Mystery Trip

How Flight 370 Could Have Become a Zombie

Damage to the electronic nerve center might have left the airplane brain-dead but physically capable of flying for hours on autopilot until it ran out of fuel.

03.19.14 11:05 PM ET

Consider this one statement about Flight MH370: It flew until it ran out of gas.

On its own that’s a pretty remarkable image. The whole story of this flight begins to look different if you begin at the end, not the beginning.

The end is alarmingly simple and final. It is not surrounded by disputes about who did what and when. There is only one source for it, the “pings” indicating that the airplane was still “alive” in the air, and (roughly) for how long. If we reverse-engineer the progress of the Boeing 777 from this single moment of clarity, where does it begin to get complicated?

An airplane doesn’t normally fly until it runs out of gas if pilots are flying it. It’s not being “flown” at all in that sense; it’s flying itself. In order to do that it has to be in stable equilibrium – “inherent stability” is a quality that airplane designers are required to give the airplane.

This composure comes to an abrupt end when the fuel runs out. Both of the 777’s engines would not quit at the same time. Fuel would dribble to a stop in one before it did in the other.  At that point, with a loss of symmetry in the power provided by the engines, the airplane banks sharply and dives, into the water.

Given this scenario, we must assume that the crew and passengers were either unconscious or dead. Six or more hours have passed (the time varies according to interpretation of the pinging received by a satellite) during which the airplane has flow at least 3,000 miles.

Following the reverse timeline, this “clean” hypothesis – the undisputed physical characteristics of the airplane in flight – only gets “dirty” when it meets the cluster of events following the last call from the pilots at 1:19 a.m. In other words, what could have brought the airplane to the point where it headed off to its fate?

The whole picture of the 777’s behavior once it departed from its direct route to Beijing has, from the start, been colored by the quasi-criminalization of the investigation. Suspicion has been directed at two targets, the pilots and the passengers.

Sinister motives are given to what few fragments of information we have – the disappearance of signals from the transponder that fixes the jet’s position, the similar ending of signals from its maintenance monitoring system, the presence of a flight simulator in the pilot’s home and the deletion of some files from it – and the latest source of alarm: The fact that the 777’s sudden change of direction was programmed into its flight management system.

The more the information is slanted in this direction, the less easy it becomes to achieve an independent forensic focus. There is certainly a need to interrogate the facts for a criminal or terrorist interpretation – we live in a world of plotters – but there is equally a need to rigorously see if there are not alternative explanations innocent of malignant design.

Experts I have talked to believe that investigators should be, and probably are, including in their scenarios at least one that would be accidental and not criminal. This would have its origins not on the flight deck or in the cabin but in the belly of the 777 – in either the cargo hold or the electronics bay or both.

Electronically, the brain center of the 777 is in its Airplane Information Management System, AIMS, in the electronics bay. This handles the management of the flight itself – how the airplane is flown in real time – as well as the cockpit information displays, monitoring of all its conditions including the cabin climate, and the receipt and dispatch of data.

In the 777, the AIMS was designed with robust backup ability, what is called “deferred maintenance operation.”  If there is a failure in any one of its systems it can continue to operate for as many as 30 days before needing maintenance.

The airplane’s two umbilical links to the ground that have featured so critically in this case, the transponder and the ACARS monitoring device, relay their signals through external antennas. The transponder has two antennas under the forward section of the fuselage and the ACARS antenna is at the top of the rear fuselage (its design and position vary with models).

Get The Beast In Your Inbox!
By clicking "Subscribe," you agree to have read the Terms of Use and Privacy Policy
Thank You!
You are now subscribed to the Daily Digest and Cheat Sheet. We will not share your email with anyone for any reason

There are at least two locations that could be responsible for the loss of these communications – either by an electrical fault, failure or fire in the electronics bay itself, or as a result of some kind of explosion or fire in the cargo hold that affected the electronics bay. The AIMS units are the gateway for all communications to and from the flight deck. Experts believe, for example, that it is feasible that the loss of both the transponder and the ACARS signals could be explained by this kind of disruption, while backup systems still ensured that the airplane could fly.

Another explanation is that a certain kind of combustion in the cargo hold could rapidly introduce toxic fumes and smoke into the cabin and flight deck. The National Transportation Safety Board found that there was an unusually large consignment of lithium-ion batteries on the cargo manifest.

These batteries were for consumer electronic products like laptops and cell phones, not the much more powerful industrial-strength lithium-ion batteries that provide power to the systems of the Boeing 787 Dreamliner, batteries that caused fire emergencies and the grounding the 787 fleet.

An expert on lithium-ion technology, Dr. Victor Ettel, told me: “The organic electrolyte in lithium-ion batteries decomposes at very high temperatures, generating very toxic fumes typically containing compounds of fluorine and arsenic. Regulation packaging would inhibit the potential of open fire and therefore enhance the probability of generating toxic gases.”

