Newly emerged details concerning Malaysia Airlines flight 370’s electrical system indicate that whoever took over the plane was technically sophisticated, possessing greater knowledge of Boeing 777 avionics than most commercial line pilots. They also suggest that the plane’s captain, Zaharie Ahmad Shah, was not responsible for taking the plane.
The new information comes via Michael Exner, a satellite industry veteran who has been one of the most prominent independent experts investigating the airliner’s disappearance. Several days ago Exner gained access to a major US airline’s professional-grade flight simulator facility, where he was able to run flight profiles accompanied by two veteran 777 pilots. “This is a state-of-the-art 777 simulator, level D, part of one of the most modern training facilities on earth,” Exner says.
A little background. As is well known, approximately forty minutes after its departure from Kuala Lumpur for Beijing, someone turned off all communications between MH370 and the outside world. Around the same time the plane turned sharply to the left and headed back over the Malayan Peninsula. Among the systems that were shut off were satellite communications; the transponder; and two automatic reporting systems, ACARS and ADS-B. The plane went dark just as it entered the space between two air-traffic control zones and was temporarily unmonitored, a sign that whoever planned the diversion wished to avoid detection and was well versed in international air traffic control procedures.
For approximately the next hour, MH370’s progress was visible only to military radar. The plane flew straight and fast between established navigational points, indicating that the aircraft had not suffered mechanical accident. At 18.22 UTC the plane was heading west out into the Indian Ocean when it passed out of range of military radar. At that point, the plane became effectively invisible. Shrouded in night, with approximately six hours’ fuel aboard, the plane could have reached any point within a 3000-mile radius and no one on the ground would have been any wiser. But it did not stay dark. Less than a minute later, MH370’s satellite communications system was switched back on.
Over the span of several minutes, between 18.25 and 18.28, the Satellite Data Unit (SDU) transmitted a flurry of brief electronic messages with Inmarsat satellite 3F-1, which occupies a geosynchronous orbit above the Indian Ocean. In a report issued this June, the Australian Transport Safety Board stated that the signals were “generated as part of a Log-on sequence after the terminal has likely been power cycled.”
Until now, it has not been publicly known how such a power-cycling could have taken place.
At the simulator facility, Exner reports, he was able to confirm “that there is no way to turn off the primary power to the satcom from the cockpit. It is not even described in the flight manuals. The only way to do is to find an obscure circuit breaker in the equipment bay [i.e. the Electronic and Equipment bay, or E/E bay, is the airplane’s main electronic nerve center].” Both of the pilots accompanying him told Exner that “pilots are not trained to know that detail.”
Why the satellite communications system was turned back on is unknown. The system was never used; no outgoing telephone calls were placed, no text messages were sent, and two inbound calls from Malaysia Airlines to the plane went unanswered. Aproximately every hour for the next six hours, however, a geostationary communications satellite sent electronic handshake signals, and the SDU aboard the plane responded, confirming that the system was still active and logged on. Though the signals contained no messages per se, the frequency at which they were sent, and the time it took to send and receive them, have been used to determine the plane’s probable direction of travel.
The fact that the SDU was turned back on provides a window into the circumstances of the hijack. For one thing, since the SDU integrates information from other parts of the plane’s computer system, we know that the plane’s electronics were substantially functional, and perhaps entirely so. Second, the fact that the perpetrator (or perpetrators) knew how to access this compartment and how to toggle the correct switches suggests a high degree of technical sophistication.
Further evidence of the hijacker’s sophistication comes from the fact that they also managed to turn of the ACARS reporting system. This is can be done from the cockpit, but only by those with specialized knowledge. “Disabling it is no simple thing,” Emirates Airline CEO Tim Clark told Der Spiegel recently, “and our pilots are not trained to do so.”
For all its importance, the 777 E/E bay is surprisingly accessible to members of the flying public. The hatch, generally left unlocked, is set in the floor at the front of the first class cabin, near the galley and the lavatories. You can see a video of a pilot accessing the E/E bay inflight here. (In Airbus jets, the hatch is located on the far side of the locked cockpit door.) Once inside, an intruder would have immediate physical access to the computer systems that control communication, navigation, and flight surfaces. A device called a Portable Maintenance Access Terminal allows ground crew to plug into the computer system to test systems and upload software.
The security implications of leaving the plane’s nerve-center freely accessible have not gone unnoticed. Matt Wuillemin, an Australian former 777 pilot, wrote a master’s thesis on the vulnerability in June 2013 and submitted it various industry groups in the hope of spurring action, such as the installation of locks. In his thesis, Wuillemin notes that in addition to the Flight Control Computers, the E/E bay also houses the oxygen cylinders that supply the flight crews’ masks in case of a depressurization event and the controls for the system that locks the flight deck door. “Information is publicly available online describing the cockpit defences and systems located within this compartment,” Wuillemin notes. “This hatch may therefore be accessible inflight to a knowledgeable and malevolent passenger with catastrophic consequences.”
