Care and Feeding of the CFA!
The previous articles in this series on the CFA defined the concept, presented the history, presented evidence of the outstanding performance of AM Broadcast versions of this antenna, and finally presented a construction article (January 1999 issue) to allow any Ham to build and use one on 75/80 Meters. Unfortunately, that article raised more questions than it answered.
The construction details presented certainly should not, and have not, raised any questions. All of the questions are related to the phasing/matching network presented in the article, and rightly so, because that article contained errors. This article will set the record straight. However, let me say first that a number of Hams have reported issues that need to be addressed, to assist experimenters understanding of what is happening.
Narrow Bandwidth
It is easy to adjust a network such that a high voltage develops on the E plate and sufficient voltage is applied to the D plate to cause low VSWR over a very narrow bandwidth. The performance is many dB down from a good antenna, in the range of about 20 dB, based on my experience. When this happens, the CFA is not operating properly, since the radiation resistance of the elements has not developed due to improper phasing and proper voltage ratios to meet the criteria for transductance to 377 ohms of free space. When the radiation resistance in both elements develops, it will be low, with the actual values depending on CFA size and frequency. Low is defined as much less than a few hundred ohms. That resistance is reflected in the phasing/matching network and results in a low Q (wide bandwidth) antenna system. When we talk about the antenna system, we are talking about the complete system including the phasing/matching network and the physical CFA.
Hot Toroids
Since the radiation resistance has not developed, very high reactive circulating currents are present, thus causing heating in the toroids. Changing to wire wound coils without toroid cores allows the heat to escape easier, but is not a cure for the problem.
Due to numerous Hams reporting difficulty with the system in the construction article, I (W5QJR) was asked to build one and relate my problems/cures to assist the readers. I have built many CFAs, but did not build the one in the construction article due to press of time before publication deadline. The following relates my activity during the process of construction and tuning, and is reported to enhance the understanding of the CFA, and thus promote experimentation with this new concept.
Construction
Since I had no plans to use this particular CFA as a permanent antenna, it was decided to construct it with minimum effort, thus creating a “disposable” antenna. The CFA lends itself to easy construction in a very small space, when the antenna is only for experimentation. The CFA operates on low voltage and low current, allowing construction techniques that may not be familiar to Hams. Indeed, the CFA is not familiar and no other antenna may be constructed in this manner. This is Heaven on earth for antenna experimenters.
E Plate Construction
As I have done numerous times, I obtained a large cardboard box and began to cut circles the required diameter of the E plate (cylinder). Then I “rolled” the circles with stiff paper (file folders) to form a cylinder somewhat longer than the specified dimension. Actually, I made it the length of the E Plate plus the spacing between the E and D plate. Next, I covered part of the cylinder with aluminum foil to match the specified dimension of the E plate. The remaining cardboard then became the spacer/insulator. Since there is no “high” voltage and we will be operating in the house away from weather, this paper doll construction process is adequate, and it’s utter simplicity is an elegant solution to minimum cost antennas for experimentation.
D Plate Construction
The D Plate is simply a round piece of cardboard covered on one side with aluminum foil.
Ground Plane
Another piece of cardboard covered with aluminum foil serves very well as the ground plane. If the groundplane is round, it should have a radius at least equal to the height of the CFA. If it is square, then the width should be at least twice the height of the CFA. A larger area is good, but not necessary and, if very large, will enhance low angle radiation. The width of the aluminum foil that was readily available was less than the required width of the ground plane. By overlapping strips of foil and putting tape over the joint, adequate conductivity is achieved.
Assembly
After a wire is attached to each element (I simply attached the wire using tape), a strip of cardboard was cut whose width equaled the specified spacing between the groundplane and the D plate. It was then bent into a circle and set on the ground plane. The D plate was set on that, then the E plate was set on top the D plate. That completes the assembly, and the entire construction and assembly only takes a few minutes. This approach allows easy, inexpensive experimentation before building one from metal and gold plating it before you had proven to your self that it works. Also, you may wish to experiment with a different size CFA. Few of us would buy a car without driving it. This is the equivalent – – test a “disposable” model first.
Wiring
Questions have been raised on how to get the wires between the network and the elements. If the network is below the CFA, then poke a hole in the groundplane and run the wires up to the elements. For an experimental model like mine, they just sort of drape around until they get there. Remember, a piece of wire 1 foot (1/3 meter) long is less than 1.5 electrical degrees at these frequencies. Just a slight change in the tuning will compensate for that, so it is not an important issue. My CFA is at the far end of a large table.
The Network
I built the network per the schematic diagram in the previous article and quickly learned that there were errors. A couple of e-mails with Dr. Stewart and a couple of phone calls with Mr. Hately quickly identified and corrected the errors. All of us want to apologize for any inconvenience it caused you, and we are sure you will be pleased with the circuit included in this article. In addition to correcting the errors, changes were made to ease construction and tuning for experimenters.
Input Transformer
With reference to Figure 1, Note the unusual arrangement of the input transformer, designated L1, which provides several features. First, to minimize losses, it is a single winding on a toroid (T130-2) used as an autotransformer, and produces two out of phase signals to the two elements of the CFA. Secondly, it also allows variation of the ratio of applied voltages to the E and D plates, while simultaneously changing the impedance reflected into the “primary” of the transformer. This is accomplished by moving the “ground” tap. The transformer has a total of 20 turns of #14 wire, tapped at 12 turns from the bottom. For best VSWR, it may be necessary to move the tap up or down one turn.
