Amateur Radio (G3TXQ)- HexBeam Matching
We saw in the Basic Performance section that our benchmark 20m HexBeam delivers useful performance ( > 10dB F/B ratio ) over a 200KHz range from 14.060 MHz to 14.260 MHz. The chart on the right shows how the SWR of the beam falls from 3.8:1 to 1.6:1 across this frequency range. At the lower frequency limit the beam's impedance is [13 + j3] Ohms, rising to [38 + j16] Ohms at the upper frequency limit. At the frequency of best F/B (14.140 MHz), the impedance is [25 + j14] - an SWR of 2.2:1
It may be that, if your rig has an automatic ATU or that you are prepared to operate through an external ATU, you can cope with this range of impedances without needing to introduce any matching at the antenna. If not, there are a couple (at least) of simple matching techniques that can be used with a monoband HexBeam
1. Series matching section ( 25 Ohm line )
On the Smith Chart to the right we have plotted at Point 1 the impedance presented by the HexBeam at 14.140 MHz [25 + j14]. The chart shows the effect of introducing a series section of transmission line having a Characteristic Impedance (Zo) of 25 Ohms and an electrical length of 0.1 Lambda. The impedance has been transformed to [43 + j0] at Point 2 - a reduction in SWR from 2.2:1 to 1.2:1
If we now introduce this series transmission line section into our model we can see its effect on SWR across the frequency range of interest. The SWR has been improved to below 2:1 in all but the bottom 20KHz of the range. The SWR at 14.060 MHz could probably be reduced to below 2:1, at the cost of a small increase at 14.260 MHz, by small adjustments to the length of the series matching section.
25 Ohm line is not something that is readily available commercially, but can be fabricated easily by "parallel connecting" two identical lengths of 50 Ohm line. Connect the inners of the coax together, and the braids together, at either end of the run. Make sure the two lengths of line have the same velocity factor!
To summarise the technique:
Add two parallelled lengths of 50 Ohm coax between the beam and your main 50 Ohm feedline; make the parallelled lines one tenth of a wavelength long, remembering to take account of the velocity factor. For example, at 14.140 MHz the length would be about 4.6ft.
2. Beta matching using a shortened Driver
In the section concerning Driver characteristics we learned that the Driver of the HexBeam could be shortened or lengthened considerably without compromising the performance of the antenna. Shortening the Driver caused an increasingly capacitive component at the feedpoint; we noted that for a Reflector / Driver ratio of 1.04 the impedance was about [25 - j25] Ohms which makes it a prime candidate for Beta matching. This impedance is plotted as Point 1 on the Smith Chart to the right.
Adding parallel inductance is equivalent to moving around a circle of "constant conductance" on the Smith Chart, so we are easily able to transform Point 1 into Point 2 - a new impedance of [48 + j0]Ohms and an SWR close to unity - by placing a coil across the feedpoint of the antenna. This inductor needs to have a reactance of 50 Ohms - about 0.57uH at 14MHz.
If we now model the shortened Driver with the parallel inductor we can see the effect on SWR across the frequency range of interest. The SWR has been improved to below 2:1 in all but the bottom 10KHz of the range. Small adjustments to the inductor would likely improve the SWR at 14.140 MHz to below 2:1 at the cost of some small increase to the SWR at 14.260 MHz.
To summarise the technique:
Make the Driver 4% shorter than the Reflector and add an inductor with a reactance of 50 Ohms across the antenna feedpoint (e.g. 0.57uH for 20m)
Baluns
Many HexBeam users wonder whether or not to include a balun in their feedline. I recommend the use of a 1:1 choke balun: when I added one to my beam I noticed that measured SWRs were more "repeatable" and less dependent on the route taken by the feedline. I use a ferrite choke balun, but a "coiled coax" choke would likely be as effective. If you build one don't overdo the number of turns: for 20m through 10m, 6 turns of RG213 on a 4" diameter former is plenty.