Wuillemin reports that, among others, he sent his thesis to Emirates’ Tim Clark. A vice president for engineering at Emirates responded that the airline did not perceive the hatch to be a security risk, since the area is monitored by cabin crew and surveillance cameras. Wuillemin notes that cabin crew are often called away to duty elsewhere, and that the surveillance cameras are only routinely monitored when someone is seeking entry to the cockpit; he adds:
Emirates considered the possible requirement for crew to access the area should there be a ‘small’ in-flight fire. Research indicated there is no procedure, checklist or protocol (manufacturer, regulator or operator) to support this latter position. In fact, Emirates Operations manuals (at that time) specifically prohibited crew accessing this area in flight. Emirates amended the Operations manual recently and re-phrased the section to ‘enter only in an emergency’.
The fact that someone must have entered the E/E bay during MH370’s disappearance diminishes the likelihood of one of the more popular MH370 theories: that the captain barred himself in the cockpit before absconding with the plane. Even if he locked the copilot on the far side of the door and depressurized the cabin to incapacitate everyone aboard, emergency oxygen masks would have deployed and provided those in the cabin with enough air to prevent Zaharie from leaving the cockpit before the next ACARS message was scheduled to be sent at 17:37, 18 minutes after the flight crew sent its last transmission, “Goodnight, Malaysia 370” at 17:19.
It’s conceivable that Zaharie could have acted in advance by leaving the cockpit, descending into the E/E bay, pulling the circuit breakers on the satcom system and then returning to the cockpit to lock himself in before making the final radio call and diverting the plane to the west, depressurizing the cabin, and waiting until everyone was dead before returning to the E/E bay to turn the SDU back on. But if his goal was to maintain radio silence he could have achieved the same effect much more simply by using cockpit to controls to deselect the SDU without turning it off.
As it happens, Wuillemin’s efforts to draw attention to the potential hazards afforded by unlocked E/E bay hatches proved too little, too late. MH370 went missing just two months after he submitted his work to the Australian government.
@sk999
If the turn south was at IGOGU: it would mean again a major turn at ATC zone border. Quite remarkable.
What of the in-flight enertainment system?
According to the ATSB report, the signals at 18:27 came from the in-flight entertainment system. A perusal of the packets from 18:25:34 (starting with the pair of log-on and log-off messages) show the exact same pattern as at the beginning of the Data Communications Log, when the plane was parked at the gate at KL. The relevant packets from that time are at 16:00:13 and 16:01:16 to 16:01:29. (Intervening unrelated messages are presumably from ACARS). So the IFE behaved identically both times. Whether it was because the IFE was being powered up as well or whether it was responding to the SATCOM’s change in status is an open question. Just another bit of information to note.
“No route in ATSB Wide Area has both steady speed and steady bearing. Cannot fly there on auto-pilot.”
https://twitter.com/DrBobbyUlich/status/520923785499660288
Any route can be flown on autopilot.
Was a heading indicated by the Maldives eye witnesses ? Would be interested if aircraft heading south, or towards Madagascar or Eastern Africa …
@VictorI: you’re right, of course. I had s38 on the brain when I wrote earlier, but yes, Go Phoenix has your s34 covered, doesn’t it…
Under your scenario, do you have MH370 take on more fuel at BA? Or do you trust the ATSB’s Jun.26 performance limit (which indicates enough fuel to accommodate a 2nd landing) more than its Oct.8 PL (which seemingly doesn’t)?
Finally: I tried adrift.org.au from [s34, e94] – I can’t get your scenario’s debris up to Indonesia, either. Do you have any take on that bizarre drift analysis reference, Victor – either in general, or as it might pertain to your landing scenario in particular?
@ Gysbreght:
I think we can agree on that.
Dr. Bobby’s assertion is qualified by “ATSB Wide Area”. As you can see from the convo attached to the tweet, there was back-and-forth about it.
If you (or others) have more to add, it would be appreciated.
@ Myron, sk999 Niels:
Re MH370’s (possible) heading and sk999’s interesting comment re a possible turn at IGOGU, please take a look at this diagram (posted here previously):
https://twitter.com/idannyb/status/506978985167704064
Rough, but you will note that Kate Tee’s reported sighting appears to be in the vicinity of IGOGU. I need to ask her.
And re-upping @albertoriva’s comment in his IBTimes article (see my Maldives post in previous thread):
“the autopilot would have kept them flying straight and level on the last compass heading. (Which would have taken MH370 more or less over Kuda Huvadhoo, by the way.)”
