A Halo Antenna for 6m and 10m

Matt Roberts - matt-at-kk5jy-dot-net

Published: 2021-01-26

Updated: 2023-02-02

During the 2020 Sporadic E (E_S) season in the US, I used a 6m halo antenna to quickly get on 6m, and work hundreds of stations. The halo is a very effective antenna, with a small size, good efficiency, and a nearly omnidirectional horizontally-polarized pattern at low elevation angles. It can be easily built using very inexpensive materials.

For 10m E_S work that same summer, I used a simple dipole antenna made from CB whips. That antenna worked fine, but with a 16' wingspan, it was not nearly as space friendly as the halo, nor was it the best form factor for portable or mobile operation.

Since my interest in 6m and 10m is almost exclusively chasing E_S using digimodes, I started searching for a single antenna that I could easily raise during peak E_S months, but would also give me a good horizontal pattern for chasing VHF DX across both bands.

The 6m halo is closely related to the Cobweb antenna, popular for high-band HF work. The Cobweb is basically several halo antennas in parallel, usually fed with a wideband transformer connected to coaxial cable. After modeling a number of different antenna options, I realized that a two-band halo had all the features I wanted for an E_S antenna, both for home use and on the road:

Single feedline
Easy match to 50Ω cable
Low-angle (DX) pattern at reasonable mast height
Horizontally polarized to match E_S conditions
Omnidirectional (or nearly so)
Physically small and lightweight for portable use

To my knowledge, nobody is making such an antenna commercially, so I decided to build one. Construction costs were very similar to that of the 6m halo antenna, at roughly $25 US, all-in sans coax. When it comes to antennas, "build" is usually better than "buy," anyway. ;-)

The Model

The design is very similar to the 6m halo. The antenna is really two halo antennas, sharing a single feed point, feed line, matching network, and physical support structure.

The only real point of compromise with this antenna is the installation height. For a single elevation lobe at its lowest elevation, a 6m halo has a target installation height of about 12', while the 10m halo has a target installation height of just over 20'. Raising either one higher will lower the elevation of peak gain of the lowest lobe, but it also creates secondary lobes due to ground reflections, with nulls between the lobes.

I chose a modest installation height that was a good compromise between the best single-lobe heights for the two bands. This gives a single lobe on 10m, while still preserving a reasonably wide lower lobe on 6m.

The EZ-NEC+ model for the design is shown below, at its initial installation height of 15', with elevation plots along the main lobe axis, and SWR sweeps for both bands using a reasonable compromise match, as measured at the model antenna feed point:

Figure 1: EZ-NEC+ Model and Data
E_S Halo at 15'

As with beams and other horizontally-polarized antennas, there is certainly nothing wrong with raising the halo antenna to great heights. As the antenna is raised, the lowest lobe in the elevation plane will descend further towards the horizon, giving better DX performance near that angle. That said, modest installation heights can still give excellent results during band openings.

Each of the antenna elements is a dipole, bent into a near-loop shape. The resulting pattern is similar to that of a dipole, but with the side-facing nulls largely filled in. The antenna is therefore omnidirectional, with positive gain in all directions, and ~4dB of eccentricity from front to side—that is, the gain normal to the feed point is ~4dB higher than the gain to the sides.

When I compare the model of a 10m halo at 15' to that of a rigid dipole at 20', the halo has only ~1dB lower gain at its peak elevation normal to the feed point. However, the dipole is four times wider, physically. So a 75% savings on space comes at a cost of only 1dB, an overall omnidirectional pattern, and a lower physical installation height.

I'll gladly take that trade.

The Physical Antenna

The two-band halo is very simlar to the 6m-only version. I used a piece of enamel-painted lumber as the hub for four, 48" fiberglass electric fence poles, used as spreaders. The poles are bolted to the wooden hub using common stainless steel P-clamps. The P-clamps I used had soft insulating inserts, commonly used to prevent chafing of insulation when the clamps are used to hold cables, but plain metal clamps are probably fine for this use. Also, P-clamps and hose clamps hold the wires to the frame at the proper points, and Dacron cord pulls the wire ends into tension. Many Cobweb and Hexbeam antennas use similar construction techniques and materials.

The 10m wire perimeter is 52" per side, giving the antenna an overall wingspan of just over 4'4", or 1.3m. That's a great size for an antenna that can fit in the back of a truck or SUV without requiring disassembly. The components are relatively lightweight, also good for portability, and also providing a minimal load for any reasonably sized mast, allowing even modest masts to safely support the antenna at significant height, if desired.

The 6m wire perimeter is just under 30" per side.

The elements were joined at a plastic box that converts an N connector to 1/4" fastners for wire lugs. A fifth fiberglass pole was lashed across the spreader poles on the feed point side, and the junction box was attached to it. This greatly reduces the amount of movement of the box, as well as the mechanical load placed on the wire lugs.

The Electrical Antenna

Like the 6m halo, the feedpoint impedance is quite low—less than 10Ω on each band. Commercial Cobwebs cover anywhere from one to three octaves, making use of a wideband transformer almost necessary for a good match across all the supported bands. I dislike transmitting through transformers, because it is difficult to make an efficient wideband transformer for high power levels.

Fortunately, the 6m + 10m combination is less than one octave, so I chose to keep the hairpin matching arrangement used on the 6m-only version. While this requires selection of a hairpin value that is a compromise between them, 10m and 6m are close enough that it is easy to find a single inductance value that gives a very good match to 50Ω cable on both bands.

