Summary of Facts: Egypt Air flight 804

20 May 16 20 Comments

Let me start by saying that the first week after a catastrophe is always a hotbed of rumours. When an aircraft is lost at sea, it usually takes a few days to a week to find the wreckage and recover any information; however ever since Malaysia Airlines flight 370, MISSING AIRCRAFT!! is headline news, with every major network offering minute-by-minute updates even though there’s nothing much to say, yet.

Meanwhile, I know that this post in its entirety will be out-of-date very soon, probably as soon as first thing tomorrow morning. It’s frustrating. And yet, as usual, I can’t stand to ignore the gossip as media outlets try to make sense of investigation reports which they only half understand.

So, here’s what we actually know so far.

On the 18th of May, Egypt Air flight MS804 departed Paris Charles de Gaulle airport normally and without issue for its scheduled flight to Cairo. There were 56 passengers and ten crew aboard the Airbus 320-200. The crew consisted of two flight crew, five cabin crew and three sky marshals.

The flight crew was experienced: The captain had 6,275 hours (with 2,101 hours on type), the first officer had 2,675 hours.

EgyptAir Airbus A320 (SU-GCC) on finals at Ataturk Airport by Mehmet Mustafa Celik
The Egypt Air Airbus A320 five years ago by Mehmet Mustafa Celik

The flight was cruising at FL370 (37,000 feet) when it entered the Aithinai FIR.

One issue is that there seems to be some confusion about the difference between Flight Information Regions and airspace.

All airspace is divided up into Flight Information Regions. These regions are generally a large area controlled by a specific Area Control Centre. Some countries, like Greece, have a single FIR, others are split into a number of regional FIRs and sometimes a single FIR will cover the territorial regions of several countries. A controlling authority administers the FIR and offers a flight information service to flights within its region.

So for example, the Athens Area Control Centre offers a flight information service for the Athinai FIR.

Athinai Flight Information Service
Athinai Flight Information Service

The important point here is that a Flight Information Region is not the same as airspace. Generally, a country’s national airspace is defined by the country’s land and territorial waters. (Interestingly, Greek airspace goes beyond this, claiming 10 nautical miles (19 km) of airspace around its borders, instead of the boundary of its territorial waters at 6 nautical miles.)

Flight Information Regions, however, are assigned and the region is administered by a country’s aviation services.. So for example, Athinai FIR covers all of Greece and the entire airspace over the Aegean Sea.

The point is that if you are in a Greek or Egyptian Flight Information Region, then you are not necessarily in Greek or Egyptian airspace; an FIR extends beyond territorial waters.

Sorry, it’s been bugging me.

That said, Egypt Air flight MS804 entered the Athinai (Athens) FIR at 02:48 local time. The Greek air traffic controller cleared the flight to continue to the exit point. The pilot appeared to be in good spirits and thanked the controller in Greek. The flight continued. There was no further conversation because there was no need for it. The flight continued along its planned route.

At 03:27 local time, the Athinai Area Control Centre called the flight crew to tell them to transfer to Cairo FIR. This is a normal hand-off as the flight leaves the Greek FIR and enters the Egyptian one. The controller received no response.

A few pilots have brought up the fact that the boundary point of the Greek and the Egyptian FIR is a well-known dead spot, where it is often not possible to communicate. The boundary point is the farthest point from both Area Control Centres.

That means that it isn’t necessarily relevant that the flight crew were not in communication with ATC at that time. If they had made a call, it seems likely that another aircraft would have heard it, even if ATC could not. However, if they could not hear the Greek controller calling them, they may well have not called out.

Nevertheless, that leaves us with a 40-minute window there where something may have happened on the flight. All we know is that the aircraft was still flying at the expected height and route at that time.

At 03:29, the aircraft flew over the boundary point between Athinai and Cairo FIRs.

The radar returns show that the aircraft then unexpectedly began to manoeuvre. According to the Greek Defence Minister, the aircraft turned left through 90° and began a rapid descent. Then it turned towards right through 360° while still descending rapidly to 15,000 feet. These manoeuvres are very specific and thus appear to be deliberate, not the result of an aircraft out of control.

My understanding is that the 90° turn is part of a common standard drill trained at various airlines. In the case of rapid depressurisation, the flight-crew makes a decisive turn off of the airway along with an immediate and rapid descent. The 360° is odder but could be an attempt to signal loss of communications or some sort of holding pattern perhaps? I’m hoping that someone can elaborate on this in the comments.

