Welcome to G7CNF's Website

The G7CNF "Monster Antenna Project"

Having used my homebrew EF0606 and EF0407 (6el 6m, 6.9m boom & 7el 4m, 6.5m boom) for nearly a year before mother nature's flatulence had an ill effect on the VHF array, I was forced into using the 6+4m Duo which is covered elsewhere on this site. True to form, as a 5+5el it behaves essentially as two monobanders with an imperceptible difference in performance compared to two discreet yagis of similar size. Having become accustomed to the longer booms, I did miss the 'ears' that these bigger brothers had. In the end it was all too much for me to bear and the Duo was consigned to storage for /p as was originally intended.

This page charts the construction of two new EF designs; the EF0607 and EF0409. These are a 7 ele 6m and a 9 ele 4m both on 9m booms. Although not particularly large for some amateurs, they are by far the largest antennas I have made in my ham carreer. To the right is the polar diagram of the 4m yagi which you can see has an impressive gain. Pop (YU7EF) is very satisfied with this model, as the Q is low giving it a wide bandwidth and uniform gain across the entire 70.0-70.5MHz band.

Like the other builds I have undertaken, I have reused the old polycarb element mounting brackets which shipped with the Moonraker yagis which were purchased as a quick "get on the air" option when I first moved to this address. As all of Pop's designs utilise insulated elements and the Moonraker brackets leave a lot to be desired in this regard, I put two layers of heat-shrink-sleeving over the bolt which passes through the element and connects directly with the underside of the boom. On a positive note, these brackets place the elements exactly at the position of 0% boom correction so all elements are per-table.

For this project I used 10SWG aluminium for the booms, which translates to 3.25mm wall thickness. This may not sound much but when you compare similar lengths of this 10SWG box against the Moonraker boom, there is a huge difference in the weight. For the elements, 16SWG wall is more than adequate.

I chose this wall thickness because in my previous build, the 0606 in particular at 6.9m was really at the limit of boom length for the thinner wall, with some notable twisting of the elements axially along the boom, during windy weather. And BOY does it get windy here!

My first job was to cut the elements to the correct length. My supplier provided for lengths in 1m increments, so I opted for 4m lengths for the 4m elements so I could have two elements extraced from each section with little waste. For the 6m elements I used 3m lengths, which also provided for the least wastage. I also ordered one 5m section which would accommodate the reflectors of both antennas. Four 5m box sections were purchased for the booms.

The first problem I had was how to join the two sections of the boom together, 5m and 4m respectively. In the Moonrakers and some other designs I have seen, there is a cylinder which slides inside both sections of boom to be mated making an 'internal reinforcement bar'. Clearly such a solution was not an option for this build as the wall thickness was so much greater the internal reinforcement would have to be impossibly strong to deal with the forces placed upon it, also without a lathe I could not achieve the external diameter required for sleeving.

Even plating the junction of the booms presented a significant weakness so another option was needed.

The solution needed to be strong enough to cope with handling the antennas prior to and during mounting when the support wires were not yet attached. Nine meters is a long boom when it is not made from a scaffold-pole!

With a wall thickness of 3.25mm there seemed only one sensible option to me, which was to sleeve one section of boom over the other.

The shorter section of box was marked at 15cm from the end and then carefully cut, diagonally along the corners to the mark. The ends were then splayed being careful not to bend too far to begin the work-hardening process. With a suitable bridge, the splayed sides were then turned back in, each set to boom centre so as to make a rigid spring.

Once completed the two sections of the boom were put together on a plane surface and carefully manipulated into a straight alignment. With chocks in place, four 5mm holes were then drilled and through-bolted with M5 zinc plated bolts. Note the alignment marks on each section to make sure that when it was time to assemble, the sections would be returned to their aligned state. Also, I then had a check on which sides to drill the element mount positions ;o)

This was then disassembled for the marking and drilling of the element locations.

XYL, Please note: This is the proper use of the patio table.... ;o) ha ha ha

Once the boom drilling was complete the next step was the dipole. Most of the EF designs use a dipole gap of 10mm. Of course the feedpoint must be waterproof, or at least the transition to the coaxial cable. I have seen and used many variations of the Yagi feedpoint arangements but I am one that likes to use materials to hand and I had some Moonraker brackets left over so decided to use these.

The dipole inner-ends were marked at 15mm, axially cut into 4 equal 'tabs'. Two of these were then splayed outwards.

The other two of the tabs were then bent inwards to the centre and mole-grips used to push the ends together and make a flat surface, two walls thick.

A 2mm hole was drilled through to accept a self-tapping screw.

An insulated crimp had its plastic cover removed and a short length of "G5RV left-overs" was soldered into it.

The crimp was then bent back on itself to 90 degrees to make it ready for mounting on the monopole.

Finally the tag was fixed into place with a galvanised self-tapper. Not shown here but the end was then carefully sealed with hot-melt glue.

Next the spare Moonraker brackets were summoned. The dipole centre is subject to some fairly high forces under windy conditions and every effort needs to be made to ensure that there is no movement on the electrical connection to ensure longevity and reliability. To this end I use the lower part of two brackets mounted back-to-back, bolted tightly to the boom to achieve this.

The monopoles are laid out on the floor with the lower mounting bracket. The 10mm dipole gap is set and then the upper bracket is placed on top. Next I use a G-clamp to press the two halves together and secure the dipole in place, temporarily.

