HRBuckley.Net

Two weeks ago I had a conversation with a CASARA navigator about the basis of the Aural Null electronic search pattern. He insisted that the technique was based on the ELT radiation pattern. In other words the ELT radiates power in all directions, as the searchers move away from the ELT the radio energy is dissipated and absorbed by the process of propagating through the atmosphere. When enough of the energy is dissipated or absorbed so that there isn’t enough left for the receiver to detect, the signal fades out. If the energy is radiated in the same strength in all directions, the radiation pattern is circular, then the signal will fade out at the same distance from the ELT regardless of which direction you are moving away from it.

On the other hand I know that the Aural Null is based on the radio horizon of the search platform. As pilots learn during basic training, VHF radio signals propagate in a straight line until they are blocked by terrain. (See: From the Ground Up Millenium Edition page 210; and AIP COM 3.5(b).) When the search platform moves past the radio horizon for the altitude it is at, the ELT signal can not propagate around the curve of the earth and the signal fades out. If the surface of the Earth in the area is reasonably flat relative to the search platform altitude, the radio horizon will be the same distance in each direction and the horizon will form a circle. I may have convinced him but I’m not sure.

Then earlier this week I received, from another CASARA navigator, a paper that discusses several search methods. The paper also states that the Aural Null is based on the ELT radiation pattern.

So let’s have a look.

The National SAR Manual gives a table of detection ranges for ELTs from aircraft at various altitudes:

The National SAR Manual correctly warns that these distances are for an ELT operating at full power. However, prior to the 121.5 MHz SARSAT packages being switched off, ELTs in courier trucks and with their antennas removed, were routinely detected by the satellites in orbits with altitudes between 850 and 1000 km. So it takes a fair amount of degradation to prevent the signal from propagating to the radio horizon for search altitudes below 10,000 feet. With some simple trig one can plot the location on the Earth under the aircraft (in red) and the location of the aircraft in the atmosphere (in blue) (X and Y scales are in nm, the Y scale is exaggerated with respect to the X scale for clarity):

ELT Detection Range

As you can see the surface of the earth curves away as we move from the ELT. An aircraft maintaining a constant altitude above mean sea level would follow this curve. The position of the aircraft at different ranges form (except for rounding error) a straight line. This means that an airplane that can detect the ELT at 200 nm and 30,000 ft would be receiving the same “beam” from the ELT as the airplane that can only detect the ELT at 30 nm and 1,000 ft (assuming both aircraft are the same direction from the ELT). If the Aural Null was based on the radiation pattern of the ELT, and one airplane lost signal at 30 nm and 1,000 ft, then the airplane at 100 nm and 10,000 ft would not be able to hear it because the signal energy would have been dissipated or absorbed. However because the Aural Null is based on line of sight propagation and the radio horizon it is entirely consistent with the detection ranges provided by the Canadian Forces.

What is troubling is the implications of a CASARA navigator believing the former over the latter. Just as in the days before Copernicus when the misconception that the Sun orbited the Earth prompted intelligent and educated people of the time to derive complex (but wrong) models to explain their observations; labouring on the misconception that the Aural Null is based on radiation pattern can lead intelligent and educated people to incorrect conclusions about electronic searches.

In case anyone needs reminding that resolution of an ELT signal is always needs to be done in a timely, efficient and correct manner; or the need to let people know when and where you are flying, and when to expect you back should read this press release from the US National Park Service.

This work is licenced under a Creative Commons Licence.

At this point in this series (which started here) I was going to ask you to apply your imagination to what can happen to an ELT during a crash. I thought this would be necessary because hard data on ELT crash survival has been difficult to come by. However, in going through Transportation Safety Board of Canada (TSBC) Aviation Reports over the past decade I’ve discovered that these reports have included steadily increasing content on the status of the ELT. The debate over ELT failure rates has been going on almost as long as ELTs have been mandated. In a NASA report published in 1990 the National Transportation Safety Board (NTSB) using 1983 through 1987 data found that in 75% of accidents the ELTs did not operate, the Air Force Rescue Coordination Center AFRCC using 1984 through 1987 data found that in 77.9%  of accidents the ELT did not operate. The NTSB further found that 88% of the failures were accident related.

Of the available TSBC Aviation Reports for the years 2008 and 2009 thirty six discuss the state and operation of the ELT. In nineteen (53%) of those reports the ELT could not provide alerting or search guidance because it was destroyed in the crash, destroyed by the post crash fire, submerged in water, unserviceable, or did not activate. Of the remaining seventeen accidents, in nine (25% of the total, 53% of the remaining), the ELT system was compromised by the ELT being separated from the antenna or antenna cable, or the antenna was damaged or destroyed. Of those nine reports, there were indications of a detectable or usable signal in four. Four of the reports indicate the ELT was capable of transmitting on 406MHz (TSO-C126), fourteen indicate the ELT was not able to transmit on 406MHz (TSO-C91 or C91a).

