Anyone attempting to work DX on Top-Band (160 Meters) soon learns of the need for a good receiving antenna. John ON4UN in his Top Band Dxing book along with Greg Ordy and L.L. Cebek on their web sites pretty thoroughly describe most of the receiving antennas in use at this time. These references are MUST reading if you are serious about improving your receiving antennas. They describe the variety of factors influencing the decision on which antenna to build for a variety of circumstances. There are literally reams and reams of information on Beverage antennas, phased arrays, and loops. I too have evaluated many different antennas both on paper and in the field. W8JI has developed RDF (relative directivity factor) and ON4UN uses both RDF and DMF (directivity merit figure) as a measure of the usefulness of a particular antenna. Both these figures are a measure of how much wanted signal in the desired direction one might expect compared to unwanted signals and noise received off the sides or sides and back of an antenna. To calculate RDF a three-dimensional analyses of the antennas average response in Decibels ratio to a common reference such as an isotropic radiator is done. The antennas peak response for the desired receiving direction is calculated also in Decibels ratio to the same reference. The average response is then subtracted from the peak response and the result is expressed as Relative Directivity Factor or RDF. This result is also a ratio as subtracting two logarithmic numbers actually is a numeric division. The resultant RDF then carries a value in Decibels. To calculate MDF an average value of the three-dimensional antenna response in db with a common reference is calculated for the back half or rear 180 degrees of the antenna only. The average rear half response in db is then subtracted from the peak response in db (same common reference) in the desired direction and the result is then the Directivity Merit Factor or DMF in db. The greater either RDF or MDF numbers, the greater potential signal to noise ratio one might expect. These factors become very useful in comparing antennas for low noise reception. Using RDF for antenna selection is useful mainly when antenna noises arrive from all directions equally. Using MDF for antenna selection is oriented more for conditions where front to back ratio is important for reducing excessive noise or signals at the rear of the antenna. Many thanks are owed to Roy, W7EL for making RDF easy to calculate with his EZNEC antenna analysis program. There are many charts on the Internet as well as Table 1 here comparing different antennas by their RDF among other things. Suffice to say that loops, K9AY loops, Flags and less than 500-foot long Beverage antennas fall below an RDF of 10 db. A 1000-foot Beverage, a 70-foot by 320-foot Broadside array, and a phased pair of 1000-foot Beverages fall between a factor of 12db and 15db. It is often said that one can hardly detect the changes in a signal when it changes less than 2db (27% power change). My tests also indicate it is often hard to tell the difference when comparing antennas with an RDF difference of 2db or less. My research also indicates that one can change very little the RDF of any of these antennas with termination resistance, ground, or geometry changes. Yes you can change front to back and to some degree elevation angles but not much in received signal to noise ratio. And so it is that any antenna that shows a potential for operating with an RDF at least 2 db greater than what one is presently using deserves attention. Unfortunately for some people, as the RDF goes up the square footage of area used goes up in square law. This large area used by better antennas makes the below 10db antennas quite common with nearly always a desire to improve them. The 8-element array I have developed and described here produces an RDF of 13.5 db with a 30 plus db front to back ratio. This is a very significant advance for an antenna that only occupies a 200-foot diameter circle. In addition it is remote switch-able in 45-degree directional increments, covering the entire compass. This antenna generally requires no radial field and works with a high impedance buffer amplifier at each antenna.
This antenna is definitely not for the faint of heart to attempt. First of all there is no technique available I am aware of to verify any performance claims I might have uttered along the way of my progress (short of flying a signal source around the antenna). I have built several 1000-foot Beverage antennas that operate concurrently with the 8-element array. I can only compare the 8-element array to their performance levels. Secondly, it is likely to say an understatement that accurately maintaining signal phase relationships throughout the signal paths of every circuit in this design is the key to its performance. There are many pitfalls to complex phased arrays such as this one and making simplified assumptions has lead to a few non-working trials. I do not believe an array of this type has ever been constructed for the amateur radio fraternity and I would like to know if there has been. (k7tjr at arrl.net). I have been using this array for nearly 8 months (today=June 1, 2007) now in its best configuration so far and I am constantly amazed at the little nuances it has. Quickly selecting directions in 45-degree steps has shown rapidly changing signal arrival directions. It has some really deep notches off the side where you can place an offending signal at times to eliminate it. Having a northern direction available which I had ignored in the past has shown that signals can and do arrive here in a more northerly direction right over the North pole, sometimes when you would least expect it. Most importantly, it has over 13db of RDF, which ranks right up there, with the best receiving antennas published. All this in mind I am convinced that it indeed works as Eznec predicts it should. It is also possible to reduce the side lobes or “clean up” the pattern of this antenna by juggling the amplitudes and phases as it was optimized for RDF alone and not any other property. This antenna hears equally as well, and sometimes better than the 1000 ft Beverages on 3000 mile plus signals. Sometimes on closer signals the Beverages seem to have a higher angle and perform slightly better. All claims anecdotal of course but based on lots of hours in the chair with a Beverage in one ear and the array in the other. I am aware that there are easier ways to do the switch decoding such as using a PIC or Basic Stamp processor however not having the time to learn those I chose to use analog methods. I also learned that it would have been possible to eliminate half of my diode-switching array by including a simple reversing relay. There are a lot of parts in this array making it quite complicated to construct without the availability of circuit boards. I have however made a circuit board design for my Hi-Z amplifier as it can also be used in many other configurations of RX arrays and even on some of the higher bands. See WWW.hizantennas.com for supplies.The signal paths of the controller are laid out so that the approximate length of each antenna signal path is the same at least not glaringly different. While researching the needed excess phase compensation I realized that the connecting coaxes exhibit nearly 1/10th degree phase lag per inch at 160 meters. All the phase errors add up. Although it might be possible to build this array without an accurate way to measure phase shifts at 160 meters I would not want to try it.
I do not live in an area fraught with high-level broadcast transmitters so I do not know how my circuitry might handle that. What BCB interference I did pick up I have eliminated and have attempted a high level of third order intercept for all circuits. Adding filters before the combining circuits is surely a phase control disaster in my opinion. The signal levels in this array are quite low and even though there are built in amplifiers, there may be a need for a few extra db of preamp. Despite the low signal levels I am confident it is producing the Eznec calculated 13.5db RDF. Although it is difficult to find a stable signal that is directly off the back of the antenna I have measured in excess of 30db front to back ratio. It hears far better than other simpler antennas such as short Beverages, loops, and etc at this location. 13.5db RDF by all measures is no insignificant antenna on 160 meters. I could list many DX countries I have heard with this antenna but that would mean nothing as only anecdotal evidence of its performance. I really look forward to the future and hope that there will be improvements in this technique. It is possible to shrink the 200-foot diameter to as little as 150-foot with some degradation of RDF performance. Increasing the diameter seems to add little to RDF as well. I have looked at this antennas sensitivity to phase and amplitude changes in its current design. Without a thorough Monte Carlo analysis of all components I found that phases within a couple degrees and amplitudes within a couple percent had less than 1db of RDF change. I have worked on this antenna off and on for over five years now and I think it is time for some fresh ideas. Rudy N6LF has suggested it may be possible for even more performance with this array by using one of the antenna analysis programs that has an optimization routine. I agree that sounds exciting but personally after 6 years working with this design it is time to go put up the rest of my antennas for other bands.