 |
 |
| K9AY Azimuth Plot (50 degrees) | K9AY Elevation Plot (0 degrees) |
Greg Ordy
Based upon my own experiences, the primary challenge to conquer in order to consistently work DX on the low bands is to improve the quality of my receiving antennas. This is not to say that other aspects of my station and my skills are unimportant. In the end, everything is important, to some degree or another, and from one time to another. But it seems as if I'm never content with my receiving antennas. Perhaps they just interest me more than other aspects of the station.
Before jumping into receiving antennas, which will be the subject of many of my web pages, I might as well throw out my 2 cents worth on some of the other related subjects. As I have said before, please consult the ON4UN book, Low-Band Dxing, for additional information on these and many more topics.
A good radio is helpful, but it is probably the most overrated aspect of the (low-band) station. While there are folks who wage war believing that their radio is the best, and all others are garbage, I believe that most contacts can be made with most radios. Operator skill and knowing how to get the most out of your radio are often overlooked, as if the radio is going to do all of the work itself. Whatever radio you use, take the time to get the most out of it. Please understand that I'm guilty of failing this challenge too. I kick myself when I figure out some aspect of my radio that I had previously glossed over. In the extreme, we become appliance operators. Of course we all have to balance and prioritize the activities in our lives, and there are times when it is indeed a better course to spend more money than to spend more time. As I get older myself, it seems as if I have far more money than time. I digress.
Here's a little test to take in that regard. Consider all of the knobs and adjustments available on your radio. How many do you turn/change in normal operation? I get the sense that the average operator adjusts very few of the controls on their radio. Again, the radio is expected to do the work. Especially as radios get more complex, one needs to really understand how to optimize the performance of the radio for a certain circumstance. I have thought about making up a hint sheet so that I remember to fiddle with all of the applicable knobs for a given situation. For example, when trying to dig out a weak signal with the receiver, do I consider the preamp, attenuator, RF gain, AF gain, filter width, IF shift, noise blanker, noise reduction, RIT, AGC, auto/manual notch, and anything else that I might have available?
Low-band operation does require a selective radio, which usually implies accessory filters. CW operation, which is essential in order to work the greatest amount of DX on the low bands, needs CW-specific filters. Trying to work CW through an SSB filter on 160 meters is just too tough, unless the conditions are very exceptional, and nobody else is on their radio. I have a personal preference for ICOM radios, and with the 756PRO I tend to set my IF-DSP CW filter widths to 1200 Hz, 350 Hz,. and 50 Hz. That 50 Hz filter really improves the signal to noise ratio, but must be balanced with wider settings for finding the action. The point is, you may need several CW filter widths, not just one.
Fortunately many of the flagship radios of recent years are now available on the used market for around $1500 (USD). They will work well into the future.
An amplifier is a necessity on the lower bands. There are few exceptions. If your callsign begins with P5 (North Korea) or A5 (Bhutan), perhaps you can get by with 100 watts, but that's another story. Especially on 160 meters, running power is necessary in order to (consistently) work DX.
I have separated transmitting antennas from receiving antennas, which I believe is a common practice on the lower bands. The antennas tend to be separate because their objectives are not the same. The objective of transmitting antennas is usually efficiency and gain. On the receiving side, signal to noise ratio is key, as opposed to gain. I'll say much more about this in the upcoming section on receiving antennas.
If I had to make transmitting antennas as simple as possible, my advice would be: put up a full-size 1/4 wavelength vertical with 60, 1/4 wavelength ground-mounted radials. The classic dipole, unless elevated to 1/2 wavelength in height, just doesn't cut it - in my opinion. Now is probably a good time to repeat one of those little sayings which I find to be true. It is, you can never have too many antennas. There will be a time and circumstance when one antenna will work better than the others you have. That relationship will change, on occasion, from minute to minute. So, if you want to put up a dipole, there will be times when it will outperform whatever other antennas you have. Still, in my opinion, I believe that an efficient vertical is a good all-around DX transmitting antenna.
I don't want to make it appear as if verticals are a simple topic. Putting up an efficient vertical, especially on 160 meters, can be a challenge. There are many issues, opinions, and disputes out there concerning verticals. Some of the sources I frequently consult are listed on my resource page.
Operator skill is especially important on the lower bands. One aspect has already been mentioned - getting the most out of your station by understanding how to operate all of the equipment effectively.