@Brock: At the time I generated the path with the landing at BA, I did some rough calculations that convinced me that the additional fuel required for the descent/ascent was balanced by the time spent on the ground so the fuel was predicted to last until 00:15 (but not achieve the same range, hence ending further north on the arc). To fully answer the fuel question, I need to go back and do the detailed calculations (allow fuel flow to vary with plane weight) as I did for the non-loitering path ending around 37.5S.
if the flight continues along a path similar as it had to reach Kuda Huvadhoo but did not land, it might have gotten to Somalia?
And for those who can’t imagine that MH370 might have vanished because a nefarious event (read: hijacking):
ICYMI:
An Al Qaeda informant testified in a US court proceeding that he’d met Malaysian jihadists (one of whom was a PILOT), who were planning a 9/11 style attack by taking control of an airplane.
“Badat, who spoke via video link and is in hiding in the UK, said the Malaysian plot was being masterminded by Khalid Sheikh Mohammed, the principal architect of 9/11.”
“his claims were first made long before the disappearance of Flight MH370.”
http://t.co/RH6s826qCw
Nihonmama – the story about the aircraft plot(Badat) was discarded by many as pertaining to a previous well known plot from years earlier. But, it contains elements that did not actually surface at the time, such as claims of a “Malaysian pilot” and a “group of Malaysians.” Plots evolve as much as they are planned.
Spencer – Seems you are still intent on misrepresenting Mike’s posts? I don’t see a contradiction, rather an extension or clarification on his part. He appears to have put some distance between himself and the simple attribution of human hand in the possible switching on/off of the SDU. My only point was that the alternatives were not as good a bet and that the number crunching represented a series of bets. Mike has been a gentleman throughout and patient with people who do not have his background. Rand has appealed for decorum in the past, as I have done before that, and I’ll do it again.
@ Myron:
At the end of that same article:
“As for the coast of Africa, lawless Somalia, an ideal place for a hijacked plane to land, is 2,000 miles from the Maldives. That would have been too far for Malaysian 370, and was also likely among the first places that spy satellites would have scoured for signs of a giant, easily visible airplane, which they did not see.”
But FWIW, Somalia was my initial thought when the plane vanished. BBO (Berbera) in Somaliland (autonomous) is one of longest runways in Africa.
IDK, still can’t throw that one away…
@ Matty:
“contains elements that did not actually surface at the time, such as claims of a ‘Malaysian pilot’ and a ‘group of Malaysians.’ Plots evolve as much as they are planned.”
You are spot-on.
Which takes me back to this:
https://twitter.com/nihonmama/status/447077518214840320
I did not then, nor do I now, nor will I ever believe that a Dornier with US Spec Ops people aboard ran out of fuel over Banda Aceh, INDONESIA.
And where were they coming from?
The Maldives.
@Gysbreght,
You posted November 8, 2014 at 5:34 PM:
“Any route can be flown on autopilot”
I disagree.
What I have said is that no route that (1) uses one LNAV mode (out of the five available) coming out of the final turn and (2) has no human making changes in the manual throttles or the auto-throttle setting after the final turn, can end in the ATSB wide area and also be consistent with the satellite data.
If you have discovered a route that violates this statement, please provide route coordinates at all handshake times.
I discount the possibility of any pilot knowing in advance the exact times of the handshakes that occurred after the turn. Those times would be unknowable because of the two satellite phone calls initiated from the ground that reset the Inmarsat timer.
For all routes that end in the ATSB Wide Area, either the speed or the bearing must be varied significantly between the handshake arcs in order to be consistent with the BTO data. You can have either one, but not both.
You are also welcome to try using any sequence of existing waypoints to define the southern route. I have not been able to match the satellite data with multiple waypoints and also end up in the ATSB Wide Area.
The only route that has both (1) bearings consistent with one LNAV mode, and (2) speed consistent with manual throttle or auto-throttle in ECON, LRC, MRC, or Mach# modes is considerably west of the ATSB Wide Area.
My conclusion is that active human control of the aircraft speed and/or direction is required for a period of 5-6 hours after the final turn in order to generate the measured satellite data and to end up in the ATSB Wide Area.
The corollary to this is that if you believe no additional human input to the aircraft controls occurred after 18:40 UTC, 9M-MRO could not have ended up in the ATSB Wide Area.
I would be interested in evaluating proposed routes from anyone that believes they can disprove my statements above. Just post waypoints or coordinates and the speed control mode used. So far, all of the routes proposed by the ATSB, Inmarsat, and the IG require human intervention throughout the flight, either in bearing or in speed in order to be consistent with the satellite data.
