Insights from New MH370 Tracking Data

Cockpit flight displays while approaching IGARI

More precise tracking data has recently become available that gives new insights about how MH370 was flown just before the transponder was disabled near the waypoint IGARI. The data was broadcast by the aircraft and received by a Malaysian ATC receiver at Terengganu. It is similar to the data that has been previously available from the aircraft tracking site FlightRadar24, except that the spacing between data points is shorter, and the data are the raw values that were actually broadcast by the aircraft. As a result, more details about the flight can be extracted.

The new tracking data was transmitted by the aircraft’s Automatic Dependent Surveillance – Broadcast (ADS-B) system, which broadcasts the GPS-derived position and altitude, as well as other parameters, as often as about every one-half of a second. Because of the inherent accuracy of GPS, the ADS-B position is more accurate than the traditional radar systems, which derive the position from the timing and angular direction of received signals from the aircraft using a rotating antenna.

A file containing the ADS-B data for MH370 is available here.

Path Near IGARI

The figure below shows that the aircraft was on a track to pass over waypoint IGARI. However, before reaching IGARI, the aircraft began to turn towards waypoint BITOD. The radius of the turn indicates that the bank angle was around 15°. The start of the turn before reaching the waypoint and the bank angle are both consistent with the aircraft following an automated route in “LNAV” mode in which the flight computers were programmed to fly “direct” to IGARI and then to BITOD. (IGARI is a compulsory “flyby” waypoint.) This suggests that at the time the transponder was disabled after 17:20:34.55, the autopilot was still engaged and the aircraft was flying in accordance with the flight plan. (The Safety Investigation Report states that the Mode S symbol dropped off the radar display at 17:20:36, which is close in timing to the last ADS-B point.) This is evidence that the deviation from the flight plan occurred after the transponder was disabled.

Comparison with Military Radar Data

The radar data that we have after the transponder was disabled consists of “primary surveillance radar” (PSR) data from civilian and military sources. Although the civilian radar data sets from the Kota Bharu and Butterworth radar sites have become available through unofficial channels, the range of the civilian primary radar sites is much less than the military primary radar sites, so that after the transponder was disabled near IGARI, only military radar captured the turnback to Malaysia. (The range of the Kota Bharu ATC radar is nominally 60 NM. IGARI is about 90 NM from Kota Bharu, and therefore not visible to the Kota Bharu ATC radar.)

Unfortunately, after many requests, the raw military data of MH370 has never been released by the Malaysian authorities. What we do have are low resolution images from official reports that depict the path of the aircraft. We also have filtered speed and track data that has been digitized from graphical data presented in a report from Australian authorities. Neither of these data sets provides the level of resolution and accuracy that would allow us to independently study the details of the path after passing IGARI, including the details of the turnback.

The figure below shows the military radar data (fuzzy yellow line) that was officially released in a low-resolution image, and enlarged here to show the path near IGARI. In the image, the bulls-eye was labeled “Last secondary radar data 1722”. For reference, the waypoints IGARI and BITOD were added to the image, as well as the ADS-B data (red) and the IGARI-BITOD route leg (black). The box (orange) around IGARI represents the much smaller area shown in the previous image.

There are some obvious discrepancies between the military data shown above and the ADS-B data shown in the previous figure. For one, the military data shows the turn towards BITOD starting after passing over IGARI. This “flyover” of IGARI is not consistent with how an aircraft following an automated path during the cruise part of the flight would turn between waypoints. If this path was actually flown, it would suggest that the navigational mode was not LNAV near IGARI.

The location of the last ATC radar point (the bulls-eye) is also different in this image from what the ADS-B data suggests. The ADS-B data shows that the transponder was disabled when MH370 was abeam IGARI. However, the image above shows the last ATC radar point occurring after the turn towards BITOD was completed. This could be because what is shown was extrapolated from the final transponder reply in what is referred to as “coasting”.

The military data also shows an impossibly sharp turn to the left occurred. Despite numerous requests, Malaysian officials have never provided an explanation for the false depiction of this turn.

These discrepancies indicate that the military data near the turnback should be used with caution. It’s possible that the radar installation that captured the turnback was Western Hill on Penang Island, and the turnback was near the maximum range of the radar site. (IGARI is about 220 NM from Western Hill.) If so, the inaccuracies might be from limitations of the military radar coverage this area. As such, the path depicted in the image may have been extrapolated from missing or inaccurate data, and should be assigned an appropriate level of uncertainty.

Finally, shown in the figure are the “entry and exit waypoints” of the turn that were supplied by the military and used by Malaysian safety investigators to study how the aircraft was flown after IGARI. As described in the Safety Investigation Report, simulations performed by the Malaysian investigative team matched the timing and position of the entry and exit waypoints of the turn only when the aircraft was manually flown with a steep bank angle of around 35°. However, considering the suspected inaccuracies in the military data, the conclusion that the turnback was manually flown should be re-visited. For instance, if the turn was begun prior to the entry waypoint, it would be possible to reach the exit waypoint at the proper time with a bank angle of 25°, which is a selectable bank angle when either of the autopilot modes “Heading Select” or “Track Select” is chosen.