A second aspect is being aware of propagation so that you know when it's time to work a certain part of the world. The term gray-line will become very familiar. Gray-line refers to exploiting the propagation enhancement that often occurs around sunrise and sunset. Ideally, both ends of the contact are in sunrise or sunset conditions. In general, low-band DX contacts are not possible when either station is in full daylight. So, part of the challenge becomes finding times of mutual darkness. In the extreme, this may only be a few days or weeks of the year. For example, one of my favorite paths is to work JA's (Japan), long path, at my sunset, and their sunrise. I direct my antennas to the southeast (from Ohio) in order to work Japan. It's no end of fun. This is only possible for a few weeks of the year, right around January. At other times of the year, one of the two sides of the contact will be in daylight.
Again, the ON4UN book covers gray-line and propagation in detail. There are many other sources available on the Internet.
Given the use of separate transmit and receive antennas, and an amplifier, the issue of station balance becomes important. By balance, I mean the comparative ability to hear the stations that can hear you. The extremes should be avoided. They are no fun. If anything, err on the side of being able to hear more stations than can hear you. Can far more stations appear to easily hear you while you have a hard time hearing them? If so, it's time to improve your receiving antenna(s). If you can hear lots and lots of stations, but they never return your calls, perhaps you need an amplifier or a better transmitting antenna. This issue is usually not as important on the higher bands, where station balance is easier to achieve because of using the same antenna for transmitting and receiving..
When I listen to my receiver, I hear three different categories of signals. One category, the smallest if I'm chasing DX, are the stations I want to hear because I want to work them. The second category of signals are other stations that I'm really not that interested in hearing, and I certainly don't want them interfering with my ability to hear the DX. The third category of signals I hear are noise signals. The noise signals can be either natural phenomena, or man made. They may be at a specific frequency and location, or they may arrive from literally all directions and all frequencies, as if they were hundreds or thousands of random transmitters - all trying to cover up the one signal you are trying to hear.
The noise issue is not as important on frequencies above 14 MHz. Even a wide-open band can be relatively quiet. On the lower bands, however, noise usually goes hand in hand with DX. During the day, the D-layer of the atmosphere, charged up by solar energy, tends to absorb all radio signals at the desired frequencies. No noise, but also no propagation. As sunset approaches, DX may become possible, especially at low wave angles, but as the sun sets, the noise machine starts.
Over time, I have come to view noise as the real problem, the real enemy, the real challenge. Noise on the lower bands acts as if it were hundreds or thousands of interfering and undesirable stations whose signals arrive at my antenna from nearly all directions - at once.
At the least, listening to noise is simply hard on the ears. You get numb in the head, and start to drool. Well, perhaps it's not that bad, but it's no fun.
So how does one separate the handful of signals they want to receive from the many signals that are not desired, and the noise which may be arriving from all directions?
I have come to believe that the best answer to this question is to use a highly directional receiving antenna. If this were 20 meters (14 MHz), I might think of a full-sized 5-element Yagi, located 80 feet off of the ground. When scaled to 160 meters, however, you would have a 300 foot boom with 230 foot elements, located 400 feet up in the air. I'm going to call that excessive. So, the brute force approach won't work. Fortunately there are many alternatives.
A directional antenna is characterized by it's response pattern. Ideally, a low-band receiving antenna would have a pattern that resembles a flashlight beam. All of the signal, the energy, the light, would be directed in a narrow beam towards the destination. UHF antennas such as dish antennas can approximate the flashlight beam. But, down at the short-wave frequencies, practical antennas are not large enough to act like a flashlight, and the antennas are too close to the earth, which interacts with the antenna, changing the pattern.
Historically, the most important aspects of an antenna pattern have been summarized by the gain of the main lobe, and the depth of the nulls off of the back or side of the main lobe. This will always be an incomplete description of the antenna, since the complete description requires specifying the response at a nearly infinite number of azimuth and elevation angle permutations. Until the relatively recent introduction of antenna modeling software, the complete character of an antenna pattern was very difficult to obtain, and required a large number of tedious computations. As a result, only a relatively small number of basic antenna systems were detailed. Fortunately, antenna modeling software is now available from a number of sources, at a relatively inexpensive price.
Using the EZNEC modeling package, here are two screen captures of the K9AY terminated loop antenna, unveiled in the September, 1997 issue of QST magazine.
|
|
|
| K9AY Azimuth Plot (50 degrees) | K9AY Elevation Plot (0 degrees) |
This antenna model was evaluated at 1.83 MHz, which is within the DX segment of the 160 meter amateur band. The azimuth and elevation plots represent slices through the 3-dimensional antenna pattern. Conceptually it is somewhat like looking at the image slices of the human body taken in a medical scanner. From these slices we can focus on certain details, and develop an overall sense of the body, oops, I mean antenna.
