Peter Cossins VK3BFG

Way back in 1974, DL7KM published an article in a German magazine called FUNK-TECHNIK (I think ?) for a two metre ‘Dopplequad’ or Dual Quad.

Since each leg a quad antenna is one quarter wavelength virtually in free space, the design is easily and successfully scaleable to other frequencies.

I decided to build a prototype antenna at 444.25 MHz to see how it would perform for possible further use by the Melbourne ATV Repeater VK3RTV.

Mike, VK3YOV had already built a bay of four for the 50 cm sound transmitter using a scaled down version of the original design which was successfully installed some years ago. This was a masterpiece of mechanical engineering constructed with the aid of commercial manufacturing facilities.

My objective was to experiment with materials that could be obtained from any local hardware store, construct some prototypes and make some reasonably accurate comparative measurements .

The 2 meter design by DL7KM called for quad sides of 520 mm. Since V = , where V is the velocity of light
( metres/sec in free space), f the frequency and l wavelength, the design frequency corresponds to about 144.2 MHz.

DL7KM’s design uses two quad sections joined at the centre resulting in an impedance of approximately 60 ohms, good enough for a match to a 50/52 ohm or 75 ohm feedline. (Theoretically 1.2:1 or 1.25:1 VSWR)

A directional antenna can be made by adding three reflectors spaced behind the dual quad section. The 2 metre version designed around 144.2 MHz with one dual quad and reflectors had a stated gain of about 10.5 dB. Phasing a pair of units or four units can obviously enhance this.

Another option is to connect four quads in a box formation which will provide a stated 8dB all round looker. This arrangement must have its quads connected in parallel rather than phased.

Depending on how any of these antennas are mounted the polarisation can be either vertical or horizontal. The diagrams provided show the antennas with horizontal polarisation.

I used 2.38mm brass rod for the prototype, but I would strongly recommend 3mm brass alloy welding rod as a better option for mechanical strength. The basic quad sections were bent around a jig made from a flat piece of particleboard with nails hammered in at each corner. There is a bit of trial and error involved to get the dimension of the quad sections correct. An easy option is to bend two sections to make the quad, joining them at the centre with small gusset plates.Use silver solder for these joints.

If the wire is long enough only one join will be required which can be achieved by using a small brass sleeve made from tubing available at model shops. This option is a bit more difficult to manage and some form of mounting mechanism will still be required at the centre of the dual quads. Brass washers can be pressed into service lashed to the quads with thin wire strands and silver soldered. There are other bending options using sleeving to join sections and you may like to experiment to achieve the best result for yourself.

When the dual section is completed, carefully check the dimensions and symmetry of the assembly.

DL7KM used a wood frame for his 2 metre version but for 70 cm I decided to use standard UV resistant PVC water pipe together with the range of fittings available.

The hub was constructed from an adapter. This was partially filled with potting epoxy and drilled to accommodate the feedline, matching stub and mounting screws for the quad element.

An end cap provided a final water seal for the hub. The remaining assembly is self explanatory from the photographs. The T Section is drilled and reamed so that the pipe will fit through. With the exception of the endcap, all the fittings are glued together with the standard PVC glue. Holes for the reflector elements are drilled for a force fit and small pieces of brass tubing are soldered each side to keep them in place. Don’t apply too much heat or you will melt the plastic tube. The big advantage of the technique is that the feedline, matching stub and driven element connection is very weather proof.

To avoid crushing the pipe with the mounting U Bolts, a small piece of the pipe can be glued outside or alternatively two half sections of aluminium tubing which is slightly undersize clamped by the U Bolts around the pipe will do the trick.

A ‘Pawsey Stub’ is soldered to the end of the RG58 feedline which in turn is fed down the centre of the boom. The feedline and matching stub is connected to the centre of the quad section and brass screws through the gusset plates fix the whole assembly to the hub. Keep all connections as short as possible and convert the cable to a larger low loss variety as soon as possible using suitable in-line N connectors. The co-axial cable can be connected directly to the centre of the quad sections without the Pawsey Stub, but I understand there may be a bit of skewing of the radiation pattern as a result of this.

The Pawsey Stub can be made of a quarter wavelength of small diameter teflon co-axial cable. This can be any impedance as the inner is not used. The advantage of using teflon co-axial cable is that it is very heat resistant. It is very important that the stub is insulated from the main feedline throughout its entire length and that the teflon outer does not come in contact with the RG58 outer. When soldering to the RG58 braid at the bottom of the stub, be careful not to use too much heat otherwise you may damage the integrity of the RG58. If you wish, the whole stub assembly can be sealed using heatshrink tubing.

The weather shield end cap is recessed slightly to completely cover the centre of the quad sections and is held in place by two PK screws.

It is also important to drill holes in the bottom of the pipes to allow water which may condense inside to escape.

The mechanical materials to build one dual quad are:

  1. PVC
  2. One T-Section
  3. Two end caps
  4. One adapter
  5. One end cap for adaptor

For VK3RTV I set up a bay of four as shown in the photographs. As for some arrays designed to spread signals in a desired broad pattern, each antenna is connected in parallel with an equal length of cable. The resultant impedance of 15 ohms is matched to the feedline with T Section matching network. The other alternative would be to construct a Q section .

With all multiple antennas it is very important to keep the sense of the connections consistent. This means that each antenna has cable attached in the same way ……. ie if one particular side of the Pawsey Stub is connected to the right side of the quad section, connect the stub in exactly the same way to the right side of every quad section.

General Dimensions ………….

is a quarter wavelength in free space at the design frequency.

A is the spacing between the reflector elements

B is the length of the reflector elements

C is the horizontal spacing between a pair of quads

D is the vertical spacing between a pair of quads in a four quad array.

E is spacing between a box array of four quads

F is the spacing of the driven element to the reflector

































The new frequency re-arrangements for VK3RTV has resulted in a downlink on 444.25 Mhz so I was able to make some comparative measurements with other antennas at the home location against a known constant signal. Given that the bay of four had offset antennas fed in parallel to produce a 1800 + pattern, they performed very well.