The G7FEK  A compact multi-band HF wire antenna.

G7FEK multi band wire HF antenna

The antenna shown above is my own homebrew version of the G7FEK antenna design. This was built entirely from scrap materials so was a very low cost antenna project. Although it looks quite large on the photo it is actually very compact for an HF antenna that works well on the 80m amateur band. It is about 47ft  (14.4m) end to end across the top horizontal radiating elements with 25ft  (7.6m) of vertical twin conductor made with scrap copper wire. The vertical twin-conductor section is fed from the bottom with both conductors joined at the bottom feed point. This means the vertical conductors are used as linear loaded vertical radiators and the whole antenna is effectively a pair of Marconi inverted L antennas  nested back to back.

 The antenna itself is a 'T' shape, the poles and guy ropes at the left and right of the photo are only there to support and tension the G7FEK so that its feed point at the bottom of the vertical twin conductor section is about one foot (30cm) above the ground.  The original design of the G7FEK includes balancing counterpoise wires below the antenna connected to the outer braid of coaxial feed line at the antenna feed point. I do not use counterpoise wires on my G7FEK as I have found it works better for me at this particular location using a long 3/4 inch  (about 22mm diameter) copper ground rod driven about 7ft (2m) into the ground with a chunky earth wire connecting it to the outer braid of the coaxial feeder at the feed point.


As with all of my antennas I use common mode chokes on the coaxial feeder to stop common mode currents flowing back down the outer shield of the feeder and radiating, You only want the antenna to radiate, not the feed line. I also have common mode chokes on coaxial feed lines before they enter the shack, this not only helps stop the feed lines radiating which has negative effects such as signal loss and interference but it also keeps common mode currents from flowing back into the shack as that can cause a range of problems for computers and sensitive test equipment.

In direct comparison 'A' - 'B' swiching tests against other multi-band HF antennas I have found the G7FEK to be a good low noise receiving antenna with very good transmission characteristics. There have been a few other antenna designs that were better on one or two bands but they were all much larger, between 95 and 270 ft long, and they were all deployed at greater heights than the G7FEK.  

For anyone with limited space to put up an 80m antenna I can recommend giving the G7FEK a try.

 if You have less than 47 ft to string up the top horizontal wires You can hang it up so that as much as possible is in a straight line but part of the long leg of the horizontal radiator turns 90 degrees to a third support. This will change the shape of the antennas lobe pattern and may slightly reduce efficiency but it will still allow You to have a lot of fun on most of the amateur HF bands from 80m to 10m. You may also get it to work on 6m band using an ATU or other impedance matching device but it is not particularly efficient on 6m band. 

I have used this as my primary HF station multi-band HF antenna for just over three years of low power 'QRP' operation and have been very happy with its performance with over 18,000 confirmed QSOs across 150 countries in the log using my primary MM7WAB callsign at the time of writing this article.

I also used it running 5W QRP to work over 2,000 stations in 82 countries using the callsign MQ7WAB during the month of June 2022 as part of the Queen Elizabeth II Jubilee celebrations. 

The fact that I live in a rural location 700ft up a hill and am lucky to have excellent ground conductivity due to high ironstone content in this area are certainly contributing factors in my success with my HF station.

 Two element Moxon for 2m VHF band.

Two element Moxon beam for 2m VHF band in use on portable station.

A VHF Moxon is another very cheap and easy to build project. This one was constructed from repurposed scrap materials, a wooden curtain pole, an old wooden coat hanger, some old 2.5mm Red & Black Twin-and-Earth cable with external cable sheath stripped off to provide two lengths of insulated copper wire for the driven element and reflector. I also happened to have a short length of coaxial cable left over from another project that already had a BNC connector fitted so that was checked with a meter and then soldered to the two halves of the driven element.  Total antenna cost was Zero and took less than an hour to build. 

 If You had to buy the bits the connector would probably be the most expensive part unless You used a long coaxial feeder cable to use the antenna at a fixed location or on top of a tall pole. 