Timing of Events Near IGARI

According to the Safety Investigation Report, the final radio transmission from MH370 occurred at 17:19:30. The following list shows the timing of this event along with the timing of the last three ADS-B points:

17:19:30 Last radio transmission (“Good night Malaysian Three Seven Zero”)
17:20:33.61 Last ADS-B point with altitude reported
17:20:34.15 First ADS-B point with no altitude reported
17:20:34.55 Last ADS-B point, no altitude reported

In a Boeing 777, the transponder may be disabled in the cockpit with a mode selector switch located on the pedestal between the left and right seats. The selector switch for the transponder, shown below with the label (1), would be set to standby (STBY):

During the time that the transponder was operating with altitudes reported, it was either in the switch position designated “TA/RA” for full functionality, or “TA ONLY” to suppress Resolution Alerts, or “ALT ON” to suppress both Traffic and Resolution Advisories. (TAs and RAs are part of the aircraft’s traffic collision avoidance system.) In order to select STBY from one of these three positions, it’s required to pass the intermediary position labeled “ALT OFF”. In this position, the transponder is replying to interrogations and is transmitting ADS-B data. However, there is no altitude data included in the replies and broadcasts.

Looking at the final ADS-B messages, we see that the altitude is missing for the last two messages, spanning a time of less than one-half of a second. This could mean that the intermediate switch position ALT OFF was captured as the selector switch was rotated to the standby position.

The time interval between the last radio transmission from the crew and the first message with no altitude reporting is 64 seconds. If the diversion from the flight path was caused by a third party forcing their way into the cockpit and taking control, those events would have to have occurred in 64 seconds or less. It is very unlikely that this could have been achieved by a third party in such a short amount of time.

Summary

New tracking data that has only recently become available gives us new insights as to how MH370 was flown before it disappeared from ATC radar:

  • At the time the transponder was disabled near IGARI, the ADS-B data shows a path that is consistent with normal automated flight in LNAV mode following a programmed route from IGARI to BITOD.
  • Discrepancies near IGARI between the ADS-B data and the military radar data suggest that the radar data has inaccuracies possibly because of the range limits of the radar installation on Western Hill on Penang Island.
  • The Malaysian investigators’ conclusion that the turnback after IGARI was manually flown should be re-visited in light of the demonstrated limited accuracy of the military radar data near IGARI.
  • The final ADS-B points may have captured an intermediate switch position as the transponder’s selector switch was rotated towards the standby position.
  • After the last radio transmission from MH370, the maximum time available to disable the transponder and divert the aircraft was 64 seconds. That leaves an impractically small amount of time for a third party to enter the cockpit and take control.

We now have better ADS-B data for understanding how MH370 was flown up to the point that the transponder was disabled at 17:20:35. This complements the primary radar data from Kota Bharu that starts at 17:30:33. However, we are still missing the military radar data that would cover the 10-minute gap between these two data sets. That gap includes the left turn at the start of the diversion that put MH370 on a course back over Malaysia.

It is important for Malaysia to release this closely-held military radar data so that other investigators that are working to solve this mystery can perform independent analyses of how the aircraft was flown during the turn back to Malaysia. Whatever strategic reasons there might have originally been for withholding the military data are no longer relevant more than 5 years after the disappearance.

Fellow IG members Mike Exner and Don Thompson have provided valuable comments to this article.

281 Responses to “Insights from New MH370 Tracking Data”

  1. David says:

    @Andrew, Victor. I note that the crew in the Lion penultimate flight were able to switch the trim off twice, albeit perhaps in different circumstances.

  2. oddball says:

    @Victor,

    Excellent news and report. I’d like to add a small point:

    … a third party forcing their way into the cockpit and taking control, those events would have to have occurred in 64 seconds or less. It is very unlikely that this could have been achieved by a third party in such a short amount of time.

    Presumably, irrespective of the happy coincidence of timing, a third party forcing his way into the cockpit could only do so without creating a substantial disturbance if he had been waiting outside the cockpit door for a pilot or co-pilot to come out for a comfort break. Doesn’t the door have a security viewer, such that even this could only occur if the PNF didn’t bother to check for someone lurking outside before opening it? Or, said hijacker was known or invited in. Only those, I think, offer good explanations for the precise timing. Otherwise, a break-in would be required, and that, surely, would allow time for a Mayday call?

    Even then, is the PF going to correctly and immediately disable the transponder for the convenience of a hijacker, unless said hijacker knew exactly what was required and very specifically ordered it? Otherwise, any sane pilot would leave it on.

    If I was an average hijacker, I’d want to get settled in first, and make sure that no one “touched anything” until that time. As you say, disabling the transponder must have been the first action taken, and what kind of hijacker is going to worry about that in the first <63 seconds? Only someone who knows modern passenger aviation systems very well and probably has no need to contemplate hijacking a plane, unless there is very big money involved.

    A "military style" operation / heist could have done it. But then, there's airport security and passenger screening to get around as well. For that, we'd need Hollywood, or a good CT!

    Do we yet know what the undeclared cargo actually was?

  3. TBill says:

    @Victor
    Thank you for the new higher resolution data at IGARI. Your presentation is concise and excellent, and I see no further comment on the details.

