ANOTHER ARTICLE IN THE EH ANTENNA SERIES --

THE EH DIPOLE ANTENNA WITH THE L+T AND STAR TYPE OF MATCHING & PHASING NETWORKS

(Originally published in the journal "Amateur Radio" in July 2004)

by Lloyd Butler VK5BR




(Figures redrawn for AR Journal by Bill Roper VK3BR)


INTRODUCTION


In my previous articles on the EH antenna, I have mainly concentrated on the antenna which uses a matching and phasing circuit network defined as the L+L network.

I will now discuss two other methods of matching and phasing used by the EH Antenna inventor, Ted Hart W5QJR. Ted has called these methods the L+T system and the Star system.

In the L+L antenna, I described how two E fields were formed, the primary field from a voltage developed differentially and a secondary field from a 90 degree phase shifted voltage developed longitudinally. (Displacement current from the secondary E field generates the H field in phase with the primary E field). In the L+L antenna the two dipole cylinders are fed by the balanced form of the L+L network. However in the L+T and Star antennas, the lower cylinder is connected directly to the 50 ohm coax shield and the system is unbalanced. Clearly, the idea of one field voltage developed differentially and the other longitudinally does not apply to these antennas and a different explanation must apply.



The L+T System


Figure 1 shows a typical L+T network with component values taken from the EH Antenna Calculator on the EH Internet site. The values were set for a 14.1 MHz EH dipole with 9pf of capacitance. The L+T name becomes apparent when L1 is considered as two separate inductors for the L and the T sections combined. The network is particularly applicable to an EH antenna which Ted Hart has called the Backpacker and which he and an associate have distributed in kit form throughout USA. The series antenna radiation resistance for this antenna has been defined by Ted as around 30 ohms.

Figure 1 - The L+T Matching & Phasing Network


To understand the network better, I have broken down the complete network into individual L sections which are coupled together. (See figure 2). To do this, L1 ad L2 are each split into two separate inductive reactances in series and C2 is split into two separate capacitive reactances in parallel. (The reactance values when combined closely fit the reactances derived from original form of figure 1.).

Figure 2 - The L+T Network broken down into separate L networks and last antenna Resonating Inductor


The first L network, C1-L1A forms a match from the 50 ohm input to 25 ohms at A. The second L network L1B-C2A forms a match from the 25 ohms to 200 ohms at B. The third L network C2B-L2A forms a match from the 200 ohms to 34.5 ohms at C.
(Reactance values are easily calulated from formulae in the VK3IY article, reference 5). If the source at the input looking back into the coax cable is 50 ohms resistance, then looking back from C into the network will be 34.5 ohms resistance.

Now we come to L2B. The reactance of L2B is simply made equal to the antenna capacitive reactance Xa so the antenna is then resonated and the output of the network at C sees a resistance equal to the sum of the radiation resistance and any loss resistance (mainly in the inductor).


The Secondary E Field


The next question is how do we get a 90 degree phase shift and how is a secondary E field applied. As the network at C looks as a resistance, we can substitute it for a resistive source feeding a series resonant circuit formed by L2B and Xa as in figure 3.

Figure 3
90 Degree Phase Shift developed across Series Resonating Inductor
& resultant potentials to develop two E Fields 90 Degrees Phase Separated


A characteristic of this circuit is that there is a 90 degree phase shift across the inductor. There is clearly a primary E field developed by the potential between the top cylinder and the general common reference of the bottom cylinder and coax shield. My theory is that there is also a 90 degree phase shifted secondary E field formed between the potential on the inductor around the vicinity of the C junction to the common reference. As in the L+L antenna, the displacement current from this secondary field generates the H field in phase with the primary E field.

In using this method of generating a 90 degree phase shift, an interesting characteristic is that it only works if the source is resistive. In figure 4 an inductive reactance has been added to the source. Observe that to bring the circuit to resonance, the inductive reactance of L2B must be less and the phase shift across it must be less than 90 degrees.

Figure 4
Effect of Reactance added to Source
Series Resonating Inductor now has less than 90 degrees across it.


