bld_cfa.gif (23641 bytes)HOW TO BUILD A 75/80 METER CFA
by
Maurice Hately - GM3HAT, Dr. Brian Stewart - MM1DVD,
Dr. Fathi Kabbary and Ted Hart - W5QJR

a_cfa.jpg (1311 bytes)fter the preceding series of articles on the theory of Crossed-Field-Antennas (CFA), I am sure most of you will think this is the most important article in the CFA series. It presents the details on how to build and tune a CFA for 75/80 Meters. Actually, this is a two-part article. antenneX editor, Richard Morrow, K5CNF has taken the details and devised methods to simplify construction of this CFA. That article is entitled CFA Construction Guide for 75/80 Meters and will be found in this same January 1999 issue of antenneX. As supporting documentation, readers are encouraged to obtain a copy of US Patent # 5,155,495 which may be found at the IBM Patent Server <www.patents.ibm.com>.

Mr. Hately and Dr. Stewart selected the 75/80-Meter ground plane version of the CFA as one of the least complicated for construction by Hams. This same physical unit works well on 160 and 40 Meters as well as 75/80 meters, but requires significant changes in the phasing/matching network. In addition, the toroids create problems on 160 meters due to being lossy at the lower frequencies. All of you who are interested in experimenting will find this CFA easy to build, tune and achieve high performance. With that experience behind you, some trial and error will allow you to achieve comparable performance on other bands. antenneX will be interested in heating about the reports from you on your experiences. The best phasing/matching network designs reported by eXperimenters for the other bands will be reported to allow sharing of that information.

PHYSICAL DIMENSIONS
NOTE: The dimensions used by the developers are in metric units. To convert cm to inches, divide cm by 2.54. Approximate values are presented, exact values can be determined by the equation above. However, the approximate values shown are sufficiently accurate.

cfa-fig1.gif (5584 bytes)

With reference to Figure 1, the following physical dimensions are presented:

Ground Plane: 1 meter square (approximately 40 inches on a side) made from mesh or sheet aluminum.

D Plate: 40 cm (15.8 inch) diameter aluminum, with a hole of about 6 cm (2 1/4 inches) diameter cut out in the center. Position the D plate 10 cm (4 inches) above the ground plane with a suitable insulating material.

E Plate: Cylinder height 25 cm (about 10 inches), Cylinder diameter 20 cm (about 8 inches). Position the E plate (cylinder) 10 cm (about 4 inches) above the D plate, using a suitable insulating material. Construct the E cylinder from aluminum, or if desired, aluminum mesh.

Phasing/Matching Network: The schematic diagram presented in Figure 2 details the phasing/matching network. The values presented on the schematic are those necessary for operation on the 75/80-Meter Ham band. Many Hams have reported that they have tried to build a CFA but were not successful. Unless the CFA is properly fed (as shown in the schematic), its performance will be poor or not at all.

Caution: The toroids and wire specified are for transmitter powers of 100 watts or less.

With reference to the schematic, the coax from the Transceiver may be any desired length. Best operation of this CFA will be achieved when it is located clear of all obstacles. Although the height is not critical, it may be necessary to mount it on the house roof or on a pole to achieve obstacle clearance. Tuning for the center of your desired operating frequency range should be accomplished with the desired length of coax. After tuning, the unit can be elevated to the desired height. Please be aware that this antenna has the same radiation pattern as a vertical wire antenna, thus it is not a good antenna for short-range communications (for distances less than about 500 ground miles or 800 km). The antenna can be mounted on its side and simulate a horizontal dipole. To operate in this manner, the height above ground will dictate the radiation pattern, the same as any horizontal antenna.

Transformer T1 provides impedance matching, thus, the number of turns of wire on the secondary is not the same as the primary. Since the secondary impedance is higher, thus the current is lower. The wire size on the secondary may be smaller than on the primary. There are two secondary windings. These should be slightly twisted prior to winding on the toroid. Be very careful to observe that the two windings are out of phase with each other. This is very important to successful operation.

Both L1 and L2 consist of 22 turns of # 20 gauge wire wound on the same type toroid as T1. Since there are three (3) toroids required, and they are a specific type of toroid to achieve the desired inductance, antennex has made a special arrangement to make these toroids available to those who need the information. The ordering information is contained in the construction article.

cfa-fig2.gif (15126 bytes)

PHYSICAL CONSIDERATIONS FOR THE PHASING/MATCHING NETWORK Transformer T1 is best positioned underneath the ground plane, but it can be positioned in the D plate gap. The primary has 4 turns of # 16 wire. The secondary is wound with 20-gauge enameled wire, and consists of 2 windings each of 9 turns with opposite ends connected to the ground plane as shown in Figure 2. These wires can then be fed up from the ground plane towards the E and D plates. The inductances of the secondaries are not too critical for successful operation of the CFA but will be of the order of a few uH.