Once the dipole is assembled, one can immediately feel the rigidity which is reassuring. To prevent water ingress and also to enhance the mechanical strength, I seal and fill the assembly with hot melt glue. Once set, I can hold the dipole at one outer end and see that there is no unwanted play.

At this point all that is left is to drill the centre holes in the elements. I also reviewed the mounting bolts for signs of chaffing on the existing heat shrink sleeving and repaired/replaced as necessary. This is a particularly important point, as any electrical contact between the element and the boom via the mounting bolts will have an adverse effect on the antenna compared to specification.

The next part was also equally critical. As I deliberately used small-section boom material in an effort to moderate the weight of the array and limit the visual impact for the neighbours. Without suitable support, not only is there significant force of moment placed on the boom mounting position (centre of gravity) but also due to the wind surface area, sideways play is also a factor.

Fully assembled on the ground, each antenna is around 10Kg. A test of the boom splicing by picking up the unsupported antenna at its centre of gravity shows the splicing to be sound. The ends drop over 0.75m from the centre position so the support is arranged. I use 3mm paracord-style nylon rope which has a breaking strain of 150Kg and is fairly neutral to the eye.

The supports provide stability to the ends of the antenna preventing vertical movement and a little further in, support midway both vertically and sideways. Even greater stability could be achieved by attaching another pair to the ends and gently tensioning them downward to the stub-mast.

I will endeavour to make a better picture!

I place the coax / balun connection in a short length of plastic drain-pipe and stuff this with bubble-wrap at each end to keep the connection dry.

The electrical connection to the antenna is made to the dipole fly-leads. As the antenna is native 50Ω it is possible to connect the coax directly to the dipole and place an RF choke at the feed-point or nearby, such as half-a-dozen turns of coax made into a 30cm loop. However doing this you are feeding an inherently balanced antenna with unbalanced feeder and this will introduce a skew on the pattern.

For this reason I use a 1:1 coaxial balun, which is made from normal coax. If space is an issue then RG58 will do but if there is any likelihood of QRO then I recommend mini8 or 10mm low loss.

Basically take your centre frequency, calculate the wavelength and multiply by the velocity factor of the coax; common types are well documented on the internet. Then work out the length of �λ and crop this off. Remember that the dimensions are to the screen end and not the centre conductor, so allow a couple of extra cm for this.

Fold the longer section of coax into 3 to save space but more importantly to allow the ends to naturally 'mate'. Connect the screens together at each end, connect the centre conductors together at one end only. At this end connect your feeder. At the opposite end, connect the two discreet centre conductors to each internal end of the dipole.

Apart from the obvious waterproofing, that is it, finished; one 1:1 balun. It is worth noting that if you leave the centre conductors too long at the dipole end, or the dipole fly-leads, if they are not perfectly balanced then the leads may add to the dipole length in which case it may be necessary to do some minor trimming. I have made it a point to make my dipole 'tuneable' by having a short length of smaller diameter tube in the end of each monopole for such a reason. Such a contingency can also aid if there is any coupling to nearby metalwork.

[I have had some emails querying the cause underlying the above statement regarding baluns and chokes]:-

Without entering into a treatise on the subject of balanced-unbalanced transformers (baluns) I will supply a simple explanation. A balanced antenna system can be defined as "any antenna where the driven elements are maintained at a potential independent of earth". The dipole on a yagi is no different than the dipole in, say an "inverted V" on HF. When the energy travels up the coaxial feeder and reaches the end, if it does not terminate in an unbalanced antenna, currents will flow down the outside of the braid of the coax which will radiate like an antenna. This radiation has a two-fold effect. 1) is that the additional radiation will "add" to the polar plot of the antenna, modifying it (hence the "skew" above) and secondly it will modify the standing wave arrangement on the entire system (two currents flowing in opposite directions in the same braid, with differing phase). In short the SWR will change. A simple test here would be to make a simple dipole for say 50MHz and connect the coax to it. Trim the dipole for best SWR then place a clip-on ferrite at the feed point (or as close as practicable) and observe the SWR, and frequency of the best impedance match you may be surprised that the result!

That concludes my construction experience with these antennas. Once mounted the antennas were measured with a miniVNa to get a picture of the matching and the 6m Yagi was almost perfectly within spec'. The 4m antenna however did suffer from some decoupling to it's 6m big brother and to the aluminium support cross-bar for the guys, mounted at the top of the mast. Some minor tweaking and the SWR was almost within spec' but the Q was slightly higher leading to a corresponding reduction in bandwidth, but having said this, the highest SWR was at the top end of the band and was just under 1.4:1 so I am not complaining. I modelled the array to check for interaction and to ensure that the 4m antenna's pattern wasn't being drastically compromised; but of course mounting two large antennas like these with only 1.8m spacing was always going to have some detrimental effect. It turns out that the pattern is not changed substantially and I have lost about 0.5dBi gain on 4m, mostly due to the proximity of the cross-bar to 4mD4.

Early tests show the antennas to have 'good ears' which to me is probably more important than the forward gain. On 4m I now have GB3ANG/b in IO86 permanently 'in' at 598km. I am very pleased with both antennas and recommend them highly to anyone wanting to build a performance station on either band.

I have yet to receive feedback from the neighbours................73

(\__/)
(='.'=)
(")_(")