@Bobby: I think you will agree (because you have in the past) that if a loitering around Sumatra is allowed, my path that includes the segment BEDAX and then due south ends in the search area (@34.25S,93.79E), is a constant TAS, and satisfies all the satellite data. The requirement of hands-off after 18:40 is an assumption, not a fact, that may or may not be true.
Nihonmama – I found a bit here on the plane you are talking about.
http://www.ainonline.com/aviation-news/ain-defense-perspective/2013-10-11/328-group-hands-over-last-ussocoms-wolfhounds
Factoring that in, Aceh and Maldives are two Islamist hotbeds so a Wolfhound popping up there is logical on one hand, but it looks odd too. What we do know from that episode and a few since is that Indon radar is now going just fine. If Indon radar saw MH370, and the ATSB are actually using their data, I think word would be out by now – just a stab, so does the search assume that the Indons are lying? Indonesia isn’t really a country with a lot of secrets. Not responding I can grasp, but insisting you saw nothing is a story with a lifespan – if it was untrue.
Of interest maybe: If special ops Wolfhounds are active around Maldives you can be sure Jindalee will be poking it’s nose in there too in support, but that would be a specific tasking.
This analysis is one of the best probable theories of disappearance of MH370. Since, there is no MAY DAY call from the cockpit, it points towards involvement of pilot(s) behind this. May the almighty give enough strength to all the dear ones of the passengers to bear with the loss…
@ Matty:
A Wolfhound is exactly what that was – which I didn’t know at the time of that tweet. The very good article you found is one that I only saw much later. And FYI, there was another blog that has loads of info about the Wolfhounds. But sometime between that tweet in March and early September, the owner of that blog made it private:
http://malaysiaflyingherald.wordpress.com/?s=Wolfhound&submit=Search
“Indonesia isn’t really a country with a lot of secrets”. A very interesting observation.
Remember that Indonesia’s Police Chief had a presser and said he knew what happened to MH370? Then Malaysia’s police chief quickly (and on Twitter) said his Indonesian counterpart was “misquoted”. Which does not apparently jive with the takeaway of all of the other news orgs that were in the room. (I posted about this one or two threads ago). We still don’t know what Indonesia’s police chief intended to say.
Don’t know if the search assumes Indonesia’s lying. But does the search have info about what Indonesian radar saw with respect to MH370 that it’s not sharing?
‘Portmanteau’ on pprune (Apr 8):
“an Indonesian military chief said ‘the aircraft was not detected flying over our territory’ which not the same as saying we never saw it… I suppose the convention is that you do not comment on what you have seen going on there. ditto Malaysia.”
Certainly appears Indonesia’s police chief was bursting at the seams to talk.
And something else, from @Petrossian50
(Nov 5):
“The Indonesian intelligence agency BIN is leaderless right now. There is probably nothing more dangerous than such a thing without reigns”
@VictorI,
Thanks for responding, Victor. Yes, I agree that the choice of “hand-on” or “hands-off” after 18:40 is an assumption. That is why I presented two scenarios – one hands-on and one hands-off.
If you would please post the coordinates of the “loitering route” you are suggesting I would be most appreciative. Please indicate which waypoints and LNAV mode are used. Alternatively, you may get my email address from Jeff and send this to me directly. Thanks, Victor.
@Jeff: for the sake of sheer academic rigour: is the data record at 17:21 – in and of itself – also consistent with a catastrophic event (e.g. massive explosion) at 17:21, or would such an event cause the drop off of comms to register differently in the data trail?
I freely admit any such event is spectacularly inconsistent with both of the next two segments of supplied path data (1 hour of primary radar, followed by 6 hours of signal data). I’m just trying to understand whether we still need to trust those delayed-release datasets in order to rule that scenario out.
Jeff, please indulge a few more questions.
Since it is always critical to distinguish between facts and assumptions, what is the basis for stating “As is well known, approximately forty minutes after its departure from Kuala Lumpur for Beijing, someone turned off all communications between MH370 and the outside world.” It is a fact that communications systems weren’t communicating, but is it established fact that this was the result of some party’s deliberate action? We may feel someone shutting things off is likely, but is there anything close to irrefutable evidence?
Moving on, there is the assertion that pilots would typically lack the knowledge of how to turn off ACARS from the cockpit. Are you familiar with an article from the New Straits Times which includes the tidbit that it was routine for MAS pilots to shut down ACARS on flights to China?
See http://www2.nst.com.my/business/todayspaper/font-color-red-missing-mh370-font-acars-cannot-be-disabled-1.521314
@Dr. Bobby Ulich,
I’m not aware of any justification for the completely arbitrary assumption that no human input was made after 18:40 UTC. Do you have any?
@PhilD,
I believe the article is saying in a somewhat distorted way that the ACARS service provider does not have VHF facilities in China. Therefore ACARS must use Satcom instead of VHF to communicate with the service provider when flying over China.