At 03:29:40 the flight signal was lost. The last return from the aircraft was at 10,000 feet in Cairo FIR, just under 7 nautical miles past the boundary point.

An accident investigation is conducted by the country in which the accident occurred. If the accident occurred in international waters, then the investigation is conducted by the country in which the aircraft is registered. This means that Egypt is in control of the investigation. Also relevant to an investigation is the registration of the aircraft (Egypt), the base of the aircraft operator (Egypt Air in Egypt), and the country(ies) responsible for the design and manufacture of the aircraft. As it is an Airbus 320-200, this means France. The French BEA have already dispatched three investigators to Egypt join the investigation.

Debris from the aircraft has been positively identified today. There were no survivors. The search and salvage operation will be focused on finding the “black box”, that is, the bright orange container(s) holding the Flight Data Recorder and the Cockpit Voice Recorder. This should be found relatively quickly although the Mediterranean has some very deep ridges in that area. If it isn’t found within a week, expect to see a lot of loud questioning as to why we still don’t have ejectable black boxes on commercial aircraft.

It is correct that the Egyptian civil aviation ministry have stated that Egypt Air 804 was probably downed by a terrorist attack; however I feel that I need to point out that recently Egyptian officials have shown a trend towards sharing sound bites before the facts are known. The French Foreign Minister has limited himself to stating that so far they have no indication as to the cause.

Category: Miscellaneous,


  • As always, your posts somehow achieve that sweet spot of being highly informative (to people like me who are not pilots), incredibly easy to read, and riveting.

    Question on this: “The radar returns show that the aircraft then unexpectedly began to manoeuvre. According to the Greek Defence Minister…”

    When you refer to the “well known dead spot,” you’re talking voice communications, but radar still tracks the planes? Is it only defense radar at that point, or do the civilian authorities track to the edges (and beyond?) of their FIRs? (Yes, I am almost entirely ignorant of how all this works.)

    • Right, the dead spot is an area where it is often difficult to receive ATC transmissions. These are often at FIR boundaries (there was also one where MH370 was last in contact) because of the distances involved. Note that doesn’t mean communication is impossible; just that pilots and controllers are aware sometimes communications break down in that area. It’s bad coverage which can be made worse based on atmospheric conditions, the quality of the aircraft receiver, etc.

      When you say defense radar, I think you mean primary (non-cooperative) vs secondary (cooperative) radar. As they had the exact height information of the aircraft, it must have been secondary radar. Using MH370 as an example again, one of the issues with tracking its original movements was that the countries in the area did not want to reveal how broad (or weak) their radar coverage was in the area. In this case, Greece has released the information of the aircraft as it was tracked from Athinai control centre, thus you could call it civilian radar, but I am not sure it’s a useful split.

  • Part of the puzzle why radar is more able to track may well be the amount of signal strength that is available from the aircraft electric supply. Meaning of course, the ability of the electrical generating units, the generators, to supply the power that can guarantee complete coverage of the VHF radio systems. The design of any aircraft always involves certain compromises. More power requires heavier generators, but also wiring that can handle the extra power. They will have to be more heavily shielded to prevent interference with other systems in the aircraft. Electrical wiring tends to set up electro-magnetic fields around them. Modern aircraft, as we all know, rely increasingly on electronics, all requiring their share of the ability of the system to supply them with adequate electrical power. And without extra shielding there is a risk that these systems may interfere with one another. It all would add weight and weight is the enemy of aircraft. Excess, unwanted weight is to be avoided and so, this is my guess but more qualified people can probably either confirm – or debunk – this: In general: VHF radio communications between aircraft and ATC are deemed adequately covered in general and the blind spots are acceptable.
    Of course, and here the terrorist theory gathers credibility, these blind spots also create areas that informed terrorists can make use of to hijack or destroy an aircraft and gain minutes before the outside world becomes aware.
    Ground radar is not bound to weight limitations and their signal strength can be many times that of the generating power of aircraft systems. Military radar systems are specifically designed, of course, to eliminate blind spots. So ground radar, working on UHF, different frequency bands than VHF, can in general detect objects or aircraft even if the VHF radios of the aircraft itself are outside of their operating range.
    And I agree that Sylvia has done an excellent job summarizing the news bytes as they are currently available (even if she pays some lip service to the now ingrained habit of calling bright orange objects “black”).
    And yes, she is also right when she mentions that there should be some method of preserving the data other than hoping that the FDR and CVR will be recovered. Ejecting the boxes before a crash may pose technical difficulties and the items may end up miles away from the actual crash site. It should be possible to transmit these data to a ground station in bursts to e.g. a satellite. Of course, if all systems are disabled because of structural failure, the most recent, vital data will be missing.
    Equally, cockpit crew may feel that their privacy will be violated to an unacceptable level if all cockpit conversations were to be transmitted to a ground station on a continuous basis.