I have tried operating the tuned antenna directly from the 50 ohm coax cable. It seemed to me that as the antenna resistance was about 30 ohms, the mis-match would not be that great. This actually worked but the interaction between antenna tuning and movement of the cable (as experienced on other EH antenna systems) was particularly high. I formed the opinion that the preceding elements of the L+T network were needed to provide a coupling buffer to reduce this effect. However, even with the buffer, reactance at the source can reflect through to cause some shift of phase in the series tuning inductor.

Concerning source reactance, one interesting bit of history concerns field tests carried out on a Backpacker EH antenna by Adam Macdonald N1GX. He found that when he eliminated the coax feed to the antenna and substituted a local battery powered oscillator at the antenna, the radiated field strength dropped dramatically. He (and others) concluded that this showed most of the radiation from the EH antenna was due to its feeder coax. However, Ted Hart strongly argued that the antenna was phase sensitive to the source and that the local oscillator had a reactive source impedance which upset its operation. The difference of opinion was never resolved except that results from my own tests, with out of balance current on the coax inhibited, did not support Adam's conclusion. (Refer to my previous articles in AR).



The Star Antenna


Ted Hart's more recent version of matching and phasing of the EH dipoles is illustrate in figure 5. He has called this the Star. Basically it works much the same as in figure 3 except that instead of injecting the source signal in series with the resonating inductor, the signal is coupled in by magnetic induction and the coupling to set the matching is controlled by the position of the input tap. As in figure 3, the primary E field across the dipole cylinders is 90 degrees phase shifted to that at the inductor input and there is a 90 degree phase shifted secondary E field between that point and reference common connected to the lower cylinder.

Figure 5
The Star EH Matching Matching System
showing 90 Degrees Phase Shift
and the two E Field Potentials developed


To get the thing working with the 90 degrees phase shift and low SWR reflected to the 50 ohms coax cable, there seems to be a need to add a small amount of inductive reactance in series with the input tap and a small coil is added as shown in figure 6. I haven't tried to evolve a theoretical explanation but I have confirmed the need for this on my own test antenna with SWR meter connected and monitoring of phase with the CRO across input and output of the network.

Figure 6
The Star antenna Matching with additional series Phase Correction Coil added.







Figure 7
An Experimental 20 metre EH Star Antenna

       

An Experimental 20 metre Star Matched Antenna


To carry out some tests on the Star matching idea, I assembled a rough model for the 20 metre band as shown in figure 7. More detail is given in the following text:

I put to use 600 mm length of 62 mm diameter PVC tube which I had on hand. Over this I fitted two metal sleeves 65 mm in diameter and 70 mm long to form the dipole cylinders. These were spaced 70 mm apart.

An open wound coil from the junk box was fitted at the bottom of the PVC tube, 130 mm down from the bottom of the lower cylinder. The coil, with a diameter of 50 mm, was wound with 22 turns of 18 SWG tinned copper wire space to a length of 80 mm. The spacing enabled it to have sufficient room between turns to tap the coil where required with crocodile clips. To obtain a good match, I found that 15 turns for the top tap and 4 turns for the lower input tap was close to the mark.

To locate the coil in place, a second PVC tube, 45 mm in diameter was fed as a tight fit through the centre of the coil and into the main PVC tube. The two PVC tubes were held together with the aid if three wood spacing pieces and self tapping screws
The coax input connector was located at the bottom of the smaller and lower tube.

The input series coil (a later addition not originally planned) was made up of 8 turns of 20 SWG tinned copper wire open wound to a length of 30 mm and with a diameter of 20 mm. The coil was left to float in space with one leg soldered to the 4 turn tap on the main coil. I found best matching was achieved with the other end tapped at 6 turns.

As an engineered example, my matching asembly would hardly take a prize but it enabled me to adjust the coil taps and carry out operational and performance tests as I needed to do.
















