CFA - Looking for Poynting Synthesis
In last month’s article (The CFA Radiation: Matching is Not Radiation!) published in the July 2000 issue of antenneX, I examined the minimum lead length CFA/tuner output as I varied the tuning, while keeping track of the amplitude and phase of the E and D plate voltages and currents. I observed no electronic artifacts as I tuned through the critical 90 degrees between E and D plate voltages. Nothing indicated anything about the correctness of the tuning. So, I decided this month to cut the tuner out of the experiment and attempt to prove or disprove the stated phenomena of synthesizing a radiation (poynting) vector out of two separate field sources. I constructed a magnetic field source and an electric field source and mounted them physically so that the fields generated pointed in different directions by 90 degrees.
The signal generator output is split into two channels with a resistive divider that has an impedance of 50 ohms at all ports. Each channel voltage is then adjusted in amplitude using switched step attenuators that allow steps as small as 1 dB. The source radiators are placed across 50 ohm loads so that the impedance of the capacitance (the E source) and inductance (H source) is much larger than 50 ohms. In this way, the voltage is forced to the desired level for each source without being effected by the reactance of the field generating structure.
An oscilloscope that is externally triggered by the RF cycle at the signal generator always begins the trace at the same point on the RF cycle. The two channels of the vertical amplifier are connected to the current sensor of the H source, and the drive voltage of the E source. Since the inductance is driven by the RF E voltage, the current is delayed 90 degrees from the drive voltage. If in phase operation between the E and H fields is wanted, a 90-degree delay is inserted in the drive to the E source by a cable. The scope checks this phase by showing the voltage and current waveforms. An attenuator between the power splitter and signal generator isolates the triggering from any effects of amplitude adjustments. The size of the voltage and current is read from the scope screen.
Figure 1 shows the block diagram of the test set up. E and H are in phase.
- H source attenuation
- H source current (from scope)
- E source attenuation
- E source voltage (from scope)
- Near field detector output (from AC vtvm)
- Far field signal (from spectrum analyzer)
For a fixed H source drive (or attenuation level), step the E source from its maximum (0 dB attenuation) to the smallest signal size (by adding attenuation) that can be seen to change the receiver outputs. Do this for a number of steps across the range of H source drive until reaching the level where the signal can no longer be detected.
The H field generator is a two-turn loop wound on a 6-inch (15 cm) PVC pipe 38 inches (call it 1 meter) long. The pipe is drilled at the top and bottom to pass two wires (approximately #14 stranded wire) across the same pipe diameter at each end. The diameters at each end are 90 degrees to one another. Thus, looking down the pipe from the top end, imagine a clock face at the top and the bottom of the pipe. The top holes are drilled at 12 o’clock and 6 o’clock (2 holes each for 2 wires about 1/4-inch apart). The bottom holes are drilled at 3 o’clock and 9 o’clock (2 holes each for 2 wires.)
The 90-degree difference between the wire crossing diameters is intended to make the polarization of the top H field, which is the desired output, be orthogonal to any radiation from the bottom wires. It is an attempt to isolate the top radiator from any canceling radiation from the bottom wires using polarization. Further, the loop is wound among the holes so that the current in each of the top wires is going across the pipe in the same direction, while at the other end, the current crossing via the bottom wires is going in opposite directions in the two wires. The sequence of holes along the wire is listed:
Start
1. 3 bottom to 9 bottom
2. 9 b to 12 top
3. 12 to 6 t
4. 6 t to 9 b
5. 9 b to 3 b
6. 3 b to 12 t
7. 12t to 6 t
8. 6 t to 3b finish
The E field generator is a 6-inch square capacitor spaced 0.4 inches (1 cm) and sized to just fit into the pipe. A 1/2-inch (1.3 cm) hole is drilled in the center of the plates for passage of the two top wires of the H field generator. The E field generator is inserted in the pipe from 3 o’clock to 9 o’clock and the H field loop is wound through it. The capacitor spacing is just the spacing of a piece of 300-ohm twinlead that connects the two capacitor plates to the bottom end of the pipe. Reversal of the E field between the plates is done by reversing the connections between the drive channel and the twinlead.