@Bobby,
I have sent you by email a path that ends with the segment BEDAX to SPOLE. As the path is directly (true) south, the mode could either be LNAV (flying by waypoint) or constant true track. (I suspect the South Pole waypoint was entered.) The path uses a constant (temperature-corrected) Mach number of 0.813 at 35,000 ft and includes our knowledge of the wind and temperature field at each minute along the path.
It would take some time to find your previous comment on this blog regarding this path, but my recollection is you have already evaluated this path and said you would change the speeds slightly to reduce the range error at 22:41 to be lower than 8.5 km. With that change, you believed the path was acceptable, understanding it would require a loiter. (I allowed no variation in Mach number at the expense of some range error.)
Note that this path does not require a reduction in speed between 22:41 and 0:11 as your path requires, which I believe is a weakness of your proposed path because it implies a fuel imbalance and early flame out of one engine, which would also reduce the range. We have had this discussion before so I won’t rehash it here.
Victor
@Gysbreght,
There is not, at this time, irrefutable proof that 9M-MRO flew without human input after 18:40. Nor is there irrefutable proof that there was human input after that time.
Two items cause me (and others, including the ATSB) to lean in the direction of a hands-off flight after 18:40. First, there are the two unanswered satellite phone calls at ~18:40 and ~23:15, which supposedly rang in the cockpit. Second, there is a great circle path that produces very steady airspeed. Thus, it is possible that the plane flew hands-off.
The issue might be resolved to some degree of certainty by the location of the aircraft wreckage, or definitively by examination of the Flight Data Recorder.
(Sorry if this is a duplicate but my browser crash on submit)
So does the Maldives sightings fit with the satellite data bfo/bto calculations and flight paths ? And from there a due south flight until fuel exhaustion ..
Dr. Bobby Ulich posted November 9, 2014 at 9:28 AM:
“Two items cause me (and others, including the ATSB) to lean in the direction of a hands-off flight after 18:40”.
Sorry, I can’t read that in the ATSB reports. In both the June Report and the October update the ATSB expresses no preference between the AP-constrained paths and the data-optimized paths. For the sole purpose of selecting a width for the search area, the June report assumes that the crew is unresponsive at fuel exhaustion. In that context, they write:
“IF ENGAGED, the autopilot could have remained engaged following the first engine flame-out but would have disengaged after the second engine flamed-out.” (my emphasis).
and:
“Note: This suggestion is made for the sole purpose of assisting to define a search area.”
@ Dr. Bobby Ulich
P.S.
Even if there was “a long period of flight under autopilot control”, where does that period begin? Why not somewhere between 18:40 and 19:41?
@ Dr. Bobby Ulich
P.P.S.
The performance boundary is not constrained by AP modes:
“The heading and speed were constrained so that all paths intersected the BTO arcs at the times of the handshakes.”
@PhilD & Gysbrecht
In an attempt to differentiate what can be disabled and/or turned off.
The discussion here concerning the SATCOM being turned off relates to the AES complex: that is the SDU and its ancillaries including the antennas. The AES carries the ACARS encapsulated messages over the satellite datalink to the ground station if satcom is the selected datalink. It is the AES “complex” that is controlled by 3 cct brkrs in the EE Bay with no direct control on the flight deck.
However, it’s the avionics’ data communications management function which routes and assigns priorities to the ACARS messages between the AES datalink & the other avionics functions. Transmission of messages by that data comms mgmt function can be readily disabled, by familiar menus on the CDU beside the pilots’ knees, to inhibit messages going out over the AES. Those familiar menus also provide the means for the crew to select VHF or SATCOM or AUTO as the datalink bearer.
It’s that selection process that is the basis of the China connection: China & ARINC deployed VHF Data Link coverage over China and SITA customers (such as MAS) may roam onto that remote ground station (RGS) network. VHF datacom use is cheaper than satcom and if an aircraft is in range of digital data link capable VHF RGS then its generally preferred.
So the ATSB stated, re-iterated by Mike, that there are a number of reasons for an in-flight AES log-on to occur. A power outage is one but there are others that don’t involve intervention in the EE Bay & might well be the result of some technical anomaly. The coincidence of the transponders ceasing to transmit & no further radio contact after 17:21 is a more complex scenario to explain.
Getting to an explanation is even more difficult when the Mlsian authorities remain intent on their policy to keep straightforward information that is normally presented as fact behind closed doors.
:Don
@GuardedDon:
Thanks for explaining the ‘China connection’. You’re saying that SITA is the ACARS service provider for MAS, that they may ‘roam’ in somebody else’s VHF Data Link network, and that is cheaper than using Satcom. With that background, can you explain this quote in the linked article:
“It is Mas procedure to switch ACARS, VHF, and High Frequency selection off but this is only for flights to China as the service provider for Mas does not cover China. Some if not all pilots switch them all off for a while and then later switch SATCOMM back on to force the system into SATCOMM mode.”
