Gradient bending refers to any situation where an electromagnetic beam passes from an area of one dielectric constant into another area of slightly different dielectric constant. If you have studied optics you should know this is how glass lenses work. If you have ever stuck a straight stick part way into a body of water, and the stick appeared bent at the point at the surface of the water, then you have seen this effect.
A vertical interface: A horizontal gradient
The same thing occurs at radio frequencies. Here is a classic example. Consider a long straight beach at the ocean shore. The air over the water is cooler and wetter than the air over the land, therefore it has a higher dielectric constant. A radio signal coming ashore will pass from an area of more dense air into an area of less dense air. Assume the signal comes ashore at an angle other than perpendicular to the shore The signal path will bend (to one side) at the shoreline, making the angle even farther from the perpendicular to the shore. Thus if you are taking a bearing from the land to the sea, there may be a slight error. This is exactly the same as light emerging from a glass lens.
A horizontal interface: A vertical gradient
One real-world example: This writer was training the owner of a new SuperDF to use it to take bearings on ships in distress at sea (from his home on a bluff above the shore). After DFing several ships, we tuned in one who was transmitting a long "shopping list" for the ship's supplies. We took the bearing, and then stopped to talk. A few minutes later we noticed that the ship was no longer "On Bearing." It had moved 5 degrees! The DFer said that was impossible, as he knew that it was a research vessel which was permanently anchored on station, and not moving. "So what happened?" he asked. I asked him to spot the ship on his map, and he did so. I then asked for the the elevation of his DF antenna above sea level. I then calculated the visual horizon at sea for his antenna. This showed that even accounting for the probable height of the ship's antenna, that the two antennas where not line of sight. I then explained.
The air above the ocean is most dense near the surface, where it is cooled by the water and carries a lot of water vapor. Radio wave travel more slowly through this dense air than through the "thinner" air higher up. This causes a bending of the radio signal slightly downward. Thus a signal transmitted from over the horizon will bend enough to be heard loud and clear. But why did the bearing shift? The answer is that these condition of air density are not constant, and vary from time to time and from place to place. Wind shifts can make a big difference. The bending can even have a side ward component, causing the apparent bearing to shift (a small amount).
Another example is sometimes referred to as tropospheric bending or temperature layer bending. Here the radio signal encounters a near horizontal layer in the atmosphere where the air density changes rapidly with altitude. If the change is from dense to less dense at increasing altitude, then the radio signal can be bent, and come back down, hitting the ground a long distance away. The writer has seen this. A signal was heard loud and clear in the San Gabriel Valley from a station in Baja California, Mexico. At the same time, a repeater on top of Mt. Palomar (6000 feet) could not hear the signal (on its input frequency). Mt. Palomar is quite a bit closer to the station than the author. What was happening was the signal was going up, hitting one of these layers, and bending down so that I could hear it. Because the layer was below the repeater, the signal never reached the repeater. Later, the layer either lifted or dissipated, and the signal was heard through the repeater, and I no longer heard it directly.
There are specific things to look for to tell if you are seeing this kind of propagation. These layers are not stable for long periods of time, nor are they perfectly flat. They move up and down, produce wiggles and bubbles in their surface, and they tilt at different angles. As a result, there can actually be several paths from the station bouncing off this layer and down to your antenna. As a result, there is a multipath condition at your antenna, and as the various individual paths change, the signal will change signal strength (from reinforcement or cancellation). This is usually a very rhythmic fluctuation of the S meter, with the time for a cycle ranging from several minutes down to a fast flutter. I saw it go from full quieting to gone in a few minutes.
Thus, if you have an interfering signal, and it behaves in this way, then you know you will be in for a long drive, should you choose to find the source! Do you really want to burn 120 miles in each direction, with the prospect that the signal will go off the air before you get there? Instead, telephone other hunters in areas where it might be coming from (on that bearing). I know of one incident where this was the case, and the hunter who was called made the find (which later resulted in an FCC bust!)
We have already discussed the dielectric bending which produces high ground wave loss. This same bending can be produced locally by dense bushes or trees. A radio signal can pass through areas devoid of bushes while nearby passing though dense vegetation, resulting in differing velocity of the radio signal. This bends the wave front. If you are DFing from an area where this bend is present it can result in a bearing error. The bending is from the path without vegetation toward the path with the vegetation. This tells us that we should seek out location to take bearing from that are uniform in environment. The best place is at the edge of a bald hill, facing the direction the signal is coming from.
In a multipath situation RF energy is traveling from the source to your RDF antenna by more than one path. One path may be a direct path, straight from the radiating antenna to your RDF antenna. It is also possible that there is no direct path at all; all energy is reaching the RDF antenna by two or more scattered (or reflected) paths. At each particular exact location of the receiving antenna, these multiple paths will combine, obeying laws of superposition of sine waves. The net combined wave can be a strong signal where all paths combine in phase, or a very weak signal when the net combination results in cancellation. Moving the antenna only a few inches or feet can result in a very large change in signal strength. This is the mechanism that causes a signal to flutter in strength or momentarily disappear while you are mobile in motion.
The objects producing the multipath can be at any distance from you. They can be a few feet away, many tens of miles away, or anywhere in between.
In addition to the amplitude variation that multipath can cause, there is another effect which is important to phase sensitive systems like SuperDF. If you were to measure the relative RF phase over many nearby (fixed) points in space, you would find that the apparent phase front of the arriving signal takes on a wavy shape, rather than a straight line (for a distant transmitter). See the figure below for a two dimensional representation of this phase front distortion. The view point is from directly overhead, looking down at the ground. The numbers 0 and 180 refer to the phase angle. This is a "freeze-frame" view. The wave fronts are really traveling at the speed of light in the directions of the colored arrows.
Display the Figure . Down load
MultPath.GIF
image.
Two wave fronts coming from different directions combine to
show a sine wave like composite wave front.
"A" in the figure shows a bearing that would be produced by a SuperDF moving at a reasonable road speed. It doesn't matter in what direction it is moving. Note that it is the bearing of the stronger path. Bearings B, C, and D indicate three different bearings that would be obtained at their respective locations taken by a stationary SuperDF.
George Russ Andrews
President
George R. Andrews (Russ, K6BMG)
BMG Engineering, Inc.
9935 Garibaldi Avenue
Temple City, CA
91780, USA
Voice 1(626)285-6963
Fax 1(626)285-1684 (24 hour automatic)
America OnLine: Grandrews
Web: http://members.aol.com/bmgenginc
(7 Feb 1996)
Send E-mail to grandrews@aol.com.
(A message window will open.)
Return to This article,
Contents.
Go to Next Page of this
article.
Go to Prior Page of this
article.
Return to TOP PAGE.