    Seems to me the perpetrator accomplished something special: completing the U-turn within Malaysian military radar coverage, such that the radar system was fooled to not sound an intruding flight alarm. Unlikely a 3rd party hijacker would be able to take over cockpit and conduct such an intricate manuever in 64 secs or less.

    Also unlikely that emergency response to a fire or other mechanical event would create a U-turn with those special properties. In combination with your recent report on the Penang fly-by, it seems obvious the most likely pilot was ZS.

    Even the ZS supporter, airline pilot Juanda Ismail, noticed that MH370’s crossing of the ILS approach to Penang occurs at a similar location and similar intercept angle to what would be flown in the course reversal procedure for ILS04. Which he thinks therefore suggests expert pilot ZS is in the cockpit to Penang.

    So the military radar data is sensitive to Malaysia because: (1) it shows how to outsmart Malaysia radar, and even though Malaysia might have fixed that loophole by now, (2) it also shows whoever was flying MH370 knew how to outsmart Malaysian military radar. The latter contradicts Malaysia’s denial that ZS or equivlent experienced pilot (are they missing any other expert pilots?) took the aircraft.

  4. Kin says:

    i am 99% suspiciously The plane want to RTB, but cannot make it, a long time ago, many media sosial, i wrote and suggested to find from last signal military radar, around above Banda Aceh – Andaman Sea and Islands, but no one search to that location until now, And They searched near Australia, i don’t know, what’s radar detected near Australia ?? If the Pilot want to go to Australia, why was not turn right, go down then straight to Australia ?? instead of turn left, go around near Banda Aceh go down then straight to Australia, wasn’t too far ? and some debris was found on Reunion Islands, slash to Banda Aceh… i speechless on that day, all big searching was focusing to near Australia … shook my head
    …..

  5. Ventus45 says:

    @Victor, @TBill

    So, someone with intricate knowledge of Western Hill PSR coverage at FL350 was required.
    Other than someone in the military, that could only have been accurately deduced by a “local” over a long period of time.
    I think it helps explain why FL350 was request at “clearance delivery”, even though FL330 had been planned and filed.

  6. Andrew says:

    @Victor

    RE: “Peter Lemme was issued a subpoena by a grand jury…”

    Yes, it was mentioned in The Seattle Times a few days ago:
    Grand jury subpoena shows sweep of criminal probe into Boeing’s 737 MAX certification

  7. david Opperman says:

    Excellent report, I would like to see the power supply wiring schematic, for this device, and the boot up sequence. It is possible that it affects the information content.

  8. Andrew says:

    @DennisW

    RE: ‘I agree that the 737 Max is not “unstable”, but it might be “uncontrollable” which is a more severe condition IMO…In the case of the Max there are indications that perhaps a combination of circumstances can result in an uncontrollable state – that is a state where the pilots have few or no available means of recovery.’

    I agree. As I mentioned previously, it seems the flight path must first be stabilised by using the electric trim to countermand the MCAS inputs, before the trim is deactivated via the cutout switches. The manual trim can then be used to trim the stabiliser as required. However, recovery might not be possible if MCAS is allowed to move the stabiliser to a significant nose low position before the cutout switches are selected, particularly at low altitude. That last point was not mentioned in Boeing’s FCOM Bulletin or the FAA’s Emergency Airworthiness Directive.

  9. Andrew says:

    @oddball

    RE: ‘Are you saying this quote from the Intelligencer article is false?

    “Because the FAA deemed the 737 Max too unstable to be used as a passenger aircraft, Boeing came up with an automated system…”’

    That statement sensationalises the extent of the problem. The 737 MAX engines tend to destabilise the aircraft while manoeuvring at high AOA, making it less stable, which is not allowed by the regulations. However, less stable is not the same as unstable. The following article has a good explanation:
    https://leehamnews.com/2019/02/08/bjorns-corner-pitch-stability-part-9/#more-29378

  10. Andrew says:

    @DennisW

    In my earlier post I said “the flight path must first be stabilised…”. I should have said “the control column forces must first be neutralised…”.

    The sentence should read:

    “As I mentioned previously, it seems the control column forces must first be neutralised by using the electric trim to countermand the MCAS inputs, before the trim is deactivated via the cutout switches.”

  11. Andrew says:

    Reuters:
    Ethiopian crew followed procedures: first official crash report

    The press have been briefed, but the report does not appear to have been published online.

  12. Mick Gilbert says:

    @Andrew

    The press have been briefed, but the report does not appear to have been published online.

    No, very disappointing. The Ethiopian authorities are not doing themselves any favours with all this shilly-shallying.

    They’ve got until next Tuesday to get the preliminary report out to accord with ICAO’s recommended timeframe. Rather than having airline and foreign ministry officials raising expectations around an early release date they would have been far better served by quietly and diligently working towards having the completed preliminary report finalised and released next week.

    Frankly, having some official say that the crew followed procedures is a little hollow until we see the FDR data and hopefully some of the CVR transcript. As Tom Sawyer said to the boy, ‘Well your saying so don’t make it so.’

    Just as importantly, what triggered MCAS this time around?

