A Portable 20M Vertical Phased Array

1)15 pound weight limit including vertical radiators, support
structures, phasing controller, and feedline cables.

2) Compactness to allow stowage of masts and ground stakes on the rear
deck of the kayak. No item could exceed five feet in length. (photo #8)

3) System support structure must be impervious to salt water.

Several practical criteria were also set:

1) Fast setup time for the array -- 15-20 minutes maximum

2) Use of self-supporting elements

3) No phasing adjustments required in the field

4) Convenient direction switching from a central location (tent)

The array consists of a number of components, each of which will be discussed, beginning with the antenna support structure:

Let's start at ground zero. (Photo #1) In order for the array to be self- supporting, a ground stake was made for each vertical. The ground stakes are 5 foot sections of 1-1/4" Schedule 40 PVC pipe. The I.D. of this pipe is actually about 1.4 inches, and the O.D. is about 1.7 inches. A point is hacksawed into one end of the mast. We'll call this the bottom end. A 3/4 inch hole is drilled through the mast at a point about 22 inches from the top end of the mast. A smaller piece of plastic waterpipe or rod with a diameter of about 0.7 inches is inserted through the 3/4 inch hole to serve as a screw handle. I cut my screw handles to a length of about 14 inches. Using the screw handle, the ground stake can be screwed into the sand a little over 2 feet. It is wise to sink the stake as deeply as possible, especially if it's a windy day. This arrangement will self-support the DK9SQ mast in winds up to 25 knots. Support lines can be tied to the mast for higher winds.

The DK9SQ fiberglass mast fits the ground stake like a glove. (Photo #2) Simply unscrew the bottom end cap off the mast and insert the bottom section of the mast over the sand stake. The bottom section of the mast will slide down until it is stopped by the screw handle. (Guess what happens to the smaller inner mast sections in the meanwhile?) :-)

Now that you've got the big picture, I would recommend the following sequence:

1) Extend the first section of the mast and twistlock it into place.

2) Take the end cap off the base section of the mast and balance the base on your foot or a clean tarp, keeping the mast vertical.

3) Velcro the antenna wire to the top section of the mast.

4) Extend successive mast sections, twistlock as you go.

5) With the mast fully extended, insert the base section over the ground stake. The mast presents minimal wind loading and is light, so this is not a difficult task even in moderate winds.

6) The antenna wire can be loosely wrapped around the mast several times
or velcroed at several points in order to keep it from flying awayduring high winds. For 20M halfwave elements, I use 34'1" lengths of #22 stranded hookup wire.

The halfwave vertical elements present a base impedance of several thousand ohms. A simple L/C network is used to transform the high feedpoint impedance to 50 ohms. (Photo # 3) (Schematic #1) For fixed station use, switches S2 and S3 may be omitted, with taps being soldered directly to L3 and L4.

Equal length feedlines (each 1/2 wavelength long) are used between the feedpoint matching networks and the phasing controller. I used 22'6" lengths of RG-174 coax to keep the weight and bulk of the total array to a minimum. The feedline losses were deemed to be an acceptable tradeoff.

The vertical elements are spaced 34 feet apart, with the operating site (tent) positioned midway between the elements. (Photo #4) The use of 1/2 wavelength feedlines ensures sufficient length to reach the phasing controller in the tent.

The phasing controller (Photos # 5, 6, 7) & (Schematic #2) is the heart of the system. As can be seen from the schematic, the system is quite simple. Transformer T1 is responsible for providing current to the two verticals.

Let's call Ant #1 the reference vertical, driven from the middle transformer winding. It is important to note that Ant #1 connects to the side of the winding that has the dot on it. The other side of this winding connects to chassis ground. Now let's take a look at the winding that drives Ant #2. The sole purpose of switch contacts S1c and S1d is to allow this third winding to "float." This means that Ant #2 can connect to either the dot side of the winding or to the opposite side of the winding. Whichever side of the transformer that does not connect to Ant #2 is connected to chassis ground via the switch contacts.

