Safety‎ > ‎

A Wing and a Prayer

posted 3 Sep 2013, 15:08 by Support SarMobile   [ updated 26 Sep 2013, 14:54 ]

A Wing

We haven't been able to determine exactly how the Wing Blanking technique came about, but it may have happened like this.

Many years ago, a pilot flying an ELT search, listening to the beacon signal on the airplane communications radio, executed a 360° circle but noticed at one point in the circle the ELT signal dipped momentarily in strength so that it could not be heard. It turned out that this dip occurred when the bottom of the airplane pointed towards the ELT location. The pilot thought perhaps this effect could be used on searches, but what could be causing it. Again we don't know for sure, but we suspect this happened with a fairly common high wing airplane, a Cessna 172, with the communications antenna mounted on top of the fuselage but aligned with the wings. Perhaps with the 'line-of-sight' propagation of ELT radio signals in mind the pilot realized that with the airplane banked for the turn, at one point the wing would block the line-of-sight from the beacon to the airplane antenna. It seems obvious that the wing must be blocking the signal. The wing is constructed from aluminium, which conducts electricity and can reflect radio signals, right? Unfortunately it isn't that straight forward.

In order to block, or reflect electromagnetic signals such as radio wave or light, an object must be made of the right material, and be large enough. We compare the size of the object to the wavelength of the signal. Objects that are much smaller than one wavelength will have essentially no effect on the signal. Objects much larger than one wavelength will reflect, absorb or pass the signal depending on the materials it is made of. Objects that are neither much smaller, nor much larger, may have some effect of the signal, again depending on what it is made of. So far we haven't assigned any specific values to these sizes, so let's start. In professional radio technology circles it is widely accepted that much smaller is around one tenth of a wavelength and much larger is ten wavelengths. So to reliably and effectively block the beacon signal the wing would have to be as large as, if not larger than ten wavelengths. The ELT beacon frequency is 121.5 MHz and has a wavelength of 2.47 meters. According to Wikipedia, a Cessna 172 has a wingspan of 11 meters and a chord (the distance of from the front of the wing to the back of the wing) of 1.5 meters. So can an object measuring 1.5 by 11 meters effectively block a radio signal with a wavelength of 2.47 meters? According to generally accepted principles no it could not. But we have an experiment that you may be able to try to convince yourself.

A consumer GPS receiver, or the GPS chip in a modern smartphone, receives the GPS data signals on 1575.42 MHz. Those signals would have a wavelength of 19 cm. We have access to a car that has a steel roof that is 1.7 meters by 2.3 meters. If an airplane wing that is 0.6 wavelengths by 4.4 wavelengths can block the ELT signal, the car roof that is 8.9 wavelengths by 12.1 wavelengths should be able to block th
e GPS signal. Most consumer GPS receivers have a display that shows the positions of all satellites in view, and the strength of the signal from each satellite. If you have a smartphone you may have to install an application. By taking your GPS receiver into a car and holding it under the roof you can measure the strength of the satellite signals entering the car. With good timing you can perform this experiment while a satellite is directly overhead so that the roof is between the satellite and the receiver. Sometimes you will notice that the signal from that satellite is week, or not detectable. But often enough you will find that you can receive a strong signal even under the roof which, according to wing blanking, should be able to block it. On the left you can see a picture of GPS reception in a car. You will notice that satellite 18 is nearly directly overhead, but still the signal is quite strong. You may click on the picture to see a larger version. If you are still skeptical, to the right is a similar image taken while inside a passenger jet completely surrounded by the aluminium fuselage. Signals are still able to get to the receiver from all directions.

We have had some questions about our signal strength displays above. Luckily we have signals analysts and programmers that enjoy working on these types of problems. Ordinarily one would place an antenna in the situation to be tested in an anechoic chamber. One would then either transmit from the antenna under test and measure the signal in all directions, or transmit from all directions and measure the signal at the antenna under test. Anechoic chambers large enough to hold airplanes are very rare and expensive. For large aircraft one would normally mount the aircraft on a pole in a very remote area without any reflectors for a long distance in all directions. This was done to test the stealth characteristics of the Horten Ho 229. This is also expensive and beyond our means. However there are systems that, over time, transmits a microwave signal from almost all directions that can be received by commercially available receiving equipment that will report signal strength, the Global Positioning System (GPS) and the Global Navigation Satellite System (GLONASS).  With some fairly simple software one can produce signal strength and standard deviation maps.