    And while we’re on preliminary reports, where’s the NTSB’s preliminary for the Atlas B767 crash?

  13. Julia Farrington says:

    https://www.bbc.co.uk/news/business-47812225. No official report released yet

    @Victor. I’ve yet to read your new blog above. I’m looking forward to it. Thank you for your continuing efforts to discover the truth about MH370.

    @oddball. Thank you for replying to my posts!

  14. Don Thompson says:

    Concerning the 737 MAX. Peter Lemme has noted that the function of the 737 stablizer trim switches changed between the NG design and the MAX design. Close comnparison of the STAB TRIM switches on NG vs MAX aisle stand shows the difference in nomenclature on the switches.

    On the MAX the STAB TRIM cut-out switches are labelled PRI (left switch) and B/U (right switch).

    On the NG and Classic series the STAB TRIM cut-out switches are labelled MAIN ELEC (left switch) and AUTO PILOT (right switch).

    My interpretation supported by FCOM description is that, on a NG/Classic aircraft, the AUTO PILOT (that is the AFDS Flight Control Computer) may be isolated/cut-out from commanding powered stablizer drive while still enabling pilot command over powered stab drive using their control wheel thumb-switches.

    Whereas the MAX is quite different. The STAB TRIM switches on the MAX only permit stab trim command from either or both Flight Control Computers (PRI or/and B/U) to be isolated/cut-out.

    My assumption for the MAX operations is that both PRI and B/U STAB TRIM switches must be switched to cut-out so as to fallback to manual winding of the STAB TRIM wheel. There is no interim fall back that isolates FCC trim drive while retaining powered control wheel stabiliser trim commands by the pilot.

    Boeing’s Nov 6, 2018 FCOM Bulletin states “Manual stabilizer trim can be used after the STAB TRIM CUTOUT switches are moved to CUTOUT. Note that this sentence states ‘switches‘.

    This change in stab trim cut-out behaviour must have been a significant contributing factor in how the pilots evaluated and reacted to the situation. If it was that the LNI or ETH pilot simply reached down and selected the right switch to cut-out, expecting (from his NG memory) CUT-OUT of the AUTO PILOT (FCC) trimming command he would have been surprised that automatic trim continued. Ergo, confusion is seeded.

  15. Julia Farrington says:

    @Victor Have now read your latest blog.

    I did not know that the transponder could be turned off in the cockpit by the method you describe. Would the co pilot be aware it was being turned off at the time or later would he notice? I am assuming the co pilot ought to be aware of even the minutest of changes to the electrical system.

    I think you and others are saying that the reason the Malaysian authorities will not release some of the military radar data is most likely because it’s not accurate (& therefore shame involved) and not because they know the data reveals facts that they do not want the world to know.

  16. TBill says:

    @Ventus45 @Victor
    “…So, someone with intricate knowledge of Western Hill PSR coverage at FL350 was required….I think it helps explain why FL350 was request at “clearance delivery”, even though FL330 had been planned and filed….”

    Thank you for bringing that altitude change to my attention. Yes I am also fixated on the higher altitude as an intentional maneuver for radar line-of-sight to keep visible on the radar.

    An alternate explanation of Victor’s observation that the altitude reporting was OFF for the last two ADS-B points, is that the aircraft was possibly starting climb at IGARI so the pilot switched the transponder to ALT OFF. I would like to know the time-stamp of the “Entry Waypoint” to see if there is slow down due to ascent. Not sure why the pilot wanted to paint the radar with altitude-less points to IGARI but we can imagine maybe the co-pilot made a speedy exit from the cockpit, and ZS was immediately free to make his move.

  17. Victor Iannello says:

    @ventus45: Why would a higher flight level be desirable if the objective was to hide the turnback after IGARI?

  18. sk999 says:

    The last ADS-B point is approximately 1 second after MH370 passed abeam of IGARI. If one were viewing the CDU ACT RTE LEGS page, my understanding is that the IGARI waypoint would have just scrolled off the top of the screen.

  19. Victor Iannello says:

    @sk999: Exactly. My guess is that the transponder was switched off just as BITOD became the active waypoint on the legs page as well as on the navigational display.

  20. paul smithson says:

    Thank you, Victor and contributors for an excellent post. A few quick observations and questions.

    1. A 15 degree bank turn from 027 to 058 at ~480kt TAS should take about 36 seconds. The aircraft was a little over half way through this turn at the time this datastream ceases, with no indication at that point that the turn was anything but ongoing as planned. About 16s later it would have completed the turn on to 058/059 on the IGARI-BITOD leg.
    2. Would the datastream described above indicate if any of the following had occurred: change in navigation mode, disengagement of A/P, change in commanded speed, change in alt, error messages from avionics systems?
    3. Can you confirm if any of the above did occur, or indeed any other sign of something being awry (other than alt=0) before the datastream ceases?

  21. paul smithson says:

    one more, if I may:
    Does the datastream also give us speed readout (any of mach/TAS/CAS/GS) up to time of disappearance?

  22. Victor Iannello says:

    @Paul Smithson: Very good questions. I don’t know the answer. Don has been working on that. Perhaps he will comment on the status of that work.