If Ant #2 is driven from the dot side of the third winding, then both Ant #1 and Ant #2 are driven "in-phase." If Ant #2 is driven by the lower end of the winding, then the antennas are driven "180 degrees out-of-phase." All we are doing here is taking conventional transformer theory and putting it to work in an antenna environment. Switch S1 allows instant direction switching to take place. More about that later…

Inductor L2 is used to compensate for the wiring associated with S1c and S1d. The use of this small amount of additional inductance enables the feedlines to see precisely controlled phase angles.

Interestingly, the two antennas behave quite differently when they are fed "in-phase" and "out-of-phase." As a result, the primary winding of T1 sees a significantly different reactive component as S1 is switched from one position to the other. Inductor L1, in association with C1/C2 or C3/C4 (depending on the position of S1), allows the transceiver to see an impedance of 50 ohms.

I would highly recommend the implementation of an RF bridge circuit into the same enclosure as the phasing controller. The RF bridge has excellent low power response, easily measuring QRPp power levels. My power meter is calibrated for a full scale reading of 250 mW (although the meter can be scaled to any desired QRP power level). Meter readings are accurate down to under 50 mW. The bridge serves another important role: It allows the L/C network in the phasing controller to be adjusted for 1:1 SWR for both settings of phase switch S1. I used the circuit in Figure 3-6 of DeMaw's "QRP Notebook" (ARRL 1986 ed., p. 49, blue cover).

As has been alluded to in the previous paragraphs, this array uses two halfwave radiators which are separated by one halfwave. EZNEC was used to predict various array parameters including gain, feedpoint impedance, and radiation patterns.

The elevation plot compares identical arrays over average soil (inner plot) and salt water (outer plot). For elevation angles below 5 degrees, the salt water provides quite significant gain relative to the typical backyard vertical installation.

When both elements are fed "in-phase," the resulting pattern is "bidirectional broadside." Picture the verticals as being erected on an N/S line. The radiation lobes shoot E/W, broadside to the N/S axis that the vertical elements are physically lined up on. The EZNEC plot again demonstrates the advantage of operating over salt water, predicting a gain advantage of around 15 dBi!

When the elements are fed "180 degrees out-of-phase," the resulting radiation pattern is "bidirectional endfire." Assuming the array is physically lined up on our N/S line, the radiation pattern is also N/S, out the two ends of the array. Note that the lobes are fatter (with consequently less forward gain) than the lobes in the broadside pattern.

The overlay plot demonstrates the predicted coverage of the array as it is switched from phase 1 to phase 2. Front-to-side ratios of 12-20 dB have been frequently observed when switching the array while listening to fellow Spartan Sprinters. On some occasions, stations have "disappeared" from the frequency upon switching phases. Most stations demonstrate a change from one to three "S units" as the phase is switched, depending on their heading relative to the array.

One can come up with a close estimate of how this array compares with a single vertical by observing the points of intersection between the outer edges of the broadside and endfire lobes in the overlay plot. This plot shows that the points of intersection are about 4 dBi down from the broadside lobes and about 2 dBi down from the endfire lobes. The implications are important. First, a single vertical over salt water is a very good performer. Secondly, a phased array can offer substantial discrimination to QRM over a single vertical-- typically 6-20 dBi. The ability to reduce QRM is oftentimes more important that the extra 2-4 dBi of gain provided by the array.

No radials are used with this array. Theoretically, no radials are needed at the base of a halfwave vertical if the vertical is operated over excellent ground. It took me a long time (and a lot of testing) to accept this fact. This provides a real benefit in time savings. The entire array can be set up by one person in about twenty minutes. Take down time is about fifteen minutes.

I have been quite pleased with the array. It has stood up well to the salt water environment and the performance is excellent. I receive periodic unsolicited reports from fellow sprinters who state that the signal from the array is usually comparable to most of the 5 watt stations. The array also passes the "truly portable" test, ensuring its rightful place on future QRP Kayaking trips.