The model predicts a "best" compromise matching inductor value somewhere around ~0.09μH. The real antenna seemed to have the best match with 10" to 12" of wire in a single turn across the feed point. With the hairpin adjusted, and about 150' of low-loss feedline attached, the antenna provided the SWR curves shown below:

Figure 2: SWR Mesurements

For this prototype, I used two wires connected to the feed point, one with an alligator clip on the end, to allow me to easily adjust the hairpin length to obtain the best overall match. I found it best to adjust the hairpin first, before adjusting element lengths, when tuning up the antenna. Small hairpin adjustments can make rather large changes to the 50Ω center frequency for each band, so adjusting the hairpin first greatly speeds the process of tuning.

So how does it play?

When the antenna was first installed in late January 2021, we were still several weeks from the start of the main summertime E_S season in North America. Even so, every new antenna project deserves a spin around the block. Running 50W, the antenna immedidatley made 10m contacts between 70 and 9,200 miles. Similarly, the first 6m contacts ranged between 60 and 1,100 miles, with excellent signal reports exchanged in both directions.

Over the next few weeks, the antenna made hundreds contacts on 10m and 6m, at various distances. The signal quality and strength seen during E_S openings is absolutely incredible. If you haven't worked a few E_S openings, you are really missing out. The beautiful SNR and very strong received signals—even between stations using relatively low power and small antennas—is enough to make even the best-equipped HF operator drool.

Now that summertime summertime E_S openings are increasingly common, the antenna has acquitted itself quite well. Despite its relatively modest size and installation height, and at just 100W, the antenna rakes in contact after contact on both bands. After more than 1,000 10m contacts, the average received SNR reports on FT8 are within 1dB of reports sent. That tells me that the antenna is quite competitive with the those used by most other E_S chasers.

Housecleaning

Over the first few months of use, I made a few improvements to the antenna:

New Insulators: The rope spacers were replaced with ceramic spacers, held to the wire ends with zip ties. This allows the antenna to use high power without melting the spacer. The problem with polymers is that it isn't the best insulator at high voltages, especially in wet weather. The ceramic insulator should provide enough isolation between the wire ends to keep power from flowing between them.

New Choke: I added some beads to the top portion of the feedline, close to the feedpoint, to act as a coaxial choke. This didn't measurably affect the match or the SWR, but it is cheap insurance against I₃ current issues in the future.

The updated antenna is shown at top-right.

Updated Hairpin Clamp

Matching Network: Unfortunately, most consumer-grade alligator clips are made from plated steel. This means that they will rust over time when exposed to the elements, making the clip used in the hairpin a poor long-term solution.

In place of the clip, I used a short bolt with two fender washers, topped by a split washer and wing nut—all 1/4" stainless steel hardware. The fender washers form a small compression clamp, between which the long wire from the hairpin can be securely held. A 1/4" crimp lug attaches the bolt to the short wire and the feedpoint. This arrangement allows the hairpin to be easily adjusted when needed, but also holds the desired length securely when the desired value is reached.

The updated clamp is shown at right.

A similar arrangement was used at the end of each element wire, to allow their lengths to be easily adjusted. A P-clip attached to each bolt holds the zip-ties to the ceramic insulator.

Hub Repair: I often use wood for strutural purposes, especially in prototypes, because it's easy to work with, easy to paint, relatively lightweight, and nonconductive. However, even the best-painted wood tends to absorb moisture when used outdoors, and this moisture can cause it to warp and split. Such was the case with the antenna shown above. After a few weeks of snow and rain, the wooden hub started to deform, which deformed the entire antenna.

So the next clean-up step was to add two short pieces of angled steel, running along two sides of the wooden hub, across the wood grain, to pull the wood panel back into a flat shape. This worked well, and immediately restored the pristine shape of the antenna. If I chose to build this antenna from scratch, I would use a piece of 1' square aluminum, 1/8" thick, in place of the wooden hub. Such material is available at any welding shop relatively cheaply, and would last for decades, long after the fiberglass has turned to dust. I have used such material for antenna hubs and mounts before, and it's a nice material for long-term installations.

Rebuilt Antenna
with Aluminum Hub

If the antenna goes up again next year, I will probably replace the hub with aluminum.

Hub Replacement: Eventually, "next year" arrived, and the wooden hub was again in poor shape, so I used a piece of 12" square aluminum plate to rebuild the antenna hub. This makes the antenna significantly stronger, lighter, more durable, and should increase the longevity by many years.

The replacement was straightforward, and an updated picture of the construction is shown at right. I still used insulated P-clamps to hold the fiberglass rod in place, and it is all bolted together using 1/4" stainless hardware.

The new hub is drilled differently than the old one. The P-clamps were arranged so that they could hold opposite spreader rods next to each other, rather than end-to-end. This allowed me to slide the rods inward to shorten them without cutting the fiberglass. This shrank the diagonal wingspan by several inches. In order to fit the opposing rods, I used an extra 1/4" nut beneath two of them, which elevated the P-clamps and arms maybe half a centimeter, allowing one pair of spreader rods to pass over the other.

Electrically, the antenna is identical to the previous versions, but the shorter arms make it even more compact.

Copyright (C) 2021-2023 by Matt Roberts, KK5JY. All Rights Reserved.