How do we use this information to help us evaluate the quality and applicability of the antenna? The answer to that question will depend upon the particular application. In this case, I'm going to focus on low-band receiving antennas. There are many other applications, and the nature of the evaluation is quite subjective and does change. These are also my opinions, and not very original at that. This is a blend of my experience coupled with information obtained from many sources. Please consult my resources page for a number of them.
The measurement of forward or maximum gain suggests how strong the desired signal will be - assuming that you can point the antenna in the correct azimuth direction, and that the signal arrives at the same elevation angle. In this case, the maximum gain appears to be approximately -24.33 dBi. What? Isn't that 24.33 dB weaker than an isotropic radiator? That's over 200 times less signal than an isotropic antenna. Isn't that unacceptable? As a transmitting antenna it most certainly is. But, for low-band receiving antennas, it is very acceptable. That turns out to be usually true because gain says nothing about the ability of a receiving antenna to reject signals from undesired directions. Referring back to my three categories of signals I mentioned at the start of this section, I would say that gain is important for the signal that I want to receive, but expresses nothing about rejecting specific signals I don't want to hear, and it says nothing about rejecting noise.
The maximum gain of an antenna will be pointed in some single direction. Is that a useful direction? A direction is expressed as an azimuth rotation and an elevation angle. Assuming that we can choose to point the antenna in any azimuth direction by how we build or rotate it, we are left to consider the elevation angle (at maximum gain), which is approximately 24 degrees in this example. That's the green line in the plots. Is 24 degrees a useful elevation, for DX? It's pretty good, perhaps just a little too high. A number of sources have investigated the typical arrival angle of DX signals from a set of locations around the world, on each amateur band. Painting with a very broad brush, DX elevation angles range between 30 degrees and 0 degrees, tending to become lower as frequency rises. Please remember that the elevation angle information is statistically significant, but there are always exceptions. The ARRL Antenna Book contains lots of good information on this topic.
In addition to specifying the direction of maximum gain, it is common to specify the beamwidth of the main antenna lobe. By convention, this is the angular wedge that is not more than 3 dB less than the maximum gain. In the previous example, the purple lines mark the -3 dB points. The text below the plot calls out the specific angles. In the azimuth plot, the 3 dB beamwidth is 160 degrees, and the elevation plot beamwidth is 62.1 degrees. This is conceptually a measure of how well the pattern is focused, smaller angles implying a tighter or narrower focus. This information helps us understand how accurately we need to point the antenna, if we want to listen to a given part of the world. Not to digress too much, but when we talk about pointing antennas, we need to consider the great circle route, that is, the direction we would fly an airplane to achieve the shortest distance between two points on a sphere (like the Earth). This is not the same as directions (bearing) implied on most common maps, such as those using the Mercator projection. From here in Ohio, the true direction to India is 20 degrees, London is 50 degrees, and Capetown is 110 degrees. The moral of the story is, use the correct map projection when pointing your antenna.
As you might imagine, we would like our low-band receiving antenna to have a very narrow beamwidth, pointed at a low elevation angle, under 30 degrees. That is a challenge, and most of my web pages on this topic will be directed towards approximating that goal. The penalty for achieving this goal is that you may need to have many antennas oriented in different directions, if you want to cover the globe. Because low-band antennas tend to be physically large, it is difficult if not impossible to rotate them. An alternative to building multiple antennas is to have a single antenna system which can be rotated electrically. This is most often done with symmetric arrays of vertical antennas, where we change the electrical relationship between fixed position antennas in order to change the antenna direction.
If we are going to talk about an antenna rejecting undesired signals, noise or otherwise, we have to consider the nulls in the pattern.
A popular metaphor for conceptualizing antenna patterns is that they are like an inflated balloon. If you want a bulge (or gain) in one direction, you achieve it by squeezing the balloon and creating constrictions or pinches in other directions. The volume of air (power) in the balloon (antenna) is remains constant, but we can obtain gain in one direction by giving it up in other directions. So, if we are trying to create a highly directional antenna, we will naturally find nulls as well as gain. Nulls will occur when an antenna system takes energy from one direction and either through cancellation or dissipation negates the energy without passing it on to the receiver.
If we look at the example antenna, it's clear that there is a substantial null. The null is in the exact rearward direction, and elevated from the horizon at an angle of 50 degrees. That null is approximately 40 dB deep, which is the same as reducing the power by a factor of 10,000. Obviously any signal, or noise, that arrives into that hole will be substantially attenuated.
