Introduction

Where an antenna radiator is operated against ground, there is always the problem of getting antenna current running as high up the antenna as possible. Often the highest current is at the base of the antenna where it can do least good for radiation with most of it lost in the earth resistance.

There are various systems of adding capacity top hats and series loading inductors placed well up the radiator line. However we still finish up with almost no current running at the apex of the antenna where high current might be most effective. But here is an interesting system which can put maximum current right at the top.

The Principle


Two plates at the end are connected so that their coupling to ground is unbalanced. This in itself has little effect in unbalancing the line currents but the capacitance between these two plates is resonated with a parallel inductor to produce a tuned circuit with as high a Q factor as possible. The line is coupled unbalanced into the inductor and the effect is to produce a high voltage across the plates equal to Q times the applied voltage. This in effect multiplies the out of balance current reflected to the line by the factor Q.

This effect became very apparent when I first started experimenting with EH antennas (which all have an unbalanced dipole connection) and I found that the current running in the centre conductor of the coax feed line was around twice that running in the outer coaxial braid. The effect was explained in more detail in my previous article (Ref. 2) and in particular in figure 4 of that article which is also reproduced in the appendix at the end of this article. The whole principle I have just outlined became even more apparent when I experimented with my balanced X3 antennas and found out how well they worked by simply unbalancing the tuned circuit formed around the X3 dipole.

But what really showed up as important was the fact that for transmission line lengths of 1/4 wave or less, the unbalanced (or common mode current component) reached a maximum close to the dipole or antenna termination connection and was still reasonably high running back down the line. So here we achieve highest current right at the top or far end of our line pair just where it can do most good simply by making a terminal unit with two plates resonated against a high Q inductor connected for an unbalance.

But we don't need to build a special type of antenna, such as the EH, to produce the phenomenon. Just concentrate on the design of a terminal unit which encourages the current imbalance in the feedline.

There is probably plenty of scope for experimentation with the size if the plates and the separation between them to get best feedline current imbalance. I chose plates 5 cm square and a length about 1% of a wavelength. The only reason for doing this was that I already had the plates available which I had recovered from previous experiments with the X3 antennas.

Terminal Unit for 40 metres

I turned my 40 metre (or 7 MHz) X3 into the terminal unit by discarding the original two coil assembly which it originally had and wound a single inductor placed between the two original 50cm square plates forming a dipole. The arrangement is shown in figures 1, 2 and 3. For the line pair (as in the original X3), open wire TV line is used and this is matched at the transmitter end using a Z Match tuner.

Figure 1 - 7MHz Terminal Unit for Reverse Feed Antennas

Figure 2 7MHz Terminal Unit

Figure 3 - Coil of 7 MHz Terminal Unit

Terminal Unit for 80 metres

I figured that the reverse feed idea might have more application to the 80 metre band as most radio amateurs usually find they don't have enough space for an 80 metre half wave and usually have to operate what wire they can fit in against ground return. For 80 metres, I tried out two plates side by side but spaced 10 cm apart. This reduced the size of the terminal unit to that of a dipole assembly and also increased the capacitance between the plates. The higher capacitance allowed me to tune the circuit to resonance with a smaller coil than would have been required for the dipole assembly.

The arrangement for the 80 metre (3.5 MHz) system is shown in figures 4, 5 and 6. Open wire line pair is a bit hard to get so that I have given an alternative as figure 8 power flex for the transmission line. I have suggested this as an possible alternative as its transmission loss is fairly low at 3.5 MHz and I have carried tests using it. However I don't recommend its use on the higher frequency bands where the losses can get quite high (particularly in the resonant feeder type mode which is used). Figure 7 is a photo of the open wire TV type line alongside the power flex.