First we will start with a stationary test point. These two pictures are the signal strength (left) and standard deviation (right) of GPS and GLONASS signals recorded in a single story wood frame structure. Signal strength is from 0 to 45 color coded from red through yellow to green. Standard deviation is from 0 to 20 coded from green through yellow to red. Black represents segments directions from which no signals were received. Direction in azimuth is 360° around the circle, North at the top. Elevation is 0° at the circumference of the circle to 90° at the center. These maps are made up of 15,523,503 individual signal strength measurements. Even with so many samples there are gaps due to the parameters of the satellite orbits relative to the earth. North is located at the top. Clearly signal strength is weaker near the horizon, which is to be expected, but there is also an area in the SSE direction that is week up to about 40° elevation. We arranged to have something actually block the signal from that direction. These images clearly show what to expect if something is blocking or reducing signal strength.

These two images are taken from a car with a hard top, steel roof. Again these images represent signal strength and standard deviation. Because the car is mobile we have the opportunity to move it around to get signals arriving from almost all directions in fewer samples. These images represent 1,792,983 individual signal strength measurements. Because we know the orientation of the car we can compute the direction, relative to the front of the car, the signals are coming from. In these images the front of the car is up. We can see there is no clear signal blockers, even though according to the Wing Blanking theory, and the Strut Mount Direction Finding Theory the roof, doors and other metal parts of the car should be blocking the signal.

To understand why this happens you really need to understand Quantum Electrodynamics, but we can find a simpler analogy. This is all about wave propagation. While the physics of radio wave propagation are very different from ocean wave propagation the macro effects are similar. The main point of confusion seems to surround the behavior of visible light, so let's start there. The visible light spectrum, from violet through red, covers wavelengths from 390 to 700 nano meters. There are one billion nano meters (abbreviated nm) in a meter. For simplicity let's use 500 nm as the wavelength for visible light. That means the wavelength of the GPS signal is 380,000 times (19×10-2÷500×10-9) as long as visible light. We can translate these numbers into ocean wave that are more familiar. Let's use a 50 meter wavelength ocean swell to represent visible light. At that scale the GPS signal would have a wavelength of 19,000,000 meters. The circumference of the earth is 40,075,000 meters. So on this scale the GPS signal is well represented by ocean tides. A break water may be constructed to prevent the ocean swell and other waves from damaging boats moored in a harbor, the tide however will be hardly effected by a break water. Scaling 2.47 m ELT signal wavelength to this model would result in a wavelength of 247 million meters, or 247 thousand kilometers. To put this is perspective the moon orbits with a radius of 385 thousand kilometers. A wave with that length would propagate around the entire earth without much effect. The lesson here is that you can't assume a radio signal, even one that propagates by 'line-of-sight' will behave in any way similar to the way visible light does.

Another way to test theories according to scientific principle is through testing predictions. Once a theory is formulated, one can use it to make predictions. If the predictions turn out to be false, it is usually an indication that the theory has problems. We are examining a theory that the wing of an airplane can block radio signals. If true then any geometry that places the antenna behind the wing from the transmitter should result in a blocked signal. A VHF air-band communications antenna is 19 inches in height. A Cessna 172 has a wingspan of 36 feet 1 inch. This puts a top mounted antenna about 17 feet 3 inches from the nearest wing tip. Simple trigonometry shows us that any signal arriving at an angle greater than 5.3° below the wing should be blocked and not received by that antenna. So the theory allows us to predict that wing blanking should work with any angle of bank over 6°, or say 15° to give a comfortable margin. Why then does the CASARA training manual call for a 30° angle of bank, and the Civil Air Patrol documentation call for a 40° angle of bank? Perhaps there is another explanation. We might also ask why high wing airplanes, with antennas mounted above the wings, don't loose contact with any control tower that is directly left or right of them. 

There is another property of a radio antenna which is probably responsible for the observations that lead to the Wing Blanking technique; that is the antenna radiation pattern. Some antennas, when transmitting, send more energy in some directions than in others. The amount of energy sent in each and every direction is know as the radiation pattern and is normally presented in graphic form. Antennas work in reception in the same way they do in transmission, just in reverse. So if an antenna transmits more energy in a particular direction, it will receive correspondingly more energy from that direction.

There has not been a lot of work put into measuring the radiation patterns of antennas on general aviation aircraft, but there has been for commercial transport and military aircraft. These studies show that, as expected, an all metal fuselage acts as the ground plane or counter poise for the mono-pole communication antennas. The shape and conductivity of the ground plane does affect the radiation pattern. An ideal ground plane will restrict all energy to one side (the same side as the mono-pole) of the ground plane. An airplane, it turns out, is not an ideal ground plane. It is not uncommon to have significant energy radiated to (or received from) as much as 30° or 40° below the lateral axis of an airplane with a top mounted antenna. Here are two diagrams of such radiation patterns superimposed on a Cessna 172.