  23. Victor Iannello says:

    @Julia Farrington: There would be a message appearing on the navigational displays of both the Captain and First Officer when TCAS was turned off.

  24. paul smithson says:

    Thanks @Victor. I look forward to hearing from Don. I see that the file you have shared does periodically provide velocity x/y, presumably groundspeed, which translates into 470.9kt and 471.3kt for the last two datapoints that included these fields. The preceding readings indicate a very slight reduction in GS from 474->473->472->471 that is consistent with expected increase in headwind as the track angle changes to windward.

  25. Victor Iannello says:

    @Paul Smithson: Yes, those speeds are ground-referenced. The groundspeed may be calculated from sqrt(Vx^2+Vy^2), and the track may be calculated from atan(Vx/Vy), or for users of Excel that wish to place the angle in the correct quadrant, atan2(Vy,Vx).

  26. TBill says:

    @Victor @sk999
    Re: My guess is that the transponder was switched off just as BITOD became the active waypoint on the legs page as well as on the navigational display.

    That is not how MicroSoft flight sim works…not sure about real B777. In FS9, the Waypoint advances exactly when the turn starts, so well ahead of Waypoint depending, on degrees of turn. I am thinking a real B777 anticipates turns a little better than flight sim, but for these gentle turns, flight sim seems pretty good match to reality.

    It is interesting that ADS-B goes off abeam IGARI, but I was not thinking that is when BITOD shows up on the screen, unless there was little delay on the part of the pilot.

  27. Victor Iannello says:

    @TBill: In the PMDG 777 model, the transition of the active waypoint from IGARI to BITOD occurs when IGARI is almost exactly abeam, and definitely not at the start of the turn. I can’t be sure of what occurs in the real world, but I would trust the PMDG 777 model over the PSS 777 model, and it also makes more sense.

  28. TBill says:

    @Victor
    OK I’ll check for a difference between FSX/PMDG and FS9/PSS777 when I get the chance. I just happen to have PSS777 saved case right there.

  29. Victor Iannello says:

    The Preliminary Report for ET302 is now available.

    Many interesting things to note, including anomalous behavior while the autopilot was engaged during the initial climb.

  30. Greg says:

    @Victor

    “The time interval between the last radio transmission from the crew and the first message with no altitude reporting is 64 seconds. If the diversion from the flight path was caused by a third party forcing their way into the cockpit and taking control, those events would have to have occurred in 64 seconds or less. It is very unlikely that this could have been achieved by a third party in such a short amount of time.”

    The third-party hijacker could have been in the cockpit for some time before IGARI. He could terrorize the pilots (pilot) and force them to execute his commands. The best moment to take control over the cockpit was the period of time just after reaching the cruising level (pilot’s visit to the toilet, drinks and meals delivered to the pilots by the cabin crew). This could happen between the first report to the ATC on maintaining FL 350 and its unreasonable repetition. This second FL350 report might have been under the control of the hijacker just like the last “good night …” message. In both transmissions, the pilot could deliberately apply deviations from the communication procedure to draw ATC attention. But the way he did it suggests the hijacker’s great aeronautical knowledge. When the FMC changed the next WPT from IGARI to BITOD the hijacker started his flight plan and the crew could somehow react in defense.

    This is not my favorite scenario. I do not have such. It’s just an option, like many others, which still can not be ruled out at this stage

  31. Victor Iannello says:

    @Greg: Yes, I suppose an earlier take-over is possible, and things were not as normal as they seemed before the transponder was disabled. The flight path and ATC exchanges seemed normal, but I understand your point.

  32. David says:

    @Andrew. “…recovery might not be possible if MCAS is allowed to move the stabiliser to a significant nose low position before the cutout switches are selected, particularly at low altitude. That last point was not mentioned in Boeing’s FCOM Bulletin or the FAA’s Emergency Airworthiness Directive.”

    Does all this not have implications for earlier 737 models in the event of a trim runaway nose down at low level?

  33. TBill says:

    @Greg
    I understand too, that is what the ZS supporters are saying, even the experienced pilot above says it sure looks like ZS was the pilot flying to Penang, so there must have been a hijacker holding him hostage. But ZS had the option to send hijack code, and why would ZS have to conduct a perfect escape that faked out Malaysian military radar coverage? There are questions about the radio transmissions and if we are getting the whole true transcript to IGARI. But I am thinking the purpose of any transcript non-disclosure was to hide the truth, which I am glad to deal with whatever the truth is.

  34. Ventus45 says:

    @Victor

    My postulate is basically this.

    Zs knew full well, that the return over the peninsula would be visible to WH-PSR, regardless of his return path cruise altitude.
    He was “only hiding” from civil ATC – and “ONLY” them, – and “NOT” from the RMAF.
    Specifically, he NEEDED to remain visible to the RMAF, so that they would know it was him, and NOT a “new” PSR target.

    The requirements were twofold.

    First, to be “forgotten” by KL ATC after the hand-off to Ho Chi Min. (Achieved)
    Second, to REMAIN VISIBLE to RMAF, at FL350, so as to make it look like an emergency turn-back.
    RMAF would assume it was still in contact with KL ATC – and do nothing. (Achieved)
    Remember H2O’s comment on “4 Corners” ? – “It was from our airspace”.

