On The Art of NDB DXing
by Sheldon Remington
©
1987-2000 All Rights Reserved
|
CHAPTER FOUR: MAN-MADE NOISE
Part Two - Mechanically-Steered Nulling Once we have done everything practical in locating and eliminating man-made noise at its sources, we'll probably still end up with an unacceptably high noise floor. Then it becomes necessary to seek solutions on the receiving end. One approach is the use of time-based noise blankers; these will be discussed in Chapter Six. Our subject this time and next is the use of directional antenna systems to distinguish spatially the desired signals from the interfering noise. This topic should be of interest even to those few who don't reside in the midst of powerlines, and to those going on DXpeditions away from the noise. This is because antenna systems with deep, steerable nulls have other applications. Examples would include nulling GWEN, TV sweep interference (unfortunately, battery-operated TV's and computers are infiltrating the remote sites nowadays), or particularly bothersome nearby transmitters. The Directionality of Powerline Noise Probably most of us, including this writer, have assumed that our locations, surrounded as they are from as they are by myriad utility lines, would be bathed in powerline noise coming simultaneously from every direction. If this were true, directional antennas wouldn't be any help at all. Well, surprisingly enough, this turns out to be just a form of DXer's paranoia! In my own case, the misconception was bolstered by a certain shortcoming of the directional antenna (a loop) I had tried; in particular, that the directionality was steerable only in one plane. This antenna could lower the noise level by a few decibels when I rotated it, but I didn't try tilting it. BCB DXers, who are probably the primary users of loops, eventually discovered that a combination of tilting and rotating would give incredibly deep nulls on local transmitters. This is due to the actual tilt of the incoming vertically-polarized surface wave. Such a loop is called an "altazimuthal"-a combination of altitude and azimuth. Adept medium-wave DXers now include such a loop in their equipment arsenals. Finally, some of us experimenters discovered that this tilt was present in powerline noise, too. Not only is it a tilted wavefront, but it apparently comes from a point in 3-dimensional space. The maze of lines surrounding us emit radiation that combines to produce a single vector sum, and therefore, we can eliminate the noise by pointing a directional antenna's null toward that point in space. This point is quite stable, although not perfectly so, varying with disturbances in the antenna's proximity. These conclusions are based on my own experiences DXing in five widely-separated apartment locations, each infested with high noise levels. Note that this nulling capability needed in our antennas is different from the directional pattern which is sought in HF and VHF work. The latter are designed to produce a single lobe of "forward gain," while suppressing other directions to a moderate degree. First of all, such gain can be achieved only with antennas whose size is comparable to a wavelength, which would be impractically large on LF. Secondly, these configurations typically yield nulls only 15-25 dB deep, while we on LF are up against a noise floor perhaps 50 dB stronger than the DX. So, what we need is a system that can produce one or more really deep nulls which can be steered to any point in space. It doesn't hurt if the lobes are very broad, since that will just let in more DX. Fortunately, there are several distinct techniques to realize such an antenna system for LF, and most of them are simple to construct. The best-known types are the loop antennas. These come in two basic categories: the multi-turn tuned loop and the single-turn untuned or broadband loop. Tuned Loop The time-worn mainstay of MF DXers is the air-core multi-turn tuned loop, but it can also be wound on a ferrite core, allowing a reduction in size. Either way, the circuit usually looks like Figure 1. Note that the receiver isn't directly connected to the tuned loop winding. Instead, it is coupled through a single-turn link winding to bring the impedance down. The tuned circuit typically has a high Q (selectivity), so the capacitor must be repeaked when you change frequency. This has the advantage of rejecting off-frequency receiver overload problems, but it requires keeping the whole antenna close to the operating position, or else using remote varactor tuning, which is a whole 'nother can of worms. Sometimes the tuning can stay fixed, i.e., for single-frequency point-to-point reception.