@ Matty:
Re “The Indonesian intelligence agency BIN is leaderless right now.”
I forgot to add: the vacancy is in part due to the recent presidential election. President-elect Joko Widodo hasn’t made his intel choice yet. But it also appears that BIN is going through a much-needed re-org.
Very interesting article:
“According to one retired intelligence officer, BIN has been able to score enough; in other word, its performance is neither good nor bad. His remarks signify that BIN has not done much of its homework to show its performance to Indonesia’s public at large. The Intelligence Law, as a legal framework to oversee intelligence, is not being implemented by our national parliament. Therefore, it likely depends very much on leadership of upcoming new presidential directive to speed up the expected reform of this particular “untouchable” security actor and place the right man on the right job.”
http://bit.ly/1xyyclz
With the flights to China its a standard operating procedure to turn off the coms but to start it up again ??
If so why all the fuss pinning fault on the pilots.. Maybe it’s a weakness exploited by hijackers. But this tidbit opens up a bunch of new questions for which scenario which caused the aircraft to go missing….
Don – I think I’ve read that “human factors” were key to interpreting the signal data all the way to the search area, and that sterile math was not enough on it’s own. It’s commendably scientific to state that some technical anomaly could have been the cause of the log-on in question, but we aren’t just diagnosing some malfunctioning machine here, we KNOW there were human factors. That is what guides a criminal investigation as much as anything.
Going suddenly conservative on the likely power interruption to the satcom looks like a smoke grenade.
@Gysbreght
There are many VHF network providers covering various regions around the globe, I assume they have reciprocal roaming arrangements. SITA has a much broader user base than ARINC.
The Transport Canada report on the Swissair SR111 accident details the datalink activity of that aircraft from departure to loss of comms: a SITA customer using ARINC’s VHF RGS network & satcom.
I don’t know the provenance or veracity of that oft repeated quote but it has a ctrl-alt-del ring to probem resolution. Also, ARINC is the only HFDL operator but it’s an incredibly slow medium.
@Matty: I don’t do ‘smoke grenades’. I have simply described what I understand, at this time, to be ‘likely’.
The challenge here is we don’t know very much. The authorities in Mlsia have decided that ICAO’s recommendation for the open reporting of facts doesn’t apply.
@Gysbreght,
In response to your post:
“Even if there was “a long period of flight under autopilot control”, where does that period begin? Why not somewhere between 18:40 and 19:41?”
I chose to use 18:40 because the BFO data indicate that the aircraft was flying south by then. That means that the last “known” human input (that resulted in the southward turn) occurred sometime before 18:40, and there is no “known” human input after then. So 18:40 is the dividing line based on the available information.
@Gysbreght,
In response to your post on November 9, 2014 at 11:38 AM:
As you say, the ATSB assumes the flight crew is unresponsive at fuel exhaustion. They have two methods of route generation, one that assumes autopilot only after the final turn with no human input, and one that allows human control prior to fuel exhaustion. So in the first case there is no human control after 18:40 and in the second case the human control ends at some unknown time before 00:15.
@Gysbreght,
In response to:
“Posted November 9, 2014 at 12:05 PM
P.P.S.
The performance boundary is not constrained by AP modes:
“The heading and speed were constrained so that all paths intersected the BTO arcs at the times of the handshakes.”
The last statement is just saying the patently obvious – that the route should be consistent with the BTO data. That tells us nothing related to performance limits.
The “performance boundary” is based on fuel exhaustion. The ATSB has never explained what assumptions were used for these maximum range calculations. I doubt these calculations involved the BTO data or the LNAV mode at all. More than likely they simply assumed a set of several speeds and altitudes, found the maximum range for each trial, drew circles with those radii, and then picked “performance boundary” points at bearings where the average speeds of the routes satisfying the satellite data happen to match the assumed speeds used in the range calculations.
Unfortunately, the ATSB does not show the performance boundary for speeds higher than those which put the end point in the ATSB search area, nor do they even identify the speeds assumed. It is possible that 9M-MRO actually flew faster and at a higher altitude than assumed in the performance boundary calculations. This could lead to increased range. Again, the ATSB has not stated what altitudes were assumed, so it is difficult to evaluate the potential errors in the “performance boundary” they present.
@Bobby: Your timing for human input is based on assumptions. Here are two comments about your reasoning:
1. The turn south just before 18:40 does not have to occur from human input. The plane could have been flying by waypoint even after the turn south.