    Now, look again at those plots I sent you yesterday.

    If he had have been lower, at FL330, the plan would not have worked.
    There would have been a significant time gap, since he would have dropped off the WH-PSR display minutes earlier, so by the time he would have “reappeared” on WH-PSR (post turn-back) it would have been treated by the RMAF as a “new target”, and ACTION TAKEN, which he did not want !

    In other words, FL350 was vital to “put the radar operator to sleep” – so to speak.
    He did.
    He simply outfoxed them.

  35. Andrew says:

    @David

    RE: “Does all this not have implications for earlier 737 models in the event of a trim runaway nose down at low level?”

    Quite possibly, but I think we also need to remember the B737 has been in production for over 50 years in various guises. During that period over 10,000 B737s have been delivered and the B737 family has accumulated more than 300 million flying hours. To my knowledge, older versions of the aircraft have had very few runaway stabiliser events, so the risk seems to be very low. Of course that doesn’t mean it can’t happen to an older aircraft.

  36. Don Thompson says:

    Paul,

    To expand on Victor’s response to your questions:

    2,3,4: the ADS-B surveillance data, as do extended squitter Mode-S replies, indicates certain avionics derived states (but not error msgs). At this time, apart from the alt reverting to 0, there’s nothing apparent that helps explain anything that is about to occur as the transponder ceased. Mach, TAS, and GS are exposed, we’re continuing to extract & assemble data.

  37. David says:

    @Andrew. Airworthiness of earlier 737 models. Thanks. While the risk is demonstrably low, I would have expected the FAA to think about a re-visit for those others, that is if there is an unaddressed hypothetical now apparent which breaches certification requirements.

  38. Andrew says:

    I think they will re-visit the problem if it is found to have been a factor in these accidents. If nothing else, I expect the issue will receive much more attention during pilot training and be prominently highlighted in the manuals.

  39. Andrew says:

    A few observations from the ET302 Preliminary Report:

    1. The L AOA output diverted to around 75° immediately after lift-off. The L stick shaker activated and remained continuously activated until the last 15 seconds of flight.

    2. The Master Caution activated shortly after lift-off and the FO called “Master Caution Anti-ice”, presumably in response to illuminated Master Caution and Anti Ice lights on the system annunciator panel, located on the glareshield. This seems to have been associated with the failure of the L AOA Heat, which occurred about the same time. Later in the flight, both pilots called “left alpha vane”, probably in response to an illuminated probe heat light, located on the probe heat panel on the forward overhead panel.

    3. There were several automatic nose down trim commands while the AP was engaged, which is normal AP behaviour in response to increasing speed.

    4. The AP disengaged about the same time the flaps retracted, at which time the first of three large automatic nose down trim commands occurred. There was a change in pitch trim with each of the first two commands, but no change with the third.

    5. The Capt responded to the first two automatic nose down trim commands by trimming nose up with the electric trim. However, the nose up trim inputs did not completely remove the previous automatic nose down inputs and he continued to hold the control column rearwards to keep the nose up.

    6. The lack of trim change following the third automatic nose down input seems to indicate the STAB TRIM CUTOUT switches were selected to CUTOUT some time between the Capt’s second nose up trim input and the third automatic nose down command.

    7. There were two more very brief nose up trim inputs towards the end of the flight, which seems to indicate the STAB TRIM CUTOUT switches were selected back to AUTO. The nose up trim commands had very little effect on the pitch trim.

    8. The nose up trim commands were followed by a final automatic nose down command, which caused a considerable change in the pitch trim.

    9. There were no further changes in pitch trim following the final nose down command. Based on Victor’s earlier comment, the STAB TRIM CUTOUT switches might subsequently have been re-selected to CUTOUT.

    Further thoughts:

    1. The extreme divergence of the L AOA output and failure of the L AOA heat might indicate physical damage to the L AOA vane during lift-off, possibly due to a bird strike.

    2. The aircraft system responses to the L AOA output are almost identical to the JT610 accident, ie stick shaker, altitude/airspeed indication disagreements, MCAS activation.

    3. The Capt did not completely remove the automatic nose down trim inputs and continued to hold the control column rearward. As previously discussed, the upward deflection of the elevator might have made manual trimming impossible, particularly at the very high speed the aircraft was flying (VMO or above) after the stab trim was deactivated. Recovery might have been possible if the speed had remained at or below 250 KIAS and not increased to VMO. However, having accelerated to such a high speed, it’s unlikely the crew could have recovered unless they climbed to a much higher altitude, which in itself might have been difficult.

  40. Pilatus says:

    @Andrew,

    What’s your take on why the column cut-out switches did not arrest stabiliser movement?

  41. Andrew says:

    @Pilatus

    I haven’t flown the B737, but my understanding is the control column cut-out switches do not affect the MCAS operation. See the following PPRuNe post by FCeng84:
    https://www.pprune.org/rumours-news/619272-ethiopian-airliner-down-africa-44.html#post10415935

  42. Andrew says:

    @Ventus45

    RE: ‘I think it helps explain why FL350 was request at “clearance delivery”, even though FL330 had been planned and filed.’

