Elongated Terminated Loops
Terminated loops have only recently become popular, even though such systems were available commercially in the 1960's and earlier. Sometime in the early 1970's, I received an advertisement from the late W1BB. That advertising brochure described arrays of receiving loop antennas marketed by a Canadian manufacturer for commercial and military use. The brochure had pictures of diamond-shaped terminated loops, individual small loops, and loop arrays that looked like a number of Flag antennas strung end-to-end where each rectangular "loop cell" fed the next loop. The brochure gave directional patterns of the arrays, and the patterns looked much better than those of other antennas. Unfortunately the price, as I recall, was in the tens of thousand dollars for even the least expensive array! Although detailed technical information was withheld, it wasn't difficult to figure out the electrical design of each array.
That early brochure started me thinking in terms of antennas other than Beverages. I experimented with small loop antennas, and eventually installed an array of eight in-line loop cells. That system allowed me to be the first eastern USA station to work Japan and other Asian countries through interference from multiple multi-KW LORAN transmitters, and as the system was refined I could work JA's on a regular basis long before the broad LORAN pulse transmitters were removed from 1900 kHz. This is a testimony to the clean pattern and wide deep nulls possible using arrays other than traditional Beverage antennas.
Are they quiet because they are loops? Not at all!
One of the most damaging rumors to good antenna science are claims that loop antennas have a mystical property that somehow rejects noise. Nothing is further from the truth, with one exception. Corona.
Loops do not have sharp points sticking out into air at high impedance points, and that can be an advantage in limited cases.
Any sharp point extended from the area of anything, even wooden ship's masts in old sailing vessels, is more subject to corona in inclement weather. Sailors feared St. Elmo's Fire long before anyone knew what electricity was, and amateurs fear it today when trying to receive. We often just don't know what to call it, and can't see it in our brighter skies, so we call it "precipitation static".
The sharp point in a dipole where the wire or tubing ends not only promotes corona by virtue of the fact it is "out in space by itself", it almost always has the disadvantage of being at a high impedance point in the system. The tiny random charge movements with little current and very high voltage (high impedance source) inherent in conductors around the corona discharge are driving a high impedance part of the system, an ideal situation to maximize transfer of tiny amounts of noise power into the system!
That is almost certainly where the rumor that loops reject noise comes from. We forget it is a function of the sharpness and impedance in the area where the antenna is sharply shaped, and instead fixate on the fact it is "a quiet loop".
Are terminated loops actually loops?
Terminated loops do not behave like small conventional loop antennas. They do not carry uniform in-phase currents throughout the area of the antenna. They don't behave like directional couplers either, they are antennas...not nearfield coupling devices coupled to other conductors (we hope). Terminated loops are really just short verticals, with the phasing inherent in the long horizontal wires. The loop's vertical areas receive the desired signals, while the horizontal conductors merely serve to act as transmission/phasing lines for the vertical (or sloping vertical) ends. A small elongated terminated loop acts like a simple two-element vertical array with integrated phasing lines.
This is true even in the K9AY Loop and Pennant antennas, which have sloped conductors. The sloped conductor behaves as a vertical (think about this when people follow the mistaken advice that sloping the last few feet of a Beverage "stops noise pickup of vertically polarized signals"), ten feet of vertical drop is still ten feet of vertical conductor exposed to vertically polarized signals. The fact the antennas work so well is testimony to how sensitive a sloped wire is to vertical polarization. If sloped wires weren't sensitive to vertical signals, the Pennant and K9AY Loop wouldn't work!
Elongated terminated loops, and arrays of elongated terminated loops, are a special form of short phased verticals where phasing and feed systems are an integral part of the antenna. This integrated system of elements, feed system, and phasing solves construction problems associated with arrays using more recognizable vertical elements. With any system there are tradeoffs, elongated loop antennas only allow very limited control of phase and unwanted high-angle response. We can't obtain optimum patterns (although a properly placed series capacitance will help) because this system only allows limited control of current distribution and phase shift.
These antennas "want to" be verticals because the earth below them and somewhat closer spacing prevents the horizontal components from being efficiently coupled to space while the earth simultaneously enhanced the vertical radiation component of the gradually sloped conductors making up the antenna. The horizontal sections act as a wide-spaced air dielectric transmission line.