Figure 4- Reverse Feed with 80 metre Terminal Unit

Figure 5 80 Metre Terminal Unit

Figure 6 - Coil of 80 metre Terminal Unit

Figure 7 - Open Wire TV Type Line and Figure 8 Power Flex

Tuning Up


The precise frequency of resonance between the plate capacitance and the shunt inductor in the Terminal Unit is not critical. It is simply set somewhere within the frequency range of the relevant band. This can be easily checked using a dip meter poked near the coil with the feedline disconnected from the coil. As the open wire feed line is allowed to become partly resonant itself, reactance is fed up the line from the Z Match to bring the antenna circuit precisely in tune and properly matched by adjustment of the Z Match.

The degree of current imbalance (or common mode current) which results in radiation from the feedline can be checked by inserting an RF ammeter in each leg of the line connected to the Terminal unit. Using the dipole type plate assembly as shown in figures 2, 3, & 4 (and with plates about 1% of a wavelength long), I usually find that the current in one line leg is about double that in the other. In the case of the 80 metre terminal unit as shown in figures 4, 5 & 6 (also with plates about 1% of a wavelength long) , the currents measured in the ratio of 2:3. Perhaps this lesser ratio means that particular design, set out to limit the physical size of the assembly and the resonating coil, could be improved, or perhaps it is just a function of the lower operating frequency of 3.5 MHz. This might be a further area of experimentation.

If the length of unbalanced feedline current extends more than !/4 wavelength down the line, the common mode current peak could be further down than the top end of the line. I suggest that if the line length is greater than 1/4 wavelength, some form of balun unit or out of balance trap be fitted at the 1/4 wave point so that the line currents beyond that point towards the transmitter are balanced.

It seems unlikely that in the usual amateur radio backyard, more than 1/4 wave of line for 80 metres would be needed, requiring a trap. Also the balanced output of the Z Match might provide a suitable limit to stop too much common mode current component getting back into the radio shack. This might not be true for 40 metres but I have not yet attempted to build a suitable trap for the twin line on that band. However I have made and tested one for 20 metres and this is shown in figures 8 and 9. Clearly for 40 metres, the coil would require enlargement for longitudinal resonance at 7 MHz. One might think of using a ferrite type of core in a coupled circuit but there can be difficulties in selecting the right core characteristics to suit the particular line impedance at the point of insertion. Get it wrong and the result can be an overheated core and for this reason I prefer to stay with the open coil .

Figure 8 - Balanced Trap for 20 Metres

Figure 9 Balanced Trap for 20 Metres 13 turns of "Figure of 8" Power Flex, spaced to 11 cm. Former - 4 cm diameter Inductance 6.5 uH Parallel Resonant at 14 MHz (Tuned with 2 x 10 pF capacitors)

Summary

Here is an interesting idea to get high current running at the top of an antenna wire by feeding the signal to the top and exaggerating an out of balance condition at the top using capacitance plates resonated with a high Q coil.

If you have been reading my most recent articles on experiments with the EH antenna (which has an unbalanced resonant termination) and the unbalanced version of my X3 antennas you will realise that I believe they are relying on the phenomenon I have just described for their good performance rather than the claimed crossed field theory.

I think that there is plenty of room for more experimentation in the best design of a terminal unit to produce the out of balance current condition down the feed line legs. I think if I were a Broadcast Station engineer, I would be interested to see whether the system I have described would generate higher field strength than some of the well established top loading antenna systems now used.

Some Interesting References (which lead up to the reasons for publishing this article)

(1) The VK5BR-X Antennas. Some modified ideas on how they work. By Lloyd Butler VK5BR - Amateur Radio, April 2006

(2) Some Different ideas on the EH Antenna. By Lloyd Butler VK5BR - Amateur Radio, June 2006. Also refer to copy on the Web page.

(3) On the Web - THE VK5BR X2-X3 Small Dipole Antennas.

(4) On the Web - Some interesting results in operating the X3 out of balance

(5) On the Web - EH & X3 Antennas (As published in QRP Club Journal "LoKey").

Appendix
Q Multiplication of out of balance coax line currents in the EH Antenna

Figure 4 of Reference 2 Shows how Antenna Q multiplies Common Mode Component on the Coax Line

Back to HomePage