    Nice theory, but may I suggest a far less clandestine explanation? Pilots can’t always flight plan their optimum cruise levels due to agreements between the ATC providers in adjacent FIRs. Those agreements dictate the levels that an aircraft can plan to fly on a particular air route. However, that doesn’t stop a pilot from requesting an alternative level from ATC once the flight plan is in the system. If the traffic situation allows, ATC will normally accommodate such a request.

    In the MH370 case, FL350 was much closer to the optimum level for the aircraft’s weight. I would suggest the pilot requested that level simply because it was more efficient than the planned level.

  43. DennisW says:

    @Andrew

    Things are not looking good for Boeing. The flight crews on the MAX crashes seem to be in the blameless category.

    @Victor

    I have difficulty with the IGARI radar comments and analytics. The plot of the military radar is obviously “derived” from the radar data and not at all representative of raw data. Who knows how that graphic was created?

  44. Andrew says:

    @DennisW

    I suspect the EK521 Final Report, when it is eventually released, will add fuel to the inferno.

  45. TBill says:

    @Victor
    OK yes the PMDG777 makes the change to BITOD abeam IGARI, whereas PSS77 makes the BITOD change earlier. So yes if PMDG is true B777 behavior, then it looks like transponder was turned off the exact second BITOD was first showing active. Parenthetically, why would a fire or 3rd party hijack happen at that moment? It looks more like the pilot was focus on the screen to see the change to BITOD waypoint.

    Both PMDG and PSS777 seem to overshoot the IGARI-to-BITOD line a little bit and then then they both curve gently back to the trend line, so the end of the turn is not quite as perfect as you show above. If the real B777 also overshoots a little bit, it might explain why the “entry waypoint” is a little offset from the exact line to BITOD.

  46. Julia Farrington says:

    @Victor. Thanks for info on the transponder.
    @Don and @ Mike Exner. I should have included you both in my appreciative comments about Victor’s blog. So thank you too for your tireless efforts to understand the technical aspects of flight of MH370.

  47. Julia Farrington says:

    I fly regularly on the older 737’s 800 series and am always thankful when I board these days to see the engines further back (than the Max8) and looking a whole lot lighter. They may not be as fuel efficient as the max but that doesn’t concern me in the light of recent tragic events. I really wish I could contribute to this blog in technical speak and my apologies for my simplistic observations but I heard yesterday that the red light/ sensor visible on the exterior at the front of the MAX failed to correct the nose dive as it was supposed to do in the new instructions to pilots so I was lead to believe that however expert the pilots are, this fault is outside anyone’s control at the moment. What I fail to understand though is that the MAX have been flown by Norwegian Air, Tui and American without fatal crashes. However it would be interesting to know how many non fatal incidents there have been with each airline with the anti stalling mechanism.

  48. Mick Gilbert says:

    @Andrew
    @DennisW

    Andrew, I suspect that the EK521 report will splash a bit of petrol up the FAA’s leg, given their steadfast refusal to do anything about autothrottle training after the Asiana 214 prang (you’ve only got to look at the history of NTSB Safety Recommendation A-14-037 to see the NTSB’s palpable frustration with the FAA). Given how close the FAA finds itself to the Boeing bonfire that’s likely to cause a bit more ‘unpleasantness’ all around.

    Did you have anything specific in mind when you suggested that Boeing will cop some stick from EK521?

    Dennis,

    Are Boeing buying their own stock to put a floor under it? I can think of no other reason for its current gravity defying feats.

    Now, if they could get the MAX to maintain altitude like that …

  49. Victor Iannello says:

    @Mick Gilbert: The lift in the Boeing stock price is probably related to some positive developments in US-China trade talks.

  50. Victor Iannello says:

    @DennisW: Many of us would like the raw military radar data so we don’t have to rely on a fuzzy line that is physically implausible. The new ADS-B data demonstrates that the path based on the military data that has been provided does not match the ADS-B data near IGARI, which calls into question the accuracy of the military data during the time interval when no other source of data exists.

  51. Victor Iannello says:

    @Andrew: I was not aware that the AoA sensor that is used to trigger MCAS alternates from left to right for each flight. That means that one flight that experiences a speed and altitude disagreement due to a faulty AoA reading might result in erroneous triggering of MCAS on the next one. I’m not sure of the logic behind that design choice as it would make troubleshooting more difficult.

  52. DennisW says:

    @Mick

    I am also surprised by the BA stock performance. I would not touch the stuff.

    https://articles.marketrealist.com/2019/04/analysts-stay-bullish-on-boeing-amid-737-max-crisis/

  53. Greg says:

    @All – Question

    In the case of MH17, the passenger oxygen system has not been activated.

    „While the sudden depressurisation of the fuselage should have triggered the deployment of masks, the inquiry found that this was prevented by the loss of electrical power.”(FlightGlobal)

    Does anyone know how the B777 pax-ox system is powered and controlled?

    It is obvious that a system based on chemical generators can not be manually turned off or blocked.

    Is the deployment of oxygen masks the only indicator of depressurisation for passengers and cabin crew (disregarding physical sensations)?

