Difficulty Locating Lost Jet’s Flight Recorders Shows Need for Upgraded Technology

The search is on for the cockpit voice and flight data recorders (CVR/FDR) from the Air France A330 lost on the night of 31 May. The airplane, with 216 passengers and 12 crew and operated as flight 447, was lost in an area of known thunderstorm activity in the so-called inter-tropical convergence zone (ITCZ) on a night flight from Rio de Janeiro to Paris.

The airplane is believed to have been lost slightly less than halfway from northwest Brazil to Cape Verde, off the coast of Africa. In other words, it plummeted to the ocean pretty much in the middle of nowhere, and the recorders lie in pretty deep waters – about 13,000 feet below the surface. The lack of ground radar coverage in oceanic airspace means it will be difficult to pinpoint exactly where the airplane went down. To be sure, the recorders have an electronic beacon to facilitate their recovery, but the signal is fairly short range, and the batteries are only good for about 30 days.

This accident may galvanize efforts to equip airliners with the means to offload critical flight data in real time, thereby avoiding the necessity of searching for flight data recorders weakly broadcasting their location at the bottom of a vast ocean.

Thunderstorms and “hard” turbulence was reported in the area, and the aircraft was at 35,000 feet and loaded with fuel for the flight to Paris. During harsh economic times, when jobs are at risk, pilots will assist in making operations more economical by flying right up in the so-called “coffin corner” of the flight envelope despite their heavy weights. That is, they will fly where the margin between aerodynamic stall and Mach buffet is minimal to yield the best fuel economy (air nautical miles per pound of fuel). When that is done in the area of 55,000 foot cloud tops of the ITCZ, the environment-related controllability hazards become very real. To be sure, even at night, one can fly around the intense cells of thunderstorm activity visually (or by radar), dodging the storm tops.

However, it’s also possible to strike clear air turbulence associated with the ITCZ, or to stumble into a thunderhead that wasn’t seen ahead (courtesy of its electrical activity and flashes of sheet lightning).

All that’s needed is some turbulence or a sudden air temperature change (affecting both airspeed and thrust output) and the pilot could well be faced with a sudden loss of control on a dark night. Only a few thousand feet below are the solid, unfriendly could tops of the ITCZ’s solid thunderstorm activity. Once inside that maelstrom of cloud below, with its severe turbulence, hail, lightning, and so forth, a successful recovery from an unusual attitude would be problematic.

It appears that the Air France jet flew through three key thunderstorm clusters. Temperature trends suggest that the entire system was at peak intensity. From a turbulence perspective, cold spots would be the areas of highest concern as they signal the location of an active updraft producing new cloud material in the upper troposphere. Planes are clear of the most dangerous weather throughout a tropical system except when directly above an active updraft area. It is estimated that the aircraft flew through about 75 miles of numerous updrafts lasting about 12 minutes of flight time. Identical tropical storm complexes have probably been crossed hundreds of times over the years without serious incident, but not heavy flights risking coffin corner encounters to save fuel.

 

False-color satellite image of weather at the time Air France flight 447 made its last transmission. The active thunderstorm areas are defined by small-scale mottled areas of cold cloud tops. Temperature trends suggest that the entire system was at peak intensity. From a turbulence perspective, these cold spots would be the areas of greatest concern, as they signify the location of an active updraft producing new cloud material in the upper troposphere.

False-color satellite image of weather at the time Air France flight 447 made its last transmission. The active thunderstorm areas are defined by small-scale mottled areas of cold cloud tops. Temperature trends suggest that the entire system was at peak intensity. From a turbulence perspective, these cold spots would be the areas of greatest concern, as they signify the location of an active updraft producing new cloud material in the upper troposphere.

Although an airliner is built to survive severe turbulence encounters with “merely” broken equipment, components, and injured passengers and crew while the wings stay on, the stresses involved in recovering from an upset or loss of aerodynamic control at altitude can quickly exceed those parameters.

Having to recover from an upset at night while possibly descending through or into the convective clouds that put you there in the first place reduces everything to a strong odds-against situation.

Recall also that the A330 is a fly-by-wire (FBW) airplane that is limited in its maneuverability by control laws. That feature is intended to forestall pilots from getting into unusual attitudes, but it can also limit a pilot’s ability to recover from inverted or spinning situations. Control laws cannot, however, prevent cataclysmic overstress on the airframe.

The ACARS (airborne communications addressing and reporting system) message sent automatically by the aircraft to the airline operations center, indicating serious electrical failure and other system problems, may well be consistent with the aircraft being overstressed and breaking up. Only a few seconds would be needed for the generated faults to self-precipitate such a message. The cascade of systems failures is about what would occur from a structural failure following an overstress stemming from a high-level loss of control (i.e., a coffin corner stall-spin entry). The “advisory regarding cabin vertical speed” tends to confirm a high rate of descent – and nil cabin pressurization because of hull rupture – towards the end of the episode.

A few items already come to mind that merit investigators attention. Is Air France (and, obviously, other airlines) allowing or instructing crews to get high early and stroking that razor’s edge of coffin corner while still at heavy all-up-weight and vulnerable to loss of control? Is the Airbus fly-by-wire design, with its hard limits on the crew’s control inputs, more vulnerable than others during unusual attitude recovery? Do crews practice tangles with coffin corner in the simulator?

And if the recorders are not located, a reconstruction of the events leading to the airplane’s loss will be vastly complicated. Investigators may wish to pursue various technologies for transmitting data as it’s captured. This is what is done for space systems (the Space Shuttle, for example, does not have “black boxes,” but transmits information continuously).

It is time to move beyond archiving aircraft data in shockproof, fireproof, waterproof boxes, according to Sy Levine of Topanga, California. “The present system is analogous to having a patient in intensive care being monitored; however, few people look at the data until the patient dies or after release from the hospital,” he said.

The concept for downloading data in real-time was born of personal experience. As the former chief engineer for a major aerospace company, Levine was stimulated to think about the problem when one of the engineers who worked for him was killed in the 1994 crash of US Air flight 427. The paucity of data enormously complicated the investigation.

In Levine’s concept, the aircraft transmits a near-constant flow of information on its status. As an example of the integration of information he envisions, items like low fuel would cause a symbol to be generate on the air traffic controller’s display.

Basically, Levine’s concept would combine multiple inputs in the aircraft with real-time communications via satellite of sensor data to a ground station. His approach – which he has dubbed the Remote Aircraft Flight Recorder & Advisory Telemetry (RAFT) system – would bring flight/cockpit recorder information out of an old recorded-on-the-aircraft methodology and plug it into a near real-time database at a remote station.

Instead of recording cockpit voice and flight data on the airplane, the RAFT concept would send the information, via satellite, to a ground station.

Instead of recording cockpit voice and flight data on the airplane, the RAFT concept would send the information, via satellite, to a ground station.

As his concept paper explains:

“The global telemetry of the [flight data recorder] parameters allows aircraft monitored data to be simply and safely stored on the ground, thus making it readily available for aircraft statistical analysis programs that enhance air carrier efficiency and safety. Also, in the advent of a crash, it provides a timely accurate global estimate of the downed aircraft’s location for search, recovery and hopefully rescue operations.”

In this case, retrieval of human remains is probably all that can be hoped for – and recovery of the CVR/FDR, whose archived data is useful only if it is pulled up from the dark waters of the Atlantic Ocean. With telemetry, that critical information would already be available.