Small Magnetic Antennas
Being an antenna “affectionado” I have been reading antenneX almost from day one. The recent protracted series of articles on the CFA antennas in particular caught my attention. Whether the theory behind these antennas is flawed or not is way above my poor head but, for sure, there is room to experiment and to try out new approaches.
For the sake of clarity, I am not a licensed radio amateur but an active SWL and antenna experimenter for many years. One of my first conclusions was that the standard design for the CFA antenna is difficult and cumbersome. Moreover, the phasing network is substantial and if ever you get the thing to work you still are faced with the problem of having to get the whole thing on your roof, making it waterproof etc.-and all that for questionable monoband performance unless you can come up with a remote control unit. This to me was not an attractive idea at all.
The Search Starts
However, there are more ways to skin a cat. After reading all the patents and publications from Maurice Hately and others, I got more interested in CFAs of the type which are easier to build, and if successful, are easier to get up on the roof or in the clear.
The first project I embarked upon was the three-wire CFA which is a CFA with the tuning unit near the radio and a three wire conductor going out to the roof. The results I obtained were quite positive, but I concluded that the total length of the three-wire conductor required was to close to a quarter wave length to differentiate it from a standard quarter wave radiator tuned against ground or the chassis of the radio.
My next experiment was to use the same tuning /phasing unit and to feed a small loop through some 5 meters of coax. (click for measurement conversions) This was described somewhere in Sprat Magazine a number of years ago and I only duplicated the reported results as described. The experience however was very satisfying and by now I was badly bitten by the CFA antenna bug.
I managed to get copies of all existing CFA patents and studied all the related patents cited in these as well. One of the patents describing the DLA or double loop antenna caught my attention in particular. Same inventor, same theory and at first glance much more easy to construct and if it works, easier to get up on the roof. I followed the forum discussions on antenneX with increased interest and can only say that I sincerely regret that the flurry of discussions have come to an end. What a pity!!
Now the Experiments
My first attempts to make a DLA antenna failed miserably but I kept trying until I saw the light: could it be that the DLA is not a real CFA but a simple magnetic antenna in disguise??? I read about the commercial version of the DLA as sold by Mr. Hately and checked the dimensions carefully.
I took a piece of coax of the same length of about one meter and measured the capacitance between the braid and the inner conductor. I also measured the inductance formed when making a loop from the coax. What I found was that the inductance when using the inner conductor of the coax for forming a loop is greater then when using the outer braid as the loops conductor. Please read this last sentence again as it is vital for understanding the rest of my story. This obviously has to do with the “Velocity Factor” caused by the dielectric loading on the inner conductor.
For these experiments I used RG 8 coax which is a lot heavier and is almost self supporting for forming a loop.
Question of Feed
Meanwhile I had read the article on the quad phaser by Harold Allen, W4MMC and I had started to communicate with him (as suggested by antenneX Publisher Jack Stone) as well as with good “Old George” (George Sharp, KC5MU) as he fondly refers to himself. It has been a sheer pleasure to exchange e-mails with both gentlemen and this has contributed much to my experiments. Then, it struck me that if you look closely at Harold’s quad phaser and you start dreaming and you replace the multi-turn coils by two one turn coils that you get something very close to the DLA design of Mr. Hately.
Looking in the ARRL Handbook I also found some similarities between the DLA and the well-known coaxial trap for dipoles. In coaxial traps one uses the capacitance between the braid and the inner conductor to form a resonant circuit. A sleepless night was the result of this observation. I had seen the light!! The DLA might be a magnetic antenna in disguise where one uses the inner conductor as the loop and the braid as the built in capacitor to tune the loop to resonance (or the other way around).
Figure 3: The double loop antenna showing the capacitance between the two loops making it look like a kind of quad phaser
Figure 4: The well-known coaxial trap
And now back to the drawing board:
I took a piece of RG8 coax and connected it as per Figure 2 after measuring the capacitance and inductance as per Figure 1. In one of my antenna books I found a formula to calculate the capacitance to resonate a given inductance:
Formula for Capacitance required:
C = 25330 / F x F xL where F = frequency in megahertz C = capacitance in picofarad L = inductance in microhenries
The formula for the frequency:
F = square root of 25330/ L x C
With the data found I quickly calculated the resonance of the loop and I was very happy to see that the calculated resonance point was spot on the one I detected by hooking up my MFJ antenna analyzer. I also double checked with a Grid Dip meter and once more by using a small coupling loop inside the larger magnetic loop. By now I was getting exited by the idea of making such simple small loops. But feeding such a small loop in the conventional way was not what I had in mind so I decided to hook it up straight to the coax of my MFJ. But hooking up is easier said than done as there are several possible ways to do so. The first way is shown in Figure 5.
I was amazed to find a very low SWR even without connecting the braid of the feed line. The resonance point was spot on the calculated one so what is going on here??? I quickly realized that the small loop had some considerable stray capacitance to ground and that the feed line was part of the resonating circuit.
I put some ferrite clamps on the feed line and was able to confirm this conclusion. I then added a very small trimming capacitor of some 4 to 5 picofarad as per Figure 5 and I had my first working magnetic loop. Performance of this little marvel hanging 5 feet above ground on my terrace was very satisfying. I also noted the very small bandwidth and directivity which are normal for magnetic antennas.
I then decided to hook up the same loop in a different manner as per Figure 6. This time I connected the center of the feed line to the cross over point of my loop and I was surprised to find a totally different resonance point at a higher value. When I verified my calculations I had to conclude that this time I was using the braid of the coax loop as the inductor and the inner conductor of the loop as the resonating plate of the capacitor.
