2x6 elements DK7ZB antenna

pa3guo dk7zb

Introduction
During 2007 I noticed that my 12 year old 2x10 elements Cushcraft antenna did not perform as it used to do. A side by side comparison of a 4 elements DL6WU vs the Cushcraft showed that something was wrong and the antenna did not function anymore as used to be the case.

After checking the cables and the antenna itself, I decided to give up on fixing the Cushcraft, and rather try to apply a more modern antenna design, eg DL6WU based. Since it had proven to be mechanically very strong, I wanted to try re-using the parts of the Cushcraft where possible

Antenna Requirements:
- Mechanically very solid (I was used to the Cushcraft, which was very well constructed)
- Suitable for satellite operations (circular polarization)
- When possible usable for EME: with only circular polarization on horizontal signals 3db is lost
- Elevation is needed: boom should not be to long (not 2 extra meters for only 2 db extra gain)

From various sources it is qouted that the DK7ZB antenna designs are very efficient. On the website of DK7ZB multiple designs are provided, I created the following summary:

DK7ZB antenna overview:
- 5 elements, 2.00 meter boom, 09.0 dbd
- 6 elements, 2.40 meter boom, 09.9 dbd
- 6 elements, 2.60 meter boom, 10.2 dbd (this is the one I selected)
- 7 elements, 3.00 meter boom, 10.7 dbd
- 7 elements, 3.30 meter boom, 11.0 dbd
- 9 elements, 5.00 meter boom, 12.4 dbd (PE1ITR has made this antenna, serves as reference)

DK7ZB 6 elements / 2.60 meter boom design specifications
On the website DK7ZB did not provide the dimensions for 10 mm diameter sized elements, which I wanted to use. Therefore I extrapolated them from the 4,6,8 mm dimensions that were given. They are provided in the table below (the orange row).

Element positions (mm):

Refl Rad Dir 1 Dir 2 Dir 3 Dir 4
0 320 675 1300 2020 2580

Element lengths, Radiator always 10mm diameter!

Diameter Refl Rad (10mm) Dir 1 Dir 2 Dir 3 Dir 4
10mm 1020mm 968mm 927mm 906mm 906mm 896mm
8mm 1020mm 968mm 927mm 906mm 906mm 896mm
6mm 1022mm 968mm 936mm 916mm 916mm 906mm
4mm 1024mm 968mm 950mm 926mm 926mm 916mm


Construction
Re-using the 2x10 elements Cushcraft antenna:
Replaced the element by a new 10mm aluminium tube. A piece of flexible plastic (yellow) waterhose provides the isolation from the boom. Quote from DK7ZB: the screw in the electrical "zero-point" has no influence to the length of the element and is connected to the boom. On the left the original Cushcraft material, on the right the concept of the new antenna.

pa3guo dk7zb pa3guo dk7zb


Due to the size of the cases for the radiators the space between both antennas became 6 cm (antennas are 6cm spaced from eachother, this is to be taken into account when both get coupled later into a single combined antenna).
pa3guo dk7zb pa3guo dk7zb

The radiator elements are built from to pieces of aluminium tube, which I re-connected by a wooden pin & glue. Inside the case a small pieve of PCB clamps the radiator to the case itself (see the picture on the right). Inbetween the PCB and the aluminium a small piece of rubber is glued, for more stability. Tip to tip length of the radiator is 968mm, including an 8 mm spacing between the 2 parts (inside the case, where the wooden pin is connecting both parts)
pa3guo dk7zb pa3guo dk7zb

The about 6x6cm case was after all quite small. On the picture you see how it all (just) fits.
pa3guo dk7zb pa3guo dk7zb

All openening in the case are made waterproof by applying cheap glue afterwards. In each case a small hole is drilled in the side that will be at the bottom when mounted later, to serve as water outlet (just in case).
pa3guo dk7zb pa3guo dk7zb

The impedance converter are built from the 2 parallel pieces of 75 ohm (RG59) coax of 1/4 wavelength. At 144.200 MHz this come down to 0.343 meter length (measured from the end of the copper shielding, not from tip to tip). At 145.000 MHz this would be 0.341 meter, I decided to make them 0.340 meter as main target is the 145.85 satellite band. They are mounted 2 cm above the boom, to avoid shorts/sparks when a high amount of RF power is applied.
pa3guo dk7zb pa3guo dk7zb


Measurements @ PA3GUO:
After assembling the first antenna on the boom that antenna was ready for a quick first test, in a horizontal set-up.
pa3guo dk7zb

First Results & Conclusions:
- GB3VHF received relatively strong, SNR of -11..-15db, used to be -17db
- ANDE received with S8, used to be only S3, big improvement !
- ANDE received with deep fading (S3..S8), same for PCSAT: I need circular polarization
- SWR: below 1:1.5, from 142-146 MHz, good enough for first version without any tuning

Encouraged by these results I finished also the second set of elements. However, as the SWR indicated the antenna was tuned to 144.5 MHz (as designed by DK7ZB), I decided to tune the second antenna towards 145.850 MHz by making all elements & the radiator 2mm shorter.