The most effective way to deal with an interfering signal is to point a null at it. Nulls, however, have several issues.
Nulls are usually very sharp. That is, the beamwidth of the null is small. In the previous example, my estimate of the 30 dB or more null beamwidth is 12 azimuth degrees by 12 elevation degrees. That's less than 1/2 percent of the total sky area. For this reason, nulls are used by folks participating in radio direction finding. It is far easier and more accurate to see the signal drop across the null as opposed to peak across the primary forward lobe (which is usually much broader). In our case, we want to hear the signal, not see it drop to nothing. The actual characteristics of a null are antenna-dependent.
Nulls can be difficult to achieve. The null can be the result of very precise control of multiple antenna element signal amplitude and phase. In the real world, small differences in antennas (size, orientation, materials) and feed systems (component tolerance) can make it impossible to realize a theoretical null.
Nulls may be frequency dependent. The very deepest null might only exist over a very narrow frequency range - perhaps 5 to 10 KHz. Beyond that range, the null still exists, but it might drop from a maximum of 30 or 40 dB down to 10 to 15 dB. Nulls from nonresonant antennas tend to hold up much better versus frequency changes.
Still, if you need to make an undesired signal disappear, there is nothing like a null. They are also a strong indication that you have successfully constructed you antenna system, and it is operating as designed. Nulls are sometimes described in terms of the antenna front to back (F/B) or front to side (F/S) ratio.
In addition to radio direction finding, nulls can be valuable if you have very specific interference that needs to be reduced. One example is receiving AM broadcast stations. Stations can be assigned the same frequency if they are relatively far apart. During the day, you will only hear the station in your local area. But at night, you might find more than one station on the same frequency. If you want to hear only one station, point a null at the other.
Victor Misek, W1WCR, in The Beverage Antenna Handbook goes into considerable detail into what he calls a steerable wave antenna. He has several designs that let you control the null angle. Adjustable systems are always attractive because you can vary them for a given situation.
Another important use for sharp nulls is in dealing with very local noise. The noise could be a defective streetlight down the road acting as a transmitter, or a poorly shielded computer that is putting out signals all over the spectrum. This sort of problem, and there are many of them, especially in urban areas, could generate a solid S9 signal level across an entire band.The offending device is probably close to your location, and close to the ground. Consider the small loop antenna, also sometimes known as the shielded loop or magnetic loop antenna. Here is a response plot at the extremely low elevation angle of 3 degrees.
|
| Loop Pattern at 3 Degrees Elevation |
At very low elevation angles, the loop has very deep broadside nulls. If you rotate the loop to knock out that local noise source, the noise level might drop back down to an acceptable level. It will appear to be nearly a miracle antenna. Some sources claim this happens because of special properties of the loop. I'm with those other folks that believe it is simply because the loop has a low angle null that's useful for dealing with local noise. [NOTE: I am greatly simplifying the the issues of an antenna in the near field versus the far field]
We have gain in the primary lobe to help us hear the desired station. We have nulls which we can use to knock down interference coming from specific directions. What's missing is an overall approach to noise, which is a special challenge because it can come from a large number of directions - literally the entire sky. No single null, or even several nulls, will make a significant dent in the overall noise problem. What I would like is a way to measure the noise response, a metric that could be computed for an antenna to help rank noise rejection performance.
In The Beverage Antenna Handbook, Victor Misek, W1WCR, develops a metric which he calls the Conic Front to Back Ratio (CFBR). He applies his metric to Beverage antennas, and determines that there are certain Cone of Silence Mode lengths where the CFBR is optimized. The term conic refers to the the 90 degree cone which he centers over the axis of the Beverage antenna. In other words, the side of the cone makes a 45 degree with respect to the horizontal antenna. One cone projects forwards, and the other cone projects rearward. CFBR is defined to be the ratio of maximum response in the forward cone to the maximum response in the rearward cone.
The third edition of ON4UN's Low-Band DXing, by ON4UN, John Devoldere, has an entire chapter on low-band receiving antenna, including design and implementation information. At the end of the chapter (section 26.2), he compares and ranks several receiving antennas using a metric which is the average F/B ratio in a rearward cone. The cone is approximated by a set of points, 56 by my count. This average F/B ratio over an area of the sky is good start towards approximating the overall noise response of a receiving antenna. He used modeling software to detemine the F/B ratio at the 56 points, but he had to manually add them and take the average. The labor intensive nature of this metric makes it difficult to easily compute. In addition, this metric does not cover the entire sky, although the cone he measures covers the rearward 70 degrees of elevation (10 degees to 80 degrees) and 120 degrees of azimuth (120 degrees to 240 degrees, where 180 degrees is directly opposite of the primary direction).