2. The time for the last known human input is not the dividing line for possible human input. As Duncan has said many times, absence of evidence is not evidence of absence. We have no satellite ping data between 18:40 and 19:41 so we have no way of knowing whether or not there was human input or additional turns via waypoints (or perhaps even a landing or a circling).
I cannot prove your assumptions are incorrect, and perhaps your assumptions are accurate. I just don’t ascribe the same certainty to your results that you do. All of our models are fraught with assumptions.
Victor
There is one scenario for the turn at IGARI which hasn’t been talked about much, but does have some supporting facts. A “Cloud to Cloud” or “Bolt from the Blue” lightning strike could have hit MH370. Though very rare, and modern aircraft are designed to prevent such damage, it has caused significant damage to a few aircraft. In spite of all the lightning prevention technology, lightning can sometimes do some very strange damage.
“Cloud to Cloud” lightning is several times more powerful and the discharge current may last ten times longer than “Cloud to Ground” lightning. It could explain the reason for the sudden loss of communications (except for the SDU), magnetizing the cockpit radio, and resulted in toxic smoke (fire) in the plane.
Supporting facts for a lightning strike:
Weather conditions in the vicinity of MH370”s flight path were ideal… partly cloudy after the passage of a large thunderstorm in the region earlier in the day.
The pilot of another plane westbound to southern Vietnam reported “clear skies but “limited scattered lightning” in the region where MH370 was flying through.
A Malaysian Airlines plane 30 minutes ahead of MH370 attempted to contact the plane on the emergency radio channel. They hear only an inaudible, garbled voice transmission…. sounds typical of radio equipment magnetized by lightning.
The oil-rig worker in the South China Sea reported in an e-mail to his employer, that he saw a plane burning at high altitude in the western sky. It was in one piece and the “flames” lasted 10 to 15 seconds.
Other facts:
A group of 8 tuna fisherman, 10 miles off shore in South China Sea near the city of Kota Bharu, observed a very fast, low flying plane with landing lights, flying west over their location about 1:30 am. Residents on shore also reported seeing and hearing an unusually fast, low flying plane with landing lights.
A retired Boeing 777 pilot says this on Rush Limbaugh radio: (March 18, 2014)
“A fire is probably the most scariest thing to happen in flight…. Our teachings as pilots when we go through school is that once you have an uncontrolled on-board fire that you have to get that airplane on the ground within 20 minutes or less or everybody’s gonna be dead. You’ll lose control of the airplane…or everybody will be dead from asphyxiation. So you immediately start descending…”
An emergency landing on Lang Kawi Island from the south/southwest on runway 03, would have been the airport of choice for these pilots.
Long, 13,000 ft runway for a 777,
Runway lights,
Instrument Landing System (ILS),
Little/no traffic late at night,
A grass runway on the right side of the runway, in case the landing gear would not drop,
The pilots were very familiar with Lang Kawi, since Malaysian Airlines uses it for pilot training.
Although the actual flight path appears to be farther south of Lang Kawi than ideal for landing there. Flying time for the 240 nm to Lang Kawi would have been more than 20 minutes. Oxygen supplies would have been nearly depleted or totally depleted. At this point, the toxic smoke could have been affecting the consciousness level of the pilots, causing the plane to fly on west without an attempted landing by the pilots.
@dhatfield
Please provide a reference or source for claim “…sounds typical of radio equipment magnetized by lightning….”
Jeff – your reporting is some of the most thorough, detailed, and well-written journalism I’ve ever seen. This is what’s missing from news these days. Exhibit A is the number of informed posts in the few weekend days following your article.
Please keep it going!
@VictorI,
Victor, I think you have misinterpreted what I was trying to say about the the last known “human input” being the turn before 18:40. I did not mean that the plane was being flown manually for that turn. I don’t know how the plane was turned, but it did that turn only because a human made it turn. The means could have been a human at the controls or, more likely, as a result of a human entering a new bearing in the MCP or a waypoint into the FMS. I stand by my statement that a human caused the plane to turn south, and that action occurred before 18:40, and that is the last known human input (action). You are correct that a human could have caused additional turns after that time, but we don’t have any evidence that they did so.
dhatfield: uhh, as a long-serving patron of this establishment, I wish to do our noble barkeep a favor and inform you that any ruminations concerning a bolt from the blue, tuna fisherman and an oil rig operator in this forum are likely to get you swiftly escorted to the door.
You can search this site for comments posted by Alex Siew a few articles back to better understand that to which I am alluding. Likewise, if you remain intent on indulging your inner lightning, you can visit the site tmfassociates.com, which is generally more of a coffee house and less of a bar, where you will find a more relaxed, bohemian atmosphere, as well others eager to engage you on the lightning-strike scenario. The caveat is that they only serve beer and wine – and toast.
Good luck.