In the previous article, the transformer was designed to allow operating the CFA in a high impedance mode. Unfortunately, the wrong cores were specified for the input transformer as well as the inductors in the circuit for operation on 80 Meters, thus the major problem with the previous circuit. One of the previously specified toroid cores (T130-2) is used in this application and works very well when used as indicated in this article. However, for the necessary inductors, we now choose to use “airwound” coils.
Inductors and Capacitors
For 75/80 meters a coil form diameter of 1+3/8 inches (3.5 cm) is used for each inductor (L2 & L3) and each has 21 turns of #14 wire close spaced. Values for the capacitors for operation at 3.8 Mhz are as follows: C1 = 152 pfd. across the E Plate inductor and C2 = 130 pfd. across the D Plate.
Tuning Procedure
You may try a shortcut by skipping step 1, but if it does not work out, then you must begin here. The goal of the tuning procedure is to achieve maximum radiated power and minimum VSWR.
STEP 1: Apply a signal to the network using low power at the desired operating frequency. With a field strength meter adjust both variable capacitors to achieve maximum radiated power. The circuits interact, thus necessitating the need to work both capacitors to achieve maximum output as indicated on a field strength meter. The interaction is due to an effective radiation resistance between the plates and ground, thus indicating that the CFA is “active”.
STEP 2: Observe the VSWR meter on the transmitter and adjust the variable capacitors as necessary to achieve a 1.0:1 VSWR. If the VSWR is low but not perfect at the desired frequency, it may be necessary to change the ground tap on the input transformer as stated previously.
Performance Parameters
The following are measured parameters at a center frequency of 3.8 Mhz:
2.0:1 VSWR Bandwidth – – – – – -> 26 kHz
3 dB Bandwidth – – – – – – – – – – – > 76 kHz
It should be noted that the bandwidth, compared to any small antenna at this frequency, is remarkably wide. If this particular CFA were tuned for twice the frequency (40 Meters) the bandwidth should be greater. To verify this, I reduced the number of turns on the inductors, adjusted the variable capacitors to tune it on 40 meters and measured a 2:1 VSWR bandwidth of 162 Khz. Conversely, if this particular CFA were tuned for 160 Meters, the expected bandwidth would be very narrow. Hence a larger antenna is desired at 160 Meters, unless you operate on a spot frequency. This information allows approximating the desired size of a CFA for a particular application. In other words, it would not be beneficial to use a CFA as large as this one for 10 Meters, unless you wanted very very wide bandwidth.
The values presented above exemplify this extraordinary antenna. It is even more fascinating to note that the only source of inefficiency in this system is a small amount of loss in the E Plate inductor. At 100 watts CW there is a small amount of heating. I do not have an accurate measure, but it is estimated that it is equivalent to two or three watts, offering a total system efficiency greater than 95%. As a point of reference, a 75 meter mobile antenna has an efficiency of less than 3%. It needs to be noted that as the bandwidth is widened, due to either increasing the size or increasing the frequency, the loss in the inductor is reduced. Thus, a larger CFA at this frequency would have greater efficiency. For reference information, 5% loss in efficiency is only 0.22 dB and there are 6 db in an S-unit. Therefore, the in-efficiency is less than the width of the needle on the S-Meter. However, a 75 Meter mobile antenna with an efficiency of 3% is 15.2 dB below 100% , or about 2.5 S-Units. Another way to compare performance is to say that 3 watts into a CFA would equal the performance of a mobile antenna with 100 watts applied.
There is another comparison that will help put the performance in perspective. On 40 Meters, this particular CFA (made of cardboard and foil) has a total height of 1.2% of a wavelength, and has an operating Q = 44. A typical 1/2 wavelength dipole has a Q of about 30. The CFA would not have to be much larger to have the Q reduced to a value much lower than that of a dipole. Although precise numbers are not available, it is recommended that experimenters start with a CFA having a height of about 1.5% to 3% at the desired operating frequency. This will produce reasonable operating bandwidth, yet will also be very small.
I will predict that given time the CFA will be the antenna of choice for most Hams. In particular, due to it’s great performance, the guys that operate QRP will really latch onto it. Also, due to it’s small size, those folks that enjoy QRP field operations will love the CFA, especially above 10 Mhz where it can become a very portable antenna. For example, a CFA having a total height of 1.5% on 20 Meters would only be about 1 foot tall. Anyone for 10 Meter portable? You would only need a total height of about 6 inches. Now we are really talking small when the antenna is smaller than the smallest QRP rig.
Anyone would have to agree that Mr. Hately, Dr. Kabbary and Dr. Stewart created a remarkable concept and put it into practice. I trust that the evidence placed before you will convince even the most ardent naysayer to confess that this is an extremely valuable contribution to science in general and to the art of communications in particular. If you then still have any doubt relative to the full capability of the CFA, I have learned that the inventors will be making a presentation to the National Broadcasters Association annual convention emphasizing the outstanding performance of the CFA. That presentation includes verification of the performance of the Egyptian CFAs by experienced broadcast engineers other than the inventors. That will occur in April 1999 at Las Vegas.
All of us wish you great pleasure experimenting with and using the Crossed Field Antenna.
Originally posted on the AntennaX Online Magazine by Maurice Hately, GM3HAT. Dr. Brian Stewart, MM1DVD and Ted Hart, W5QJR
Last Updated : 2nd May 2024