    • Was the comment about the call for ejecting the FDR/CDR “before” the crash or after? Something like release them with a float and ELT after a crash is what I imagined.

  • > (even if she pays some lip service to the now ingrained habit of calling bright orange objects “black”)

    Haha, I thought of you when I put it in quotes :D

    • Well, they do say that orange is the new black, although I hadn’t realized that that was what they had in mind.

      Any comment on reports of ACARS data indicating a smoke condition in two locations?

      • For other people’s reference, here’s the info as per Aviation Herald:

        On May 20th 2016 The Aviation Herald received information from three independent channels, that ACARS (Aircraft Communications Addressing and Reporting System) messages with following content were received from the aircraft:
        00:26Z 3044 ANTI ICE R WINDOW
        00:26Z 561200 R SLIDING WINDOW SENSOR
        00:26Z 2600 SMOKE LAVATORY SMOKE
        00:27Z 2600 AVIONICS SMOKE
        00:28Z 561100 R FIXED WINDOW SENSOR
        00:29Z 2200 AUTO FLT FCU 2 FAULT
        00:29Z 2700 F/CTL SEC 3 FAULT
        no further ACARS messages were received.
        Early May 21st 2016 the French BEA confirmed there were ACARS messages just prior to break down of communications warning however that they are insufficient to understand the causes of the accident until flight data or cockpit voice recorders have been found. Priority as of current is to find the wreckage and the recorders.

        It will be interesting to find out more. There’s still a number of possibilities but it *does* mean that this wasn’t a bomb exploding on board.

  • Sylvia,

    That sounds like it could be very bad news for Airbus.
    The aircraft I flew were not fitted with ACARS, but what you mentioned here does not seem to make sense.
    RH sliding ockpit window? At FL370? And the reports of fire – in different locations on board? Makes even less sense.

    The initial suspicion that terrorists may have been involved did make sense, this new information is a bit baffling.

    And you are right, Sylvia. I took the bait knowingly.
    It must have been many years ago, perhaps more than 30, when journaists reporting aviation incidents were in no way allowing real knowledge of the facts to stand in the way of a good story. One day I was discussing the emergence of the expression “black box” to describe the FDR and CVR. This discussion took place in a workshop with the access panels to the avionics compartment of an aircraft were removed. It was full of black boxes. We were talking about the risk that if a crash were to take place in a remote tribal area, honest locals might preserve these black boxes and use the bright orange ones for salvage.
    Fortunalely, such a thing has not yet happened.

  • The mystery deepens around this sad event.
    Apparently, according to the available news, there was a fire in the aircraft. Investigators are looking into the possibility that the pilots were attempting a controlled ditching, or at least hoping to do so. Perhaps the pilots lost control as a result of fire damage to vital systems.
    The aircraft may have broken up as a rsult of the impact, or perhaps already in mid-air.
    The real challenge for the investigators, of course, will be to determine what caused the fire. If we may believe the news, fires broke out simultaneously in different locations in the aircraft: There was mention of fire in the cockpit AND in a toilet.
    Was there a common cause, like overheating of a system, a short in the wiring or was it as a result of unlawful interference? Meaning of course: an act of sabotage by terrorists.
    Was this a systems fault, sloppy maintenance or even suicide?
    We know that terrorists do not hesitate to commit those acts in the rather strange and mistaken belief that God will reward them.
    Although the very large majority of devout Muslims are peaceful, law abiding citizens there are crackpots who have been brainwashed to think otherwise. They have caused incalculable misery and damaged the standing of their religion.
    But if this crash were the result of a suicide mission then the stakes would seem to have been raised by a very large amount.
    A quick death by crashing an aircraft into buildings or terrain is very different from dying in a burning, out of control aircraft. So far the suicide killers have been carrying out their mission with a deadly mix of disregard for lives, including their own, and a dose of cowardice. This form of escalation seems very unlikely.
    Let us hope that the cause of this crash will be determined soon.

