I had a feeling that my 20 m EFHW vertical performed better towards the end of the day, while the longer 80 m EFHW came into its own in the morning.

Feelings are useful for deciding what to investigate. They’re less useful for declaring an antenna brilliant or bloody useless. So I let both antennas listen together and compared the logs.

The result? My hunch had some support. The vertical had a clear evening advantage, the EFHW80 edged ahead during the early morning recovery, and later in the morning they were almost level.

The interesting bit is why.

Let the Flex do the listening

Two slices on the Flex 8600M monitored FT8 on 14.074 MHz simultaneously, one antenna on each input. Separate WSJT-X instances recorded the decodes.

That allowed us to compare the same transmission, at the same moment, rather than switch antennas and hope the propagation hadn’t changed while I was fiddling.

We also counted stations heard by only one antenna. Those matter: a signal that never decodes on the other antenna cannot contribute a paired SNR measurement.

Across the completed collection, we found 19,517 matched messages, with 966 identifiable stations contributing to the station-balanced SNR comparison. Each station’s results were summarised before calculating the overall median, so one enthusiastic operator repeatedly calling CQ couldn’t dominate the verdict.

The laptop interrupted proceedings by shutting down. Its missing recording periods were excluded. An absent log is not evidence that the ionosphere has packed up.

What actually happened

Between 20:00 and 23:00 UTC, the vertical recorded 38.9% more messages, heard 480 stations against 369, and had a 3 dB median advantage in the paired, station-balanced comparison.

During the 05:00–06:15 UTC morning recovery, the EFHW80 recorded 9.4% more messages, heard 244 stations against 219, and led the paired comparison by 1 dB.

Later, between approximately 07:51 and 09:34 UTC, the difference narrowed considerably. The EFHW80’s paired median advantage was 0.5 dB, and message totals differed by less than 1%.

A convincing evening showing from the vertical, followed by a modest morning advantage for the longer wire. Calling the later morning a crushing EFHW80 victory would be pushing it.

They’re both EFHWs, but they’re very different antennas

On 20 m, the vertical is approximately half a wavelength long. The EFHW80 contains roughly two wavelengths of wire.

That difference changes the current distribution and reception pattern. A longer wire operating on a higher harmonic develops multiple lobes and nulls: directions and arrival angles where it receives particularly well, and others where it receives less well. Published antenna modelling illustrates how substantially an EFHW’s pattern changes between bands.

My EFHW80 also zigzags through the garden. Its bends, vertical and horizontal sections, height and surroundings all affect the result. A textbook drawing of a straight wire is useful background, but it isn’t a measured map of my garden antenna.

The vertical has a different combination of directional response, elevation response and polarisation. It is therefore entirely reasonable for the two installations to favour different paths.

More wire does not automatically mean more useful signal. Otherwise, we could settle every antenna argument with a tape measure.

The sky changes the contest

The ionosphere changes between daylight and darkness. Solar radiation produces ionisation; after dark, recombination reduces it. Those changes affect which HF paths remain usable and how signals reach us. Conditions along the entire route matter, including daylight at the distant end.

Consequently, the morning is not simply the evening’s collection of stations returning at a different volume.

Different areas become audible. Available paths can change their arrival angles and propagation modes. Even a station heard in both periods may reach me differently.

Our leading explanation is that the changing mix of paths better suited the vertical in the evening and gave the EFHW80 more favourable opportunities in the morning.

There was some supporting evidence from individual stations, too. All four stations with paired measurements in both the evening and early recovery shifted towards the EFHW80. Four is a small sample, and shifting towards it did not always mean it became the stronger antenna, but it was an interesting clue.

Don’t forget the noise

WSJT-X reports signal-to-noise ratio, not antenna gain.

An antenna can produce a better report by collecting more signal, less noise, or both. Local interference can change as appliances, lighting and power supplies switch on and off. The two antennas may couple differently to those sources.

Polarisation changes and fading may also contribute. We cannot separate every mechanism from decode logs alone, nor can those logs verify that every receiver and decoder setting was identical.

These results compare the complete receiving installations. They identify the observed difference more confidently than they identify its cause.

What I’m taking away

The experiment supported my original observation for this collection. It did not establish a universal rule that verticals win evenings and long wires win mornings.

Nor did it give me grounds to condemn the EFHW80. I inspected it, it looked sound, and its morning performance demonstrated that it remained useful.

The practical conclusion is that both antennas deserve their place. The vertical was stronger overall, but the EFHW80 heard stations it missed and performed better during part of the morning.

The changing ionosphere, interacting with different antenna patterns, is the most plausible explanation for the crossover. Noise and receiving-chain differences remain possible contributors.

I started with a feeling. I finished with a measured difference and a much better idea of what might explain it.

And, conveniently, another reason to keep both bits of wire in the garden.

73, Kev — M7XOM