Of the four ELTs specifically identified as TSO-C126, one operated normally, one was submerged in water, and two of the ELTs had compromised antenna systems. Of these two, one was detectable by Low Earth Orbit (LEO) Search and Rescue Satellites (SARSATs) which were able to produce a position. The other was not detected by SARSAT or aircraft overflying the site, though there is no discussion in the report of the equipment, or operating procedures used by aircraft. Surprisingly similar to the numbers given in the 1990 report considering the TSO-C126 design was supposed to improve survivability. The Canadian Mission Control Centre (CMCC) is quoted in TSBC report A09Q0181:

…because the satellites are positioned so high above the Earth, if the antenna is damaged or blanketed by wreckage after the accident, the coded message cannot be captured. However, theLEO satellite can capture the coded message on 406MHz and fix the position of the signal, even if the antenna is damaged or poorly oriented. LEO satellites can sometimes receive a signal and identify its location, as was the case in this situation. [sic]

The corollary is of course that sometimes in these circumstances the LEO SARSAT can not receive the signal, as in TSBC report A09Q0111. So what does this tell us? In 1990 the best estimate was that in about 25% of air accidents the ELT was able to get a signal out to searchers. TSBC data for 2008 and 2009 confirm this except that in about 50% of accidents the ELT is not able to get a signal out, in about 25% it is, the remaining 25% is in a bit of a grey area. The ELT system has been compromised, but in in half of those there was evidence of a potentially usable signal. Would different techniques have allowed SAR crews to be able to detect and home a signal? My personal experience is that a compromised antenna system is often not a factor in airborne electronic searches when the receiving radio equipment is operated in accordance with best practices for the receipt of weak signals. For example, ELTs accidentally activated while being shipped to service depots do not have antennas attached. They are routinely located by SAR resources. I will have to keep looking.

This work is licenced under a Creative Commons Licence.

Some of my astute readers will have realized the Aural Null techniques, described in the previous article, require relatively level terrain to work. Some will also have realized that the ELT performance, the amount of power it is able to actually get ‘on the air’ must be enough to be received at the radio horizon for an Aural Null to work. But let’s take this one step at a time. First let’s look at how the crew can tell if the radio reception patten is not circular, and what can be done about that.

Let’s assume that our nice circular radio horizon is disrupted by an inconveniently placed hill that casts a shadow which is manifest as a bite taken out of the nice circular reception pattern. This is, of course, not a very realistic scenario, but it will help keep the diagrams from getting overly complex and that will help demonstrate the point. So we end up with a radio reception pattern below, and as our crew flies an Aural Null Pattern A as described earlier they will end up with the situation shown in this diagram:

An aural null performed on a non-circular pattern.

The beauty of the Aural Null techniques though, is they can be checked. Since the Aural Null depends on a circular radio horizon with the ELT at the centre and the points where the signal becomes detectable or undetectable are on the circumference, geometry allows us to check the search pattern for accuracy. There are two methods of testing. The first uses the property of a circle that requires points on the circumference (1, 2, 4 and 5) to be equidistant from the center (the crew has calculated to be at point 6). This is easily checked by using a ruler, or if you don’t mind poking yourself with sharp objects, a set of dividers:

The points on the circumference should be equidistant from the centre.

An other test, taken from Aural Null procedure B involves bisecting the chord lines formed by sets of points (1,4), (1,2), (1,5), (2,4), (2,5), and (4,5) and drawing perpendicular lines. If the pattern is circular the perpendicular lines will all intersect in the centre of the circle. If one or more of the points are not on the circumference then the perpendicular lines associated with those chord lines will be very unlikely to intersect at a signle point. We see this in the next diagram where the perpendiculars associated with point 1 are in red, and do not intersect in any single point; and those associated with the points which are on the circumference are in green and do intersect at a single point. This point is also the actual ELT location.

Using techniques from Aural Null pattern B to test the results.

So if the radio horizon pattern of the ELT is mostly, but not entirely circular the Aural Null patterns may give erroneous results, but the accuracy of the results may be tested by the crew and problematic points eliminated from the calculation, or more points acquired and tested until enough good ones are know to find the location of the ELT. In addition this shows that procedure A has a slight advantage in that if it doesn’t check out, a group of three points may still exist that provide an accurate position using procedure B.

Unfortunately it is often not that simple. Next time I will show you some properties of how radio signals work that can make life very difficult for searchers, particularly if something is limiting the amount of power the ELT is able to radiate.

I’ve been doing some data mining of published Transportation Safety Board of Canada aviation reports for an other article I’m writing. It is much more tedious than using the National Transportation Safety Board aviation accident database, but it gives me Canadian data. I was shocked to discover I had to create a data tag to capture the cases where, after a crash, the ELT was found to be turned off (In 36 reports for 2008 and 2009 mentioning ELTs I actuall found three such cases). It doesn’t really matter what you think about ELTs, if it is turned off it is cargo, not safety equipment. If you are the owner-operator-pilot, please make sure that you are complying with the ELT manufacturer’s instructions. If not, you may wish to talk to your Person Responsible for Maintenance (PRM) to see if the ELT configuration can be visually inspected prior to flight.