Matty: I hear you and would echo your concerns regarding the bias creep in terms of the search effort and our peanut gallery-driven investigation.
I would go so far as to state that, in addition to Malaysian obfuscation, I pereive confirmation bias and scientism (i.e., the misappropriate application of scientific validity testing) to be the main bugaboos with the most potential to inhibit both efforts.
On another note, it would be good to recall that Malaysia will retain authority under international convention over the recovery of human remains, as well as the remains of the aircraft, if and when any are found. This authority will extend to the flight voice and data recorders, so be prepared for additional ‘obfuscation drama’ on the part of the Malaysian authorities if and when they are found. The Malaysians will need time to analyze the FDR data for anything that could threaten their version of events (or their authority) before they release much of anything to the public, so we should prepare ourselves for months of anxious delay. The Malaysians will likely cite ‘security needs’ as well as the requirements of an ‘on-going investigation’ for the delay in announcements transmitted via Facebook, while we and, more importantly, the NOK, will be left pulling out tufts of our own hair.
The recovery of the FDR will likely not prove to be the end-all, be-all in this saga. More likely, I would imagine it being crated up for shipment to KL will only prove to be the beginning of yet another round of obfuscation, deletion, redaction and feigned ignorance on the part of the Malaysians regarding data and information concerning the flight that is in their awareness. This is why the peanut gallery must neither be dissuaded nor discouraged in its investigation of the who, what and why of MH370, no matter ‘theory fatigue’ or the threats of confirmation bias or scientism. A whistle blower, a nimble investigative journalist or the Malaysian oppostion will trump the FDR data in terms of truly cracking this singular case; I’d bet the next round of drinks on it. As for the role of the peanut gallery in all this, two words: pointing instructions.
Dr. Bobby Ulich posted November 9, 2014 at 6:31 PM :
“I chose to use 18:40 because the BFO data indicate that the aircraft was flying south by then. That means that the last “known” human input (that resulted in the southward turn) occurred sometime before 18:40, and there is no “known” human input after then. So 18:40 is the dividing line based on the available information.”
Your choice of 18:40 as the dividing line based on the BFO is arbitrary, because those same BFO’s for the remainder of the flight indicate a final course correction between 18:40 and 19:41, then a straight run due south as described in the Journal of Navigation by Inmarsat engineers.
Your understanding of ATSB’s derivation of the performance boundary is not entirely correct. The speeds are not arbitrary, the MRC speed is a given speed for each weight and altitude. The determination of the maximum range for the fuel available is not constrained by consideration of autopilot modes.
Guarded Don – “Smoke grenade” was having a little dig at the number crunchers. It just seems there are some things they don’t enjoy talking about – like debris, and the likelihood (as it stands)that the all important SDU suffered a power interruption when we know stuff was getting switched off on that plane. Don’t get me wrong, I sincerely do appreciate everything that has been done so far by volunteers.
That little issue with the SDU seems to be played down from some quarters but it’s no outlying detail(in my mind) that can just wait for the retrieval of the box. That reboot could turn out to be the ball game and I’m not game enough to bet on getting a box at all, but I’d be happily wrong.
Rand – If they retrieve a box I reckon the Malaysians will devising ways right now to lose it again.
@ Rand:
LOOOOOL.
Christ — I’m a Bohemian too, but not of the lightning variety. You mean to tell me that I’ve been in the wrong establishment all this time? And the coffee bar alternative doesn’t even serve Baileys?
As for this —
“A whistle blower, a nimble investigative journalist or the Malaysian oppostion will trump the FDR data in terms of truly cracking this singular case” —
Would like to think that the second or third options were more likely. But look at what the Malaysian government is currently doing to Anwar Ibrahim. And the dearth of real muckrakers covering aviation is nothing short of a travesty.
If only Michael Hastings were alive…
So I’ll wager on the first.
Rand: I’m weeping tears of appreciation
Unfortunately, news outlets that are picking up this story are not properly reporting the nuances. In those scrambled stories, the hypothetical scenario proposed by Jeff is attributed to Mike, who was attempting to be factual. In fact, Mike has said many times that although the ATSB has stated that the power cycling of the SATCOM is likely, there are other explanations.
There also seems to be confusion between selectively powering down the SATCOM from the EE bay and isolating the entire left bus from the cockpit, which would also disable those systems that are not redundantly protected on the right bus. Mike has tried to qualify his comments, but again the nuance is lost in the translation.
Finally, there is confusion caused by “investigators” that don’t understand the fundamental difference between deselecting the SATCOM transport in an ACARS menu from actually powering down the SATCOM, which is what Jeff was addressing. This false equivalence between the ACARS and the SATCOM has plagued the reporting from the beginning of the incident, and is now further exacerbated by incorrect investigative reporting.