The Star preference


Ted Hart seems to be concentrating on the Star method of feeding the EH antenna as the now preferred method for amateur radio use. Both the L+L and the L+T models include two variable capacitors which are adjusted to tune up the antenna. In the Star method, resonance is with the self capacitance of the antenna and there are no tuning capacitors used. Ted has stated that he considers this less difficult for the radio amateur to get the antenna working. However it does seem to me that unless the design information is tied down to very rigid specifications, there is a need for some means to easily adjust the position of coil taps to tune the antenna right on the dot. Tying down adjustments to fixed positions is also made more difficult if the out of balance current in the coax is not inhibited with a trap and tuning is allowed to interact with the placement of the coax.

So there is a challenge here. What is the design of a coil with accessible taps which can be fitted around the supporting PVC tube and which can be easily assembled by the radio amateur with limited workshop facilities?



For those who may wish to purchase an assembled EH Antenna or a Kit


I asked Ted Hart the market situation for supply of ready made EH antennas or kits for the amateur bands as at the date of preparation of this article (December 2003). The following is information he has supplied:

He says George Jones KA4Q in USA continues to supply kits for the 20 metre Backpacker (L+T) antenna. George believes it is easier for a Ham to adjust the screws on the capacitors than to adjust the inductance of the *STAR* version. Although George is associated with EH Antenna Systems, he is doing this on the side from that company. However information on his antennas can be found at the EH Antenna Web site sponsored by the company (reference 6).

Steve Galastri IK5IIR of Arno Elettronica in Italy (Reference 7) now supplies Star type EH antennas for the 40, 80 and 160 metre band. With no tuning capacitors needed, they handle powers of 2Kw of SSB and 500 w of continuous power. However he still makes L+L type antennas for the higher frequency bands. A typical Star antenna for 40 metres is shown in figure 8. Observe the metal sleeve on the outside of the protective cover used to fine tune the adjacent coil inside.

A new company (yet to be named) led by JA3FR in Japan will introduce *STAR* versions for the 80, 40 and 20 meter bands very soon. The 80 meter version contains a small motor and tuning slug to allow coverage over the range of 3.5 to 4 MHz and it has the extended cylinders to allow high angle radiation. Ted says these antennas have very high efficiency and he is very impressed with the protoypes he has received.

Figure 8
40 Metre Star EH Antenna manufactured by Arno Elettronica
(Photo courtesy Of Julie Fubbri & Steve Galastri of Arno Elettronica)

Summary


The article follows on from my previous articles in Amateur Radio to include some theory on how the L+T and Star versions of the EH Antenna work.

I am sure that if you contemplate making one of these forms of antennas, or even purchasing one, you will ask which one should I choose. All of the antennas generate out of balance current on the coax feeder which, if you desire, can be stopped with a trap. Personally, I like the L+L version as it is the nearest arrangement to a balanced antenna. It goes against my grain to see one leg of a dipole directly connected to the shield of an unbalanced transmission line as in the L+T and Star versions. However the dipole legs are isolated if you use a trap as I have discussed in a previous article.

Clearly the inventor (Ted) sees an advantage in getting rid of tuning capacitors and the problems of capacitor breakdown when using high power. (This is particulary important to him for work he is doing on antennas for high power broadcasting). Hence his more recent concentration on the Star antenna development.

However from an amateur constructor's point of view, I rather favour the opinion of George KA4Q in that it is easier for the radio amateur to adjust some capacitor screws than taps on a coil. Furthermore, there is the construction problem of how to easily make the coil with adjustable taps.

References


1. Construction of EH Antennas for 20 and 40 metres - Lloyd Butler VK5BR - Amateur Radio. April 2003.

2. More Information on the EH Antenna & how it has performed - Lloyd Butler VK5BR - Amateur Radio, November 2003

3. How much Power is actually radiated from Longitudinal current in the EH Antenna? - Lloyd Butler VK5BR - Amateur Radio, May 2004

4. Other articles on the EH Antenna by Lloyd Butler VK5BR

5. Graham Thornton VK3IY - An L of a Network - Amateur Radio March, April & May 1995.

6. The EH Antenna Web Site (Sponsored by Ted Hart W5QJR)

7. Stefano (Steve) Galastri IK5IIR- Arno Elettronica - Web site

Back to HomePage