One end is the feed for the E plate, but before being connected to the E plate, it feeds a parallel resonant LC circuit (see Figure2) which is placed in series between the E plate feed and the final connecting wire to the E-plate. The parallel circuit comprises a variable 400 pF capacitor (e.g., ganged) and a 22 turn wound (again 20 gauge wire) T130-2 toroid. The value of L should be about 5.54 uH. This tuned circuit should be secured inside the E plate cylinder using insulating material on which the tuned circuit can be mounted. A note of caution is that the capacitor should be accessible for adjustments by hand, about 30 or 40 cm (12 to 16 inches) away from the E plate. An extension made of either plastic tubing or a shaft with an insulated coupler may be used.

The other end of the feeder transformer feeds the D plate. However, it has an unusual arrangement, because the feed splits into two—see Figure 2. One branch feeds the D plate through a 22 turn winding (20 gauge wire) on a T130-2 toroid placed in series. The inductance is again about 5.54 uH. The other branch is connected to one end of a variable 400 pF capacitor which is then connected directly to the ground plane. Again, this variable capacitor should be made accessible for adjustments, and should be secured somewhere close to the D plate.

SETTINGS: For the CFA described above, on the 75/80 Meter band, appropriate settings for both capacitors should be about 330pF. The E plate capacitor is generally the more critical to set of the two capacitors.

TUNING: A field strength meter should be placed at a suitable distance from the CFA, and both capacitors should be "tweaked" accordingly until radiation is evident. As radiation occurs a drop in SWR towards 1:1 should be observed. The field strength meter need only be a very simple diode type with about 3 feet of wire as a suitable antenna. This will allow tuning using very low power, typically much less than one watt. You may desire to begin with a higher power, perhaps as much as 5 watts. As the tuning progresses, reduce the transmitter power as required to keep the field strength meter on scale. Be aware—the VSWR bridges in some transceivers will not indicate actual VSWR unless the transmitter power level is greater than a few watts. In that case, simply tune the CFA for best radiation as indicated by the field's strength meter using low power. Following that procedure, increase the transmitter power to 10 watts or more and adjust the capacitors for lowest VSWR.

If the CFA is adjusted at some frequency, say 3.9 MHz, then the next step should be to measure the range of frequencies over which the VSWR remains less than the value at which the transmitter automatically reduces the transmitter power due to high VSWR. This will be the range over which the CFA will allow operation without readjustment.

This CFA will operate at other frequencies outside the 75/80-Meter band. It may require alteration of the windings, with different taps being required in order to achieve successful radiation at other frequencies. The keen amateur is encouraged to experiment.

Editor's Note: This is one of four CFA construction articles of the 9-part series on the CFA. You will find the article vitally informative on building a CFA, but do not use the phasing/matching network described in this article. A flaw was found in the phasing/matching network and has been revised in two additional construction articles in the issue for February 1999 now in the Reading Rooms. If the reader is interested in building a CFA, the T130-2 Toroids needed are found in the Shopping Shack in the "Equipment" category. To read the other five parts of this exciting series goto: Subscribe (Subscription Option #2 has the full series)

RECEIVING
This antenna will perform well on both transmit and receive. Prior to the conception of the CFA, receive antennas were characterized by their "capture area." Even though the CFA is extremely small, it does not have a corresponding small "capture area."

All of us hope you enjoy using the CFA as much as we have enjoyed bringing these articles to you. For those of you who prefer to buy a CFA rather than build one, please stay tuned. Those details will be presented in the near future.

It has been stated that the CFA is a new concept and that various configurations are still being explored. We expect to present those variations to you as they mature.

We implore you to remember that the development of the CFA is protected by patents and any commercial gain from these antennas is solely for the benefit of the inventors.

From the days of Marconi until Mr. Hately and his coharts came along with the CFA, there were only wire antennas. These people have done an incredible job in a short time to develop and deploy Crossed-Field Antennas. Because the CFA is new and is a radical departure from conventional wire antennas, many nay-sayers have pronounced the CFA to be less than it is. Now, any one who builds this version of the CFA will certainly become a believer. I am reminded of a quote from Dr Stewart, "Antennas will never be the same.quot; -30-

Editor's Note: For your convenience, we have placed one of our new Interactive NotePads below for any questions or thoughts while fresh on your mind. Just type in your message (including E-mail address if a reply is needed) in the NotePad and press the Submit button.

Interactive NotePad


Click to Print/Save Article

Send mail to webmaster@antennex.com with questions or comments.
Copyright © 1988-1999 All rights reserved worldwide - antenneX©
Last modified: April 01, 1999