The current sensor is a toroid with 30 turns in the output winding. The H field generator wire is threaded through the toroid to form a 1-turn primary. It was found that a 32-ohm load on the secondary produced .1 volt when 100 ma (peak to peak RF current) flowed in the field generator. The scope voltage reading = the current flowing in the field generator. Since the two wires have current flowing in the same direction in the output end of the generator, twice the effective current produces the H field.
The following will present and explain one block of data: a single H source level and its associated E source levels. Then a summary table will show why I think the CFA is just two independent sources. After that. the rest of the data blocks will be listed. The intent is for others to duplicate the experiment and enlarge on the observations.
Table 1 is a typical block of data. It shows the detected radiation from a 7 milliampere current combined with an E field from a range of voltages across the E field capacitor. The first column is the attenuation in dB introduced in the H field generator channel. The second column is the resulting current in the generator loop wire. The third column is the attenuation in the E field generator channel. The fourth column is the voltage across the plates of the E field generator. The fifth column is calculated, and is:
Number = 10*log(H generator current x E generator voltage)
The next two columns are the two receivers’ outputs – the near field detector output in millivolts, and the spectrum analyzer signal display in dBm. (dB down from one milliwatt).
Looking at column 7, the first line shows the analyzer noise floor is -93 dBm. The two green background columns, #3 and #7, are the voltage adjustment and resulting response. The data show that the detected signal is below the noise floor when the E attenuator is above 30 dB. At 18 dB, the detected signal is -89 dBm. At the other end of the table, 0 dB attenuation gives a signal level of -75.6 dBm. Thus, the input dynamic range of 30 dB results in a 17.4 dB change in radiated power. The calculations (Column 4, blue background) show a 15 dB change in ExH “power”. This calculated number is proportional to the change in input power and should be one half the attenuation range in the third column – it is a check that everything is working. This data block shows that all the radiated power comes from the capacitor since the output dynamic range is nearly the same as the calculated dynamic range.
Compare this table with Table 2, where The H field attenuator is 0dB and the H field generator current is 300 ma instead of 7 ma. Here, the measured output power in column 7 stopped changing after only 10 dB E field attenuator change, and the E field generator only produced 1 dB total change in the output power. The same 10 dB attenuation change in Table 1 produced 7.5 dB change in detected power at low H levels. The detected power in Table 2 is coming mostly from the H field generator.
Table 3 is a summary table. It is constructed from the data in Tables 1 and 2, and 4 to 6. The yellow background grouping refers to changes in output due to E field change, as dilution is changed by increasing H field. The purple columns are test changes and the resulting output. The green column is the “power” calculation dynamic range of the change, and the rust-colored column is the various fixed levels of the diluting field. The blue background grouping looks at the dynamic range change in H related power as the fixed level of E field is adjusted.
Table 3 shows that both E related power and H related power can be diluted by relatively larger amounts of the power from the other field. Only when the other field is low does the total power accurately follow the change in one component. This shows that the total power is the sum rather than the product of E and H when E and H are generated from separate sources. If there were the asserted “synthesis” of ExH, the dynamic range of the total power in this experiment would not compress since each field would multiply the other by any change.
My Conclusion
My conclusion that superposition does indeed reign supreme has one caveat. Recent communications have asserted that the four Maxwell equations are simplification of Maxwell’s original set of 20 equations that was done by Oliver Heaviside (see Ed. Note below). The CFA inventors hint that the CFA really works by means of some of the lost equations, and that is why H field produced by displacement current is required. Therefore, in the next period the experiment described above will be duplicated using the conventional top hat CFA configuration, both with and without a plastic dielectric in the D plate H field generator. I hope those with access to the required equipment will try to duplicate my efforts so that some kind of consensus can be developed about the workings of a CFA.
Editor’s Note: We have been reminded that these “four equations” are not that of Maxwell but were modified by a mathematician named Oliver Heaviside. “Maxwell’s Equations” were a set of 20 equations and 20 variables set in complex quaternion notation. Quaternion notation came from Sir William Hamiliton. After Maxwell’s death Heaviside reduced them to four and in vector notations. This is what has been taught in schools ever since. Experts over the decades who have looked at the ‘new work’ and have found some 22 errors. The term “fields” came from Heavide and Hertz – not Maxwell.
Originally posted on the AntennaX Online Magazine by Joel C. Hungerford, KB1EGI
Last Updated : 21st May 2024