This is a much more satisfying explanation for this phenomenon. You will, no doubt, have noticed that changing the explanation of how the phenomenon works does not show the technique to be flawed. But we are just starting our examination of the technique in light of how radio signals actually work. We can't pick and choose which properties of radio signals we want to be in effect when we are using a particular technique. We must accept and account for all. Even though airplanes are able to maneuver in quite extreme ways, most of the time they are flying straight and level. This results in the communications antennas being orientated vertically. During a Wing Blanking maneuver the airplane will be banked over to 30° or 40° wich will result in the antennas be orientated off vertical by the same angle. We need to examine what effect this will have on the technique. Keep in mind that there is nothing about any particular high wing general aviation airplane that would lead to the conclusion that either of those two radiation patters are actually correct. The cone of silence below the airplane may be larger and require lesser bank angles, or smaller and render the Wing Blank technique completely useless. To be sure each airplane would have to be examined.

A Prayer

The property of radios signals that finally makes Wing Blanking a flawed technique is polarization. Polarization is a measurement of the orientation of the radio wave that is imposed by the transmitting antenna. For maximum signal reception, the receiving antenna must be oriented such that its polarization matches the transmitting antenna. As the polarity of the receiving antenna differs from the polarity of the transmitting antenna the received signal strength will be reduced until the polarity differs by 90°. Common examples of polarization are some types of sunglasses and flat panel computer displays. By placing polarized sunglasses lens in front of a flat panel monitor or LCD TV and rotating the lens you can see the effect that polarization mismatch has on electromagnetic radiation. The display emits light polarized a a particular angle, the sunglasses lens only passes light which matches its polarization. Here are three images showing the lens orientated at 0°, 45° and 90° relative to the polarization of the display:
If the search airplane is flying straight and level, its antenna will be oriented and polarized vertically. Even if the ELT antenna is not oriented vertically the relative orientation will remain the same regardless of the heading of the search airplane. This is good if the airplane is using proper direction finding antennas, or performing an Aural Search Technique. Similarly if the ELT antenna is oriented vertically (as it likely is during training or false alarms) then even when the search airplane is banked to turn the polarization will not change the signal strength since the relative polarization will be the angle of bank. It is when the ELT antenna is not vertical and the airplane is banked to turn, as in the Wing Blank technique, the relative polarization will be different for each heading the airplane takes while turning around in a circle. 

To understand this picture a toy top. When it first starts to spin its axis of rotation will normally be vertical. As it slows down it will begin to precess. This is similar to the motion of the polarity of the airplane communications antennas as it turns a full circle. If the angle of bank of the airplane is the same as the polarization angle of the ELT antenna, at one point in the turn the relative polarization will be 0°. When the airplane is 180° around the circle from that point the relative polarization will be the sum of the ELT antenna polarization and the airplane angle of bank. If the some of these angles approaches 90° then the signal strength will be greatly reduced, just as the crew would expect during a Wing Blank turn. This could result in two signal dips indicating two potential bearing to the ELT. If the airplane antenna radiation pattern does not give a dip, the plane is banking a 30° when a bank of 40° or even 50° is needed, this could result in a completely erroneous bearing. If the sum of the angle of bank and the ELT antenna polarization is greater than 90° the relative polarization will go through the 90° dip twice for each complete circle.

So there are two significant problems with Wing Blanking:
  1. Crews have been taught that Wing Blanking works because the wing blocks the radio signal. This implies, and some documents actually state, that one can determine if an airplane is suitable by visually examining the antenna placement. Here we have shown that one must actually perform a radiation pattern measurement on the airplane to determine suitability. Since there is no exhaustive body of work done on small general aviation airplanes, once can not be sure what changes to the airplane may change the radiation pattern making a previously suitable airplane unsuitable. 
  2. There is no way to know the polarization of the ELT antenna on a crashed airplane without first finding it. To be sure of effectiveness, electronic search techniques should be polarization independent. Wing Blanking is not polarization independent. 
We are not religious, and don't apeal to a deity during technical searches; but unless the Search and Rescue organizations that might come looking for you are willing to issue public assurance that they won't be using the Wing Blank technique, you may want to pray that if you do crash, your ELT finishes up with vertical polarization.