On the Air with a CFA
Established Rules Show Cracks
Antennas have fascinated me since the time of the famous question put by many children to their fathers—”where is the little man speaking in the radio box, Papa ?” Later on, as things were clear, settled by established math, the antenna world seemed to be an axiom with the stability of a massive granite block. Now, this granite block started to crack in December 1998 when I first got the information from Anselmo (HB9ICK) and respectively from antenneX, about a new concept of a non-resonant antenna! I do not have to describe the degree of growing excitement which followed.
The pleasure was even larger to find involved our good old friend Maxwell (GM0–?), sending us warm and meaningful 73s! Well, if not resonant I said to myself, you can build one which fits everywhere (even for 160m). So I started to look for a cylinder to build the antenna described in the patent and later in the series of antenneX articles. I found a cylinder in the Mall near my house, with a diameter of 130mm made of aluminised steel, usually used for chimneys and I cut a piece of 160 mm as shown in this first Photo and the sketch, (pos. 1).
Durable Experiment
I must tell you I rejected, right from the beginning, the idea of making the antenna of cardboard first because of my conviction concerning solidity, and secondly, because I wanted to avoid any suspicion of any mechanical nature when testing. So I continued to cut from Al tin 1mm thick, the D plate with the diameter of 260cm and the square ground plane 65 x 65cm. The spacers (see sketch, pos. 6) are made of PVC and Nylon screws in order to ease adjustment of the plate distance to the ground. The SWR improved at a distance lower than 0.5d (See relative CFA dimensions from antenneX article).
The D plate has an orifice of 40mm in which I placed a plastic ring with a collar ( pos. 2) to hold the support (pos. 3 ) of the cylinder. As one can see from the drawing, the cylinder is hold in position by two threaded rods (pos. 4) fixed via the PVC rings ( pos. 5 ) on the vertical support.
That Elusive Matching Network
The second step was building the splitting/phasing circuit. I made the two coils as described in the antenneX article. I took two capacitors of the right value (air spaced 0,5mm) then winded the transformer on a T200 core and started to solder the connecting wires. The parallel L/C circuit has been located partially inside the cylinder at the upper part (see sketch and next Photos). The other components were fitted in an aluminum box screwed under the ground plane. The connecting cables to the stimulators were simple PVC insulated wires 0.75mm diameter, which passed through a 10mm hole drilled in the ground plane. The first test at 10W was negative (as usual, hi!).
Then, with the RF Analyst I could find, with some patience, a position of the capacitors where the SWR was 1:1, but at 10W nothing happened and the SWR was very bad. I tried several times the with same procedure without any success. The next try I used the full power of my TS520 (used for such experiments) and looked into the direction of my new “baby.” I was very surprised to see at every one of my 100W whistles, a tiny light coming out by reflection from the cylinder. That was it, discharges in the capacitor! I redesigned mechanically the circuit by replacing the capacitors with vacuum ones (2x Jennings, 5-100pF/15KV Photo 1, 2 and 3). I had them in the shack so one should not worry about finding the same values. It is just important to know that we have at resonance quite high voltages developed at that level. I changed also the way of tuning the circuit by using two long plastic tubes (1.5m) connected to the axes of the two trimmers.
Fine Tune Time
The first adjustment was at the top circuit (parallel L/C) which had as result a dip at the SWR Meter and then I tuned the lower capacitor for even less reflection. Repeating the procedure I succeeded to have zero power reflection. At this point I could read on the Analyst an impedance at the transformer input of nearly 50 Ohm. The first QSO was from inside the house in the attic (my shack) with modest reports. Then, I placed the whole construction through the window onto the roof, with an inclination of 45°. Re-tuning was not necessary and the QSOs received very good reports on 40m (58-59). The bandwidth tested was only 10kHz which is not terrific, having in mind the stated advantages of the CFA, but okay if you consider the reduced dimension of the system and the rather primitive phasing circuit.
Some Tips
Another constructional hint is to short circuit the cylinder with a diagonal wire as recommended also by the antenneX articles. I think this is important in order to create a uniform skin current in the cylinder wall that has as consequence of producing a uniform electrical field build up between the cylinder and ground plane.
Another experiment worth doing was to increase indirectly the antenna dimensions by increasing the operating frequency. I reduced the inductivities of the Phaser to approximately 1/4th (I took only 6 windings) and re-tuned the circuit for 14.200. Tuning was very comfortable and I quickly got an SWR of 1:1. Also I increased the surface of the Ground Plane to 1 sqr meter. The result was a bandwidth of 100KHz and comparable signals with my R5.
Summary
The experiment with a reduced size CFA for the 40m band, proved the feasibility but also displayed the problems which arise at this dimensions (l /200). It is thus advisable to build a system with dimensions between Lambda/100 and Lambda/25 in order to get satisfactory results (the theory will confirm in the future the optimum size). Experimenting with Crossed-Field Antennas represents a very exciting task not only for the sake of experimenting, but also for giving new input to our imagination and help us learn more about mother nature.
The experiments do not stop here. There are still a lot of interesting problems to solve like:
1. Find a simulation program for computing the field magnitudes and shape
2. Find a reference CFA similar to the elementary dipole
3. Imagine perhaps other geometrical configurations (directive CFAs ?!)
4. Design an inexpensive, large bandwidth, hybrid splitter/phaser circuit
5. And so on…..
I do wish all good success in achieving more experience with CFAs!
Originally posted on the AntennaX Online Magazine by Werner, DL6NDJ
Last Updated : 14th May 2024