tuner for bands where it's not resonant, and you lose some gain on harmonic bands, but the quieter receive may net you more contacts in noisy environments.

If You Want Simplicity and Don't Care About Optimization

Random wire + 9:1 unun + good tuner. It works on every band. It's never great on any band, but it's never terrible either. And installation is dead simple — run wire wherever you can, in whatever direction works.

The G5RV

An excellent antenna for 20 meters specifically. On other bands, it's mediocre-to-acceptable. If 20m is your primary band, it's a solid choice. As a general-purpose multiband antenna, the EFHW has surpassed it.

The OCF Dipole

Promises multiband operation without a tuner. In practice, it needed a tuner on most bands except 40m and sometimes 20m. The feed system complexity (4:1 balun, specific feed point) adds failure points without clear performance advantage over simpler options.

Final Thought

The differences between wire antennas at the same height are smaller than most hams expect. On their fundamental resonant frequency, all five antennas were within 3 dB of each other. The EFHW's harmonic advantage is real but modest (3-5 dB on 20m/15m).

Height matters more than type. A dipole at 50 feet beats an EFHW at 25 feet every time. Spend your energy getting wire UP rather than agonizing over which wire design is theoretically optimal. Any wire antenna at a reasonable height works the world.