More recently, I have become interested in the work of Tom Rauch, W8JI. On his Receiving Antenna Design Basics web page, he describes a metric called the receiving directivity factor (RDF). This metric uses the average antenna gain, which is computed by EZNEC, in the Windows version, version 3.0. In fact, I upgraded from version 2.0 of EZNEC to 3.0 just to get this data [I have come to believe that the software: NEC-Win Plus also provides average gain information]. As the name implies, the average antenna gain is the sum of the gain in all of the computed directions divided by the number of directions. The RDF is equal to the maximum gain minus the average gain. As I update my various receiving antenna models, I have been noting the RDF, and that information will end up on a web page. RDF is nice in that it is easy to compute (so long as you have the software), and includes the entire sky. Of course noise is not necessarily uniformly distributed, so the inclusion of the entire sky is not necessarily the perfect solution, either. In the end, there will no perfect, one size fits all metric.
I'm hoping that the RDF will be a valuable metric to help rank different antennas, and different antenna configurations. So long as the antenna can be reasonably modeled, the RDF can be easily computed
The average antenna gain is displayed on the bottom of the main EZNEC window when you choose the 3D pattern plot. You must be in 3D pattern plot mode. For the original antenna on this page, here is a screen capture showing the average antenna gain.
 |
| Average Antenna Gain from EZNEC |
Let's calculate the RDF for this antenna. The formula is maximum gain minus average. RDF = -24.33 dB - (-31.89 dB) = 7.56 dB. This is quite an interesting metric, although it relies upon antenna modeling, which opens up yet more issues. Still, it's a good way to make progress towards understanding what makes a good receiving antenna.
Here is the 3D plot for the K9AY antenna. It's a little hard to see when presented as a static picture. In the program, you can rotate the pattern in real-time, and that makes the shape very easy to absorb.
 |
| 3D Antenna Pattern Plot |
I have tried to highlight the deep null. While it is dramatic, it is clear that it is really a small part of the overall pattern.
Without getting too far ahead of myself, let me note that I have seen the RDF range from approximately 5 to 15 dB. 5 dB is the RDF for a vertical, which is considered to be a poor receiving antenna. An antenna with a 15 dB RDF is truly exceptional. The RDF of a 1600 foot long (487 meters) Beverage on 1.83 MHz is slightly less than 15 dB. Practical receiving antennas tend to have a computed RDF of between 7 and 12 dB.
That begs the question, does a dB matter? If you read another page of mine, you know that I believe the answer is yes. And when we improve the RDF by a mere 1 dB, what we are saying is that we gain (on the average) 1 dB additional signal strength separation between the main direction of the antenna, and all of the other directions in the sky.
There are times of the year when the atmospheric noise is at a minimum. These are often during the dead of winter. It may also be true that noise additionally declines during the years at the bottom of the 11-year sunspot cycle. At these times, a much wider range of antennas may provide good results. In fact, part of the fun and challenge of the low-bands is that while there may be trends, there are no rules. If you are having trouble making a contact, and you have any alternative antennas, they are worth a try, even if they don't make sense. On occasion, a low dipole will beat a vertical (for DX; for local contacts, the dipole may always beat the vertical). Signals may arrive at an azimuth angle which is not the great circle direction. This is a so-called skewed path. Strange things can and do happen. That's part of the fun, and even part of the charm.
Remember, you can never have too many antennas.
When considering a new receiving antenna, here are the issues that I tend to ponder.
What is the RDF, and how does it compare to other receiving antennas, especially whatever I have outside at the moment?
Are there any deep nulls in the pattern, and how could I exploit those to minimize local noise or undesirable signals from a known direction?
What is the elevation angle of the main lobe? Is it low enough to be useful for DX?
What is the overall response at higher elevation angles? If there is too much high angle response, local noise tends to rise.
If the RDF is not substantially higher than my existing antennas, does the proposed antenna have higher gain? When the noise level drops, gain does matter.
If this antenna turns out the be a good performer, how many more will I need to build to cover the globe, or, what do I need to do to make the antenna steerable?
Can I fit the antenna on my property? Perhaps I should consider this a little higher on the list, but one can dream, can't they?
Back to my Experimentation Page