Lower inductance and same capacitance indeed should result in a higher resonance point. Again the calculation was spot on. However, without connecting the braid of the feed line the SWR was a lot higher than in the first case. It almost made me quit this approach until I concluded that the stray capacitance of this loop was less than in the first case. I added the tuning capacitor as per Figure 6 and again I was in business although the resonance point this time was way outside my preferred 20-0meter band. I had to use a larger, longer loop to find a resonance point in the 20-meter band. Again the performance of this small loop is much better than what you would expect for such a small thing.
Must be a Magnetic Loop
Reading more about magnetic loops and their efficiency I concluded that these were indeed very small loops so, for once, I was looking to make them a little bit bigger by looking at coax with a lower capacitance per meter. It is then that I decided to go for high quality heavy twin speaker wire. The wire I bought at Home Depot gave me a capacitance of 49 pf per meter, almost ideal to build somewhat larger self-resonating loops.
I took a piece of wire 3 meters long, put it inside a plastic tube also bought at Home Depot and started to play with it. I first connected the loop exactly as per the patent of Mr. Hately and found indeed that you can get a wide resonance point when using different capacitors-one at about 40 pf and the other at some 10 pf. The resonance point was 14,2 MHz with these values.
Needless to say I was delighted with my findings. However, the patent shows the use of a step up impedance transformer and all my experiments showed me that there is no need for such a thing. On the contrary my experiments also showed that the loop could well be resonated with only one capacitor but the loop was too big to resonate at 14,2 MHz, so I had to down scale my loop.
The final and preferred set up for this magnetic loop is shown in Figure 7 which is basically the same set up of Figure 6, but using twin speaker wire instead of coax. This time I got a nice bandwidth as well of some 400 kHz covering the entire 20 meter band and all of that with a little bit of speaker wire and just one home made capacitor (you can use a short piece of coax, a piece of twisted twin conductor wire or a chunk of double printed circuit board) for the 20 meter band
I do recommend using a length of about 1.36 meter of AWG # 10 or 12 wire enclosed in a thin plastic tube which can be bought at Home Depot, or similar type of store as well for less than two dollars USD. To calculate the theoretical resonance point you must measure the inductance of the two speaker wires connected at both ends in parallel as forming one loop (I found 1.32 microhenries versus 1.52 microhenries for only one of the two wires). Also, measure the capacitance between the two wires as well as the value of the trimming capacitor (some 10 pf to possibly 30 pf). The SWR, an impressive 1.1.
My Findings
Looking back now at the battlefield of some weeks of experimenting (and not cleaning up the kitchen table in the process causing some differences of opinion with the XYL) I can recommend that others should build these small antennas and duplicate my findings:
A very small-self resonating magnetic antenna for practically no money at all which can be put on the roof easily or can even be painted black and glued to the roof shingles of a house as a totally disguised antenna.
Very promising results on receive. It is amazing to receive stations from all over the world on such a small thing. The antenna also hears very well on the 40 and even more so on the 80-meter band although the SWR on these bands is off scale. I absolutely have to build one for these bands as it looks very promising indeed.
As there are virtually no contact points to cause ohmic losses, the efficiency should also be high on transmit as well. The only contact point in the loop can be soldered to avoid ohmic losses.
If you do not have an antenna analyzer you can still build this loop by calculating its dimensions and then fine tune it by listening to the receiver for maximum signal strength by adjusting the trimmer capacitor.
My Conclusions
The “Jef Drive” as Joel Hungerford, KB1EGI called it affectionately in one of his e-mails (for which I give him full credit) is probably the simplest and cheapest way to build and feed a performing magnetic antenna for the listener as well as for the radio amateur looking for antenna solutions. For less than $5 dollars USD you can be in business!
- The “Jef Drive” probably works like a capacity balun or voltage divider but I leave this up to the more experienced theoreticians.
- The total set up is very much like a DLA antenna but can do with only one capacitor and no step up impedance transformer.
- There is no need to claim “Vector Synthesis” or a CFA working mode. Life seems to be a lot simpler.
For those who want to experiment, start off with a piece of coax or speaker wire approximately one inch longer than the dimensions described and bring down the resonance point by pruning the wire length as well as adjusting the trimmer capacitor. It is easier to cut than to add wire!!
With regards to the speaker wire loop, there is also a second resonance point at some 36 MHz (SWR 1.1). If you want to use the second resonance point for 14,2 MHz and build a somewhat larger loop to have a bigger capture area you can do so by using a length of speaker wire of 3.7 meters
Again start off with a little bit longer wire and cut back until you reach the “sweet spot” of SWR of 1.1. This size antenna should be a real performer. There is also a way to calculate the second resonance point for any other frequency provided you have the possibility to measure small inductances. If you do, take a piece of speaker wire about two meters long and measure the inductance of a loop with both wires connected in parallel and then the inductance of a loop formed by just one of the two wires.
You should of course also measure the capacitance between both wires. The difference in inductance values found should be used as the L or inductance value in the formula mentioned above. For the C or capacitance value you have to use the capacitance value between the two wires plus the value of a trimming capacitor or about 20 pf . Make this calculation two or three times with different lengths of wire and you will be able to calculate the length required for a loop of a given frequency.
My next obvious step is to build loops for different bands and to try out an array consisting of two or more units. Having two or more single band antennas on one feeder line is also high on my wish list.
For all experimenters I do highly recommend the use of speaker wire inside the plastic tubing. It is easier to handle and a little bit bigger so performance should be up.
I really enjoyed experimenting and I am very grateful to Harold Allen, good “Old George”, Joel Hungerford and Jack Stone for supporting me in the process.
Originally posted on the AntennaX Online Magazine by Dr. Jef Verborgt
Last Updated : 28th May 2024