    • Anna has posted an interesting analysis to the Facebook page which seems to conclude that a single fire could cause all of the ACARS message. He believes that the fire started in the avionics bay under the cockpit.

      “The ACARS messages transmitted by the Egyptair flight are consistent with a fire in the front, right-hand side of the aircraft. It is very unlikely that sensors from two different windows fail at the same time during normal airline operation. The sensors’ fault was detected by the [Window Heat Computer] WHC, which is located under the cockpit, again on the front, right-hand side. Also, the SEC3 FAULT indicates a problem with the no.3 SPOILER ELEVATOR COMPUTER, also located in the front avionics compartment, again under the cockpit, very close to the no.2 WHC, mentioned earlier. The AVIONICS SMOKE message is generated when there is smoke in the avionics compartment. It also seems that the smoke was also detected by the smoke detector in the lavatory just behind the cockpit. To conclude, it seems there was a fire in the avionics compartment under the cockpit, which in turn caused a short circuit in the wiring leading to the right-hand window sensors and the SEC3.”

  • To me, whose career took place mainly on “steam-driven” aircraft, this explanation still seems to make a lot of sense.
    Now, if this is confirmed by further investigation, the task the investigators will face is: Was this a wiring fault or was the fire started by some slow incendiary device?
    In other words: was there a fault in the aircraft’s electrical systems, perhaps a poor contact? This sort of fault can be mundane enough. A poor connection causing heating up of the wires. This in turn can cause damage resulting in increased temperatures in the wire, the cycle eventually overheats the wire, can even cause corrosion which will in turn speed up the process until the insulation burns through and damages adjoining wiring. Leading to a short.
    In the coincidentally also French-built Corvette the procedure was effective: Two safety-wired switches, when pulled, would leave the pilots with nothing except standby instruments. Then start a smoke-evacuation procedure. The next job would be to pull all circuit breakers, then re-instate the systems one by one.
    When the offending system was brought back on-line, of course the smoke would start all over again. This was all trained in a real aircraft, in flight, with the oxygen masks on. There were only 40 Corvettes built and there was no simulator for this type.
    Of course, the A320 cannot fly without electrical supply. The Corvette’s elevator, rudder, ailerons, spoilers even still were operated by cables and rods. I have never been trained on these modern types so I do not know if there is any manual reversion, like in the BAC 1-11 where in the event of a hydraulic failure the flight controls could be coupled directly. A very hard job physically, but possible.
    If it is a fault in the aircraft, again there are two possibilities: Was this a fault during construction or during maintenance work?
    If it originated in the production process, and eventually manifested itself after a long time lapse where eventually the fault caused a catastrophic overheat, this may be very bad news for Airbus.
    If a wire was damaged during maintenance, then all related procedures will have to be investigated and possibly revised.
    A shockingly expensive prospect but the loss of life, 66 people, was bad enough. This must never be allowed to happen again.
    On the positive side: Although aviation can never be guaranteed to be 100% safe, the industry always has used incidents and accident investigationsot just to find out who is to blame, but also to improve safety for future generations of air travelers.

  • Hi Anna, thanks.
    I had a gawk at that site. It seems to be a discussion forum like “Fear of Landing” but it does not offer the same level of professional introduction to the topic.
    Which is, in my opinion, the reason why Sylvia’s website must be rated as one of the best.
    Reading, or skimming over the replies it does not seem to offer a qualified consensus about whether or not the crew can still have some measure of control if an A320 loses all electrical power.
    Rudder alone will cause yaw. Without other forms of control, like a yaw damper, this may set up the aircraft in an unstable condition called “Dutch roll”.
    Engine power? Are the power levers feeding the commands to the engines mechanically? The jets that I have flown were Citations (500, 550, 560, 650 and the souped-up version called “Stallion”), Lear 25, SN 601 Corvette and BAC 1-11. All of them conventional. Turboprops ditto. I have flown King Airs, Turbo Commander, Metroliner, Shorts Skyvan and 360. The most modern ones were the ATR 42 and Fokker 50, but most of my turboprop experience was on the Fokker F27 , and the American variant, the FH 227. They were the ultimate “steam-driven” aircraft on which I logged nearly 5000 hours. All of these aircraft, of course, had conventional flight controls. The F27 and FH227 had pneumatics rather than hydraulics. So obviously I am but an interested outsider when it comes to fly-by-wire aircraft. And I am therefore not qualified to make a judgment about who is right in the discussions.
    What has been discussed before, not only on this forum, is the fact that modern pilots tend to rely very heavily on the automatic systems.
    On a routine line check on the F50, I was “ticked off” by the check pilot because I liked to hand-fly the approaches. He asked me why I did not use the autopilot (cat 1 only, use of the A/P is mandatory on cat 2 and cat 3 operations).
    When I replied that I liked to keep up my flying skills, his interpretation was that this was an admittance of being “not up to standard” and in need of extra training. It earned me a place in the company “sin bin” and an extra simulator session.
    Check pilots in some airlines can be extremely pedantic and humiliating colleagues is good for their ego. Fortunately there are examples of supreme professional skills that can only have come from experience on conventional aircraft. Like the A320 that suffered loss of both engines and ditched in the Potomac with no worse injuries than some wet feet. And unfortunately the opposite, where a serviceable aircraft is lost with everyone on board because the cockpit crew lacked basic flying skills.
    I don’t know if this still is the case, but certainly in the past KLM cadets were given basic aerobatic training as part of the course.