Measurements @ PE1ITR:
Measuring the characteristics of both antenna's (which are mounted on a single boom) and comparing them to the 9 elements DK7ZB of Rob, PE1ITR.

pa3guo dk7zb

We tested the first antenna in a vertical set-up, with a metal boom. The SWR was not good at all. Conclusion: a metal boom will destroy the pattern of an '+'configuration. And even though I have a non-conductive boom, the preference will be for an 'X' configuration. After turning the antenna in horizontal position the SWR became good, with a optimum at 144.150 MHz, as expected from the calculations of DK7ZB.

The antenna coax was fed towards the front of the beam, and not towards the back of the beam. We saw no influence of the cables on the performance, and such a set-up is mechanically much better. The impedance transformers look backwards to the reflectors. See below for the characteristics of the antenna.
pa3guo dk7zb

Also the second antenna was measured: SWR also good, with the optimum at 145.99 MHz, as expected (all elements made 2mm shorter).
pa3guo dk7zb

We also compared the 6 elements beam with the 9 elements beam of PE1ITR. Using a beacon the difference in S/N ratio was around 3db, which is perfectly in line with the specified gains of both antennas (10.2 versus 12.4 dbd).

Coupling both antennas
As we wanted from a construction point of view an 'X' configuration, the antennas needed to become coupled to create horizontal, vertical or circular polarization.
Rob, PE1ITR, made a concept implementation of a power divider in order to create (switchable) horizontal and vertical polarization (and later also left and right hand circular polarization.

Results are depicted below, and are very good.
pa3guo dk7zb
pa3guo dk7zb

Conclusions
- Gain of the antenna is as specified by DK7ZB (~10.2 dbd)
- Antennas are tuned to their desired frequencies (144.2 and 145.85 MHz)
- X configuration is preferred and working fine, cables can be fed towards the boom
- Coupling both antennas can provide both horizontal, vertical, left/right hand circular polarization

Finally
Based on the very good results from the session with PE1ITR, I have finished the antenna construction for an 'X' configuration. During that work I mixed up the elements of both antennas, so their original 144.3 and 145.9 optimums changed a bit (came closer, which isn't to bad anyway).

When they were mounted on the mast in their final set-up, signals were good. Cables from both antennas are fed to the polarization switch. The length of the cable of the front antenna is made 5cm longer (6cm x velocity factor 0.8 for AIRCELL7) to compensate to the difference in mounting.
Using the polarization switch box (vertical, horizontal, RHCP, LHCP) we noticed all worked fine when monitoring different sources (satellites, beacons, repeaters). Circular polarization gave compared to linear much more stable signals.
Listening to the LEO satellite ANDE illustrated this as well:
Horizont: S1..S5
Vertical: S0..S4
Circular: stable S5

We did a final set of measurements of the antennas, see below for the graphs:

1. Front antenna
pa3guo dk7zb

2. Back antenna (sitting on the boom 6cm behind the front antenna)
pa3guo dk7zb

3. Both antennas coupled and switched into horizontal polarization
pa3guo dk7zb

4. Both antennas coupled and switched into vertical polarization
pa3guo dk7zb

5. Both antennas coupled and switched into RHCP (Right Hand Circular Polarization)
pa3guo dk7zb

6. Both antennas coupled and switched into LHCP (Left Hand Circular Polarization)
pa3guo dk7zb

Results
- Gain of the 2x6 elements antenna is as specified by DK7ZB (~10.2 dbd)
- Antennas are tuned to their desired frequencies
- X configuration is preferred and giving excellent results
- Coupling both antennas can provide both horizontal, vertical, left/right hand circular polarization

To compare the 4 polarization modes I tuned to a horizontally polarized beacon at 144.418 MHz and captured the signal strenghts: pa3guo dk7zb

Finally, a picture of the new antenna, mounted at its final destination:
pa3guo dk7zb pa3guo dk7zb