Construction plans are widely available for both the air-core and the ferrite tuned loops. The best sources probably are the assorted booklets and reprint sheets offered by the BCB clubs. Send 50 cents for a National Radio Club catalog to NRC Publications Center, P.O. Box 164, Mannville, NY 13661; send $1.00 for an International Radio Club of America list of reprints and publications and their cost to Steve Ratzlaff, 295 Pettis Ave., Mountain View, CA 94041. One recommended excellent IRCA publication is A DXers Technical Guide which has a section on loop antennas. Please note that any loop plans shown in the above-mentioned BCB Club publications will be designed for operation from approximately 500-1700 kHz, i.e., the MF BCB frequency range. These loops will need at least 2-3 times the number of windings and/or amount of tuning capacitance to get them to tune the 160-535 kHz (the LowFER and NDB "bands"). Other sources include numerous back issues of the Lowdown, the LF-MF Scrapbook by Ken Cornell ($15 postpaid from 225 Baltimore Ave., Point Pleasant Beach, NJ 08742), and chapter 14 of the ARRL Antenna Book. LWCA member Jim Hagen has ferrite-loop expertise, and we would all be indebted to Jim if he would do a write-up for the Lowdown. A few firms, such as Palomar and Radio West, have offered ready-built or kit-form tuned loops in recent years. Reportedly all of these have suffered from mechanical problems, a high noise level, or both. I have no information on current availability of commercial tuned loops. Of course, such loops should be constructed so as to be tiltable as well as rotatable. Steve McDonald's loop on the cover of the February 1987 Lowdown is a good example of an air-core type, and Todd Roberts' equally good ferrite loop is pictured on Page 19 of the April 1987 Lowdown. Untuned Loops Recently, Dr. Ralph Burhans has been providing the Lowdown with extensive information about untuned loops. Besides his Radio-Electronics reprint, see April 1987, pp. 18, 31-32; March 1987, pg. 16; December 1985, pg. 19; and June 1985, pp. 8-12. LWCA members have contributed their particular realizations of Ralph's designs, such as James Borglum, March 1986, pg. 12, Ray Cole, January 1986, pg. 19; and the ubiquitous Mitchell Lee, May 1985, pp. 9-14. These loops all utilize a low-impedance preamplifier which effectively raises the single-turn loop's output to match or exceed that of a tuned loop. The lack of retuning is a boon to the busy DXer, and it also allows us to remotely locate the loop. The availability of p.c. boards and parts kits makes construction easy and inexpensive. I would like to share a couple of my own realizations of the tiltable untuned loop. Most builders have apparently felt it necessary to construct the loop from coaxial hardline. I have chosen instead to use garden-variety flexible coax cable which is not only easier to work with, but easier to find as well. In both cases, electrical performance was all I could desire. First is the "take-apart" loop I made for the Kauai DXpedition (Figure 2). Being an amateur photographer, I was already going to be bringing a sturdy tripod on the trip. Since tripods are easily adjusted altazimuthally, it seemed like an ideal temporary loop support.
The only problem I had with this loop was my own fault: I neglected to weatherproof the wood dowels, and they swelled up in the gentle Hawaiian rain. This made them exceedingly difficult to remove from the tubing. Since the loop has a small wind cross-section, I experienced no tripod instability. If such a problem did occur, it would be easy to borrow a trick from photographers, and suspend a rock or makeshift sandbag from the tripod's center post. The second design was an experiment with remotely-actuated steering without the use of motors (see Figure 3). I built it in 1984 to go atop an old Chevy for overnight DXing on a coastal hilltop in northern California. I wanted the loop to be large for sensitivity, but desired to steer it altazimuthally without leaving the operating position inside the car. I decided that the rotation could be accomplished by mounting the whole rooftop assembly on a length of conduit which penetrated through the center of the roof and dropped into a socket on the floor. Tilting was achieved via a halyard system running through the conduit (along with the feedline).
By the way, such a massive loop presents problems if you try to drive with it. The main problem, besides low hanging branches, was a gyroscopic tendency by which the loop wanted to stay pointing in the same direction, even on hairpin curves. And unless you want to see UFO reports in the next day's local paper, paint the thing dull black, don't install LED's in the preamp housing, drive it only at night, and avoid busy intersections! Even when stationary, there was considerable rotational torque exerted by the wind. Perhaps this configuration should be beefed up and used strictly at the home QTH. Of course, though my intention was the armstrong method, this would lend itself to motor control. For one thing, if you have a full 360 degrees of rotation, you will only need 90 degrees of tilting to cover all possible angles. A really hefty loop could be handled with the commercial rotor systems used for amateur satellite and moon-bounce operations. Other Movable Antennas In addition to the various loops, there are other antennas which physically steer their null. These are combination arrays in which the outputs from two antennas are electrically combined to produce a fixed pickup pattern. The whole assembly is then mechanically rotated and tilted as a unit. One such array is the cardioid loop-plus-whip system. A standard dual-null, dual-lobe loop is combined with an omnidirectional whip, often known as a "sense" antenna, to produce a single-null, single-lobe pattern. The null is sharp, and the lobe covers the rest of 3-dimensional space. This is popular in direction-finding equipment since it resolves the loop's bi-directional ambiguity. Dr. Burhans has experimented with such an array, and wrote some comments on Page 12 of the June 1985 Lowdown. The ARRL Antenna Book, Pages 14-4 and 14-5, describes a circuit using a tuned ferrite-core loop with sense antenna (Figure 4).
Another combination array is the twin-whip system. It should be simple to combine the outputs of two identical active whips and reverse the phase of one of them. The whole assembly would be broadband, sensitive, easy to steer, and very portable. Apparently, just such a system, called the Microdipole, has been built by Dr. Burhans and is undergoing torture testing at the hands of Mitch Lee. We will be eagerly awaiting further word. A final possibility for this category might be a scaled-down Adcock array. According to the Antenna Book, it produces deeper nulls on skywaves than does a loop. It is shown as 1/8-wavelength wide and 1/10-wavelength tall; these dimensions would have to be drastically reduced for it to be practical on LF. The description has no discussion of size variations, and I have no idea whether shrinkage would disturb the function. Perhaps one of our members could comment? Next time we will get into various phasing methods (Connelly, goniometer, etc.) that will electrically steer a null without physically moving the antennas. One final note: Duane W. Cook writes that the MFJ-955 preselector has
been discontinued, but they are building a better one, cheaper.
|