  • Thinking in a mechanical cause, the hypothesis of the right engine rip off, is one of the very few facts which can explain the last a/c flightpath until crash.

    1. Steady flight, 37 kft and 500 knots, means that aerodynamic center and the center of gravity make coincidence in the same point, and the resultant of lift forces compensate the a/c weight. No torque exists, and then, no rotation and no banking happen.
    2. Sudden right engine rip-off means a center of gravity displacement towards the left wing, while aerodynamic center keeps being where it was. Then a torque appears from lift and weight applied in different points. The a/c banks to the left.
    3. To cancel out this torque, aerodynamic center has to be displaced over the current center of gravity position. This is achieved by increasing lift over the left wing, (aileron down), and decreasing lift over the right one, (aileron up). Either the auto pilot or the pilot himself, reacts by banking compensating to the right.
    4. All the above, referring to the a/c crossing plane, the banking plane. What’s happening in the wings plane? When the right engine rips off the wing, the center of thrust moves to the left engine position, now the only horse of the carriage, going away from the aerodynamic center, where the resultant of wind resistance is applied. A torque appears from left engine thrust and wind resistance, both applied on different points, doing the a/c to rotate to the right in the wings plane. This rotation means higher speed in the left wing than in the right one, then, lift increase over the left wing, then banking to the right.
    5. The composition of both torques, banking and rotation, the resulting torque, leads either to bank angle increase to the right and a/c nose going down, and then, lift decreasing and spiral rotation increasing, until crash.

    The fast turn to the right by the auto pilot or the pilot, after recovery from the sudden bank to the left, due to right engine rip off, allowed the rotation torque from the unbalanced thrust to bring the a/c nose down while banking to the right and this was fatal.

    I guess that the rotation torque could have been cancelled out with the rudder to left, even while banking to right with ailerons, and once the a/c levelled, to recover its way to Cairo slowly… But, of course, this needs the pilot to know what was happened, and I’m sure he never knew.

    Best regards.

  • I don’t want to pre-judge the situation but regarding electrical faults, all modern electrical systems are protected by fuses or circuit breakers. This goes for your house, your car and especially aircraft.
    In the event of an electrical fault eg a short between wires or faulty motor, a breaker will trip and cannot be re-set unless the fault is removed.
    If a fire occurred in the area of an electrical system, I feel an external ignition source is the more likely explanation!

    • Steve, I agree, there’s no normal reason why an electrical fault should have such a devastating effect. I would go so far as to say that the only two reasons I can think of is sabotage or a serious maintenance issue. Neither is good news for Egypt Air.

  • “My understanding is that the 90° turn is part of a common standard drill trained at various airlines. In the case of rapid depressurisation, the flight-crew makes a decisive turn off of the airway along with an immediate and rapid descent.”

    I understand the need for the descent but why turn away from the airway?

    • Well, hopefully one of our commercial pilots will chime in but I think the turn depends on the operator and their standard procedure. As to the reason, I think the point is to make sure you aren’t travelling down a designated airway at the wrong height and also so that your track signals definitively that you are intentionally off your route and manoeuvring.

Post a comment:

Your email address will not be published. Required fields are marked *


This site uses Akismet to reduce spam. Learn how your comment data is processed.