The red reflector element is the back of the beam so the primary forward lobe is in the direction of the black element. Unwanted noise or relatively strong QRM can be can be reduced and sometimes nulled out completely off the back of the beam.  The forward lobe is quite wide so You don't have to be too critical with azimuth beam headings, simply sweeping the antenna back and forth it is easy to find the peak strength of a single station. This also makes it quite good for calling CQ with eight 45 degree steps easily covering a full circle from a summit or other elevated location.

I use this antenna for Portable handheld working when walking up in the hills and have carried it tied to the frame of my bike or lashed to my backpack when travelling light with just a hand held radio. It is quickly  detached from the bike or backpack and deployed as a hand held beam when I find a nice location to stop.

The coaxial feeder tail has a BNC type connector making it quick and easy to hook up various 2m radios  fitted with suitable adapters as required. I often use this Moxon beam with a hand held 2m radio lashed onto the wooden antenna handle by the radios belt clip. It is then operated with an external speaker mic plugged into the radio making it easy to handle, one hand holding the antenna and one hand for the mic.  When using hand held radios on this Moxon  I usually run 1W low power and have made many QSO's at ranges beyond 50 miles from locations up in the hills. Running low power has the added bonus of making the radios batteries last longer.

 I have also used this antenna to very good effect by sliding the antenna pole inside a length of PVC electrical conduit lashed to the bike trailer. (see photo above) I then use a short length of coaxial feeder plugged into the Kenwood TM-701E running off a small SLA battery on the bikes rear carrying rack or a car battery housed in the bike trailer for a day or more of operation. With the TM-701e on low power, 5w, I have had very good results with several contacts well over 100 miles.

I built this Moxon based on calculated dimensions from the Moxgen software and then made slight adjustments to the gap spacing between the elements to dial in the antennas center frequency to 145.450 as I primarily use it for working in FM mode. 

 Other homebrew antenna projects and handy tips.

I have also built and experimented with a wide variety of other antennas at various locations:-

 Half Squares, (HF 40m, 20m & 17m)

 Delta loops. (horizontal & vertical) (HF,VHF)

 Bifilar 'flag' (HF 160m, 80m, 40m & 30m,)

 Slopers (HF various bands) Often use center fed dipole as sloper when fixing points limited /A or /P 

 Marconi inverted L`s (various HF bands)

 LOG 'Loop On Ground' mostly for RX but also used for TX on 160m, 80m & 40m for NVIS operation.

 Base loaded, center loaded and distributed load verticals  (HF,VHF,UHF)

 Phased Arrays. Vertical and Horizontal. (HF,VHF, UHF and Microwave bands)

 Trapped dipoles (various selected HF bands)

 J-Poles (dual band VHF,UHF)

 Double-Bazooka coaxial dipoles (HF 80m,40m,20m & 17m)

 Vertical co-linear arrays (mono banders VHF,UHF & Microwave)

 Franklin co-linear arrays (VHF,UHF)

 Horizontal quad loops (HF,VHF and UHF outdoors and in attic space) 

 Fixed and portable NVIS wire antennas (mono banders for HF 160m,80m,40m & 30m)

 Horizontal V's (mono banders for HF, VHF, UHF & Microwave) 

 Horizontal OCF 'Z' (HF 160m,80m) feedpoint at 42% from end of wire worked best at my location.

 End fed (HF) : Note. OK for RX but inefficient on TX. Only used if nothing else possible. 

 534ft 'meandering wire' forming a pair of horizontal loops  (160m,80m)

 256ft 'twisted pair' wire OCF (160m - 6m) with two 130ft+ counterpoise wires.

 132ft 9in enameled copper wire OCF dipole (80m CF 3.760MHz with feedpoint 42% from end)

 2 ele Moxon rectangles (mono banders for 17m,6m,2m,70cm)

 2 ele Yagi-Uda wire array (80m) this was strung up in trees at edge of a forest for a summer season.

 3 ele Yagi-Uda wire arrays (mono banders for 17m,11m,2m)

 5 ele Yagi-Uda array (2m) 

 8 ele Yagi-Uda array (70cm)

 2 ele Quads (mono banders for 17m,10m and 2m)

 3 ele Quad (2m)

 Lazy H (20m)

 Hexbeam (17m, 15m, 12m & 10m)  

 Folded dipoles (monobanders from 80m - 70cm)  often used as driven element on Yagi-Uda arrays.