In the case of the fire at Boston Logan airport in January 2013, caustic smoke originating the lower rear electronics bay very quickly filled the cabin. Since the 787 was parked at the gate and the cabin was empty there were no casualties in the cabin (firefighters who worked in the electronics bay were injured). Had the airplane been in the air, the smoke would have been lethal to both passengers and crew.

This raises the issue of whether such an event could incapacitate passengers and crew and yet leave an airplane able to fly itself.

Lithium-ion batteries do not represent a fully mastered science. An NTSB hearing on the 787 battery emergencies revealed that Boeing had subcontracted battery design to a French company, Thales, who in turn had subcontracted it to a Japanese company. The Japanese admitted that the technology was “not mature.” (The pressure to employ lithium-ion batteries comes from their ability to deliver more power for a lot less weight.)

Boeing disputed that the events were even fires. They said it would be more accurate to describe them as the venting of gases. However they are described, the effects are serious. There is no experience of lithium-ion batteries “venting” into an airplane cabin at cruise altitude and therefore no knowledge of whether, after such an event, the chain reaction that started it would stop for lack of oxygen.

Experts who have studied the erratic trajectory of MH370 after it made its turn told me that it could be attributed to the pilots dealing with the effects of either smoke, fire or a loss of oxygen due to a slow decompression of the air inside the airplane. In the case of a decompression the first response is to lose altitude as fast as possible, to get below 10,000 feet, in order to stabilize air pressure and remove the need for oxygen masks.

Reports that, instead, the 777 soared as high as 45,000 feet before rapidly falling are viewed with deep skepticism. As I have already reported, the airplane was heavy with fuel and would have struggled to reach even 38,000 feet.

The behavior of the Air France Airbus A330 that disappeared over the South Atlantic in 2009 has some bearing on understanding what the 777’s flight pattern could reveal. The Air France pilots, mishandling the airplane after its computerized flight management system had shut down, created a high speed stall, in which the Airbus pitched up from 36,000 feet to 38,000 feet and then, compounding the problem, the pilots failed to correct the stall. They had no control over its rapid descent into the ocean.

As it approached the water the Airbus had, in fact, assumed an inherently stable attitude, with its wings level and its nose slightly up as it would be on a final approach to an airport. One expert I spoke with pointed out that if the Malaysian pilots fought to regain control after having lost it, the 777 would first have followed the kind of zigzag course at various altitudes that some radar reports indicate it did.

The 777 is an inherently stable airplane. Once stabilized by the pilots, the airplane was in a condition where the autopilot could take over and fly it until the fuel ran out – which is what happened over the Mediterranean in 2005 when a Boeing 737 of Helios Airways lost cabin pressure and flew itself for three hours.

And so here we arrive at the intersection of the two timelines: the one beginning at the end and the one beginning with the 777’s change of course.

One stubborn fact punctures this de-criminalized scenario, though. Why was there no mayday distress call from the flight deck from the time when the change of course was programmed into the computers through the whole time the pilots were dealing with an emergency, no matter what its cause?

First, we need to understand the significance of the fact that the change in the flight plan was initiated by key commands on the computer. Much is being made of this as being sinister – that the pilots were either following some dark plot of their own or under duress from an intruder or intruders on the flight deck. 

There are two ways of making a radical change of flight plan like that. The first is what the pilots did, enter it into the flight management system. The second, which takes only slightly less time, is to disconnect the autopilot and hand-fly the airplane to its new course. The pilots could have simply felt that although the emergency required them to turn toward one of the several nearest airports, in Malaysia or Vietnam, there was not enough urgency that they needed to disconnect the autopilot to do this. Hand-flying the turn at that altitude and cruise speed would not have given the passengers as smooth a ride as leaving the airplane in the hands of the flight management system which is able to micro-fly with great subtlety.

However…the failure to send a distress call still undermines the neatness of this picture.

Until we know otherwise we have to accept that no such call was made. It is the one firm remaining indictment of the pilots’ behavior.

Yet given how slippery almost every other piece of information given by Malaysian authorities has turned out to be, could this also be suspect? Did the pilots in fact make a call that was not heard or reported in the early hours of that Saturday? After all, retrieving something as simple as an accurate radar track of the flight as it left the control of Malaysia and entered Vietnamese air space has proved to be strangely difficult.

And when it comes to tracking the entire flight, to knowing for exactly how long the 777 flew while not knowing at all which direction it took, we remain dependent on the information from Inmarsat, the British company whose satellite received the continual pinging for as long as the flight lasted.

In 2014 we have suddenly lurched backward to 1937 and the world of Amelia Earhart. When she disappeared over the Pacific there were said to have been radio transmissions three hours after her last voice contact. People are still hunting for her and her Lockheed Electra today.