These antennas work best when propagation time of signals along the horizontal wires matches propagation time of the wave in space around the antenna, and when the earth or a ground system below the antenna is good. Suppression of high-angle horizontally polarized radiation, and maintaining velocity of propagation near unity, are why EWE antennas (and other forms of this family of antennas) thrive on good ground systems or good earth below the antenna. The desired earth effects are opposite those desired with Beverage antennas, an ideal situation for highly conductive soil!
Why use terminated loops?
It is easy to understand why "loop" systems, even very small loop systems, have become popular. Arrays of terminated loops vertical elements produce effective low-angle receiving performance along with a relatively clean pattern. EWE's, Flags, Pennants, and K9AY loops are effective methods of building two element vertical arrays. They are easy, small, and inexpensive! They are noticeably less directive than two phased verticals would be, because the horizontal components are not totally cancelled by ground effects and the opposing wire, but they certainly are easier to construct than phased verticals (with all their loading, grounds, and coaxial cables). It's all the standard old tradeoff we just can't seem to get away from, we always must balance complexity against performance.
Many people are working with the various arrays of elongated loop antennas, so there are few contributions I can make other than describing how or why they work. I would like to suggest it is possible to extend the arrays end-to-end for some distance without external feed systems, and well-placed reactances can be used to modify patterns. Very little work has been done in that area. I'd suggest experimenting with series capacitors, perhaps placed mid-way in phasing (the horizontal wires) areas to increase velocity of propagation through elongated loop arrays and increase directivity.
Are small loops hyper-sensitive to vertical masts?
There is no compelling evidence that any of these antennas are more sensitive to vertical metal masts than any other antenna would be. As a matter of fact, the only basis for such claims appears to come from models that fail to pass simple recommended tests for model accuracy and stability. If we build a model that is flawed and oversensitive to changes in things like the number of segments used in the model, we can expect it to be sensitive to nearly any change!
Everyone is free to say and do what they like, but other than keeping a short mast a few feet away from vertical wires and NOT connecting that mast directly to the antenna or feedline, I wouldn't hesitate for a second using metal supports. My large arrays of loops in the early 70's used metal masts, my arrays in the 80's did, and as have commercial arrays.
Feedpoint Matching
Great care must be taken in decoupling the feedline from the antenna in the balanced versions of these antennas, although the EWE (being unbalanced) is relatively immune to such problems. Keep in mind the antenna generally looks capacitive as a common-mode structure, so inductive decoupling (i.e. a choke coil of coax) can actually increase system problems. The best common-mode isolation system would be an isolated winding transformer designed with minimal capacitance between the antenna winding and the rest of the system. I use a small transformer with stacked 73 material binocular cores in feeding some of my high impedance "log-Beverage" arrays, and similar transformers should work with ~1000 ohm impedance elongated loop antennas.
Because this transformer only has a single turn primary (two turns with the balancing pass), I'm able to reduce stray capacitance to a dozen pF or less. It has excellent balance, low SWR over a wide bandwidth, and very low loss. The reasonably low transformer capacitance, when used in concert with proper feedline grounding and routing, should make the system relatively immune to common mode problems. I'd route the feedline horizontally directly away from the end of the antenna for a few dozen feet (but never a distance approaching 1/10th wavelength or longer), and then drop the feedline to ground earthing the feedline shield at that point. Decoupling beads or sleeves belong on the receiver side of the shield grounding point, not between the ground and the antenna!
I wind this transformer on three Fair Rite Products 2873000202 cores (about 1/2 inch square and 1/3 inch thick 73 material). The high impedance (secondary) winding is #26 enameled wire through Teflon tubing, while the primary is Teflon coated wire-wrap wire wound outside the tubing. The small extra pass that "dangles" on the low-Z primary winding helps balance the system, even though it adds a few pF of capacitance.
By the way, a Faraday shield will only make things worse. It will increase unwanted stray capacitance and might deteriorate the high impedance winding's balance if it is not properly grounded. The proper grounding point for a Faraday shielded primary is opposite the exit point of the primary winding, or on the secondary winding's exit side of the transformer. Most Faraday shields described for Beverage and other transformers are not only useless, they are often incorrectly grounded and actually increase unwanted coupling!
Eznec file courtesy of Roy Lewallen.