 Magnetic Loops  Selection of experimental STL designs. (20m,17m,15m,10m)

 Turnstile (2m, 70cm) 'standard' using phased dipoles and 'high gain' using phased Moxon beams.

 Curtain Array (VHF) 3 bays high X 4 bays wide.

 Various patch, slot, PIFA and multi element arrays for mobile, mesh & fixed data links (700MHz - 5GHz)

 Assorted arrays, quads, bi-quads, parabolics & waveguides for WiFi mesh & IP data links (2.4 & 5GHz)

 Assorted small RDF arrays.


 Most of the HF, VHF and UHF antennas listed above were relatively easy to build using basic hand tools and a tape measure or one metre rule.  It is always best to cut wire lengths for antennas a bit longer than needed and fold ends back to tune the length to resonance. If the folded back part is very long I simply trim it back so there is only a few inches of fold back, this is easily secured with self amalgamating tape or electrical tape and cable ties to keep the ends neat.


The patch, slot, PIFA and other Microwave antenna arrays were made using precision measuring equipment, specialist jigs, fixtures and very fine tools. It is much harder than it appears and is a great technical challenge to achieve resonance and reasonably decent efficiency especially when developing and making multi-band arrays. Making homebrew antennas for HF is much easier as the tolerances are quite large and +/- an inch or so (about 2 or 3 CM) is usually close enough compared to the tight tolerances required for Microwave developments where a +/- 1 thou  (one thousandth of an inch or 0.025 mm) can make an enormous difference in performance.

For many years all the HF,VHF and UHF homebrew antennas I made for myself and others were adjusted and tuned to length the old fashioned way using a homebrew Grid Dip Oscillator (GDO)  lots of written notes, hand drawn graphs, plotting results on Smith charts and a fairly large chunk of time and patience.  The tuning process is much quicker and easier these days using a hand held portable Nano VNA to see near real time plots from hundreds of measurements on the VNA screen. This massively reduces the time consuming manual process of taking test readings, plotting results and drawing out graphs on paper as well as removing the tedious pfaffing about with making plots on paper Smith charts that were often poor quality photo-copies. Rapid antenna development and tuning is much more fun now with the tests and measurements performed in fractions of a second on screen. 

 Double-Bazooka coaxial dipole 


The double bazooka antenna design above is optimized for NVIS (Near Vertical Incidence Skywave) operation. This design will work most efficiently for NVIS communications when strung up at the NVIS HEIGHT above good ground, or at NVIS HEIGHT above elevated reflector wires used to create an artificial ground plane. 


If this particular antenna design is strung up lower or higher above ground (or the artificial ground plane reflector wires) it will still work for NVIS operation but will operate at reduced NVIS efficiency. 

If You want to use this antenna for long range DX operation it can be strung up much higher above ground where it will behave more like a half-wave center fed dipole but with a little extra gain over a standard dipole. I have worked stations in Indonesia, Taiwan, China, Japan and Vietnam using the 40m band version of this antenna strung up about 35 ft above ground. 


The optimum height above real ground for NVIS operation will depend on the ground conductivity at each  particular location so You may have to slightly adjust the height to maximize efficiency where ground conductivity is poor.   At sandy locations You will achieve better results by using the ground reflector wires.

I have worked stations around the UK with both the 80m and 40m versions of this antenna strung at various heights from a few inches above ground to around 10ft above ground with reasonable results.
I performed a repeated series of real world on air tests in LSB voice mode with known stations at fixed locations  over a 6 month period during 2020. I always operated in LSB voice mode driving 10W into the antenna and the best results were always recorded with the antenna strung at or very close to the NVIS HEIGHT specified above.

I used the 80m version of this antenna thrown on top of the hedge down one side of my back garden as an emergency backup antenna after my main station antennas were destroyed in winter storms.
It was certainly not that efficient but it did allow me to have QSOs with stations around the UK and across the North sea with a maximum working range of about 450 to 500 miles from my QTH in Ayrshire, S.W.Scotland. 


Note: I wound an air spaced  common mode choke in the coaxial feedline to this antenna about a foot (30cm) from the feedpoint to stop common mode current travelling back along the outer braid of the feedline and ensure that only the antenna radiates, not the feedline.