  54. TBill says:

    @Greg
    Very good question re: O2 masks.
    So far MH370 discussions have said there is no cockpit control over the O2 mask drop down, but there are circuit breakers in MEC Bay.

    I don’t know if it might be possible for the pilot to disable the O2 masks by manually cutting all power to Left and Right buses.

    My personal theory is that the pilot would have wanted to suppress the O2 masks, because if intercepted, he would not want the fighter pilots to see the “spaghetti”. I am thinking all power may be off around IGARI to stop black box data recortding as well as O2 masks (if that is possible).

  55. ArthurC says:

    Can the FDR and CVR be disabled?
    If that’s the case, even if they are found, there wouldn’t be anything useful on them to solve the mystery…

  56. TBill says:

    @ArthurC
    My understanding yes the Black Boxes can be cut off, by circuit breakers in the MEC Bay, or by cutting off electric circuits in the cockpit, which is what I am thinking.

    But to paraphrase legendary crash expert Greg Feith, even if the black boxes were disabled, it will be interesting and valuable data.

  57. Andrew says:

    @Victor

    RE: “That means that one flight that experiences a speed and altitude disagreement due to a faulty AoA reading might result in erroneous triggering of MCAS on the next one. “

    Yes, I assume the speed/altitude disagreements would occur on successive flights due to the difference between the two AOA outputs, but erroneous MCAS activation would only occur on alternating flights where MCAS is using the faulty AOA output. The 737 MAX Onboard Maintenance System can probably determine which sensor is erroneous, depending on the fault.

  58. Andrew says:

    @Greg
    @TBill

    RE: “Does anyone know how the B777 pax-ox system is powered and controlled?”

    There are two conditions that will cause the release of the passenger oxygen masks:
    – The passenger oxygen switch on the overhead panel in the cockpit is selected to ON, or
    – Two of three cabin pressure sensors indicate the cabin pressure altitude is above 13,500 ft while the aircraft is airborne with an airspeed of more than 80 knots.

    If either of those conditions occurs, a relay in the standby power management panel activates the latches on the passenger oxygen boxes and the oxygen masks are released. The system needs 115V AC power from the AC Standby Bus and 28V DC power from the Capt Flt Inst Bus to operate. The chemical oxygen generators do not start producing oxygen until a passenger or crew member pulls on an associated oxygen mask.

    Several other events occur in the passenger compartment when the oxygen system is activated:
    – Aural alert
    – Passenger compartment lights illuminate full bright
    – No Smoking and Fasten Seat Belt signs illuminate
    – PA volume increases to maximum

    In the MH17 case, the destruction of the forward part of the fuselage would have caused the immediate loss of electrical power and rendered the passenger oxygen system inoperative.

    Manual deployment of the passenger oxygen masks can be initiated from the cockpit, but automatic deployment cannot be prevented.

  59. Tim says:

    Thanks Andrew,
    That’s interesting. This is the first time I’ve considered that the pax O2 system may have failed as it’s powered from the AC stby bus.

    As I believe that bus was taken out by a rupturing O2 bottle that now means the pax never had any drop down O2 after the decompression.

    And am I the only one on this site who does no support a nefarious pilot?

  60. DennisW says:

    @Victor

    @Mick Gilbert: The lift in the Boeing stock price is probably related to some positive developments in US-China trade talks.

    Analysts make the trade talk claim. However, just for fun I did a dscounted cash flow calculation (I have it automated in Goggle Sheets), and came up with ~$600/share. The company “has been” generating an enormous amount of cash. Certainly the 737MAX events will impact that cash flow, but it is anyone’s guess how much or for how long. Based on historical financials the company is significantly undervalued at ~$400/share. At $400/share the $8.22/share dividend represents ~2% which is better than leaving your money in a bank.

    Still, you can stick solar panels on your roof and get an 8% risk free rate of return (in Cali), and feel pious about it. Maybe I am too risk averse in the winter of my years.

  61. DennisW says:

    @Tim

    And am I the only one on this site who does no support a nefarious pilot?

    As far as I know.

    Off site, the Malaysian Transport Minister sympathizes with your position, but I think he is motivated by liabilty concerns. Are you with Alianz SE?

  62. Victor Iannello says:

    @Tim: I think most here believe a deliberate diversion is by far most likely, but last I checked you have @Mick Gilbert in your corner on favoring an accident scenario.

  63. Greg says:

    RE: “Does anyone know how the B777 pax-ox system is powered and controlled?”

    @Andrew
    @TBill
    Thank You for response

    @Andrew
    If I have understood correctly, these 4 events occur as part of system activation. Are there any events indicating depressurisation independent of automatic deployment? Will any events occur if only one of the electrical power sources is lost: 115 AC or 28 DC?

    @Tim
    I do not support anything, the matter is open.

  64. Ventus45 says:

    @Andrew
    Routine request for “optimum altitude” for the weight is noted, but it still seems to me that FL350 was “required” for the seduction of the RMAF, “if” ZS really was the perp.

    @Tim
    “And am I the only one on this site who does no support a nefarious pilot ?”

    I still feel that a genuine emergency is an “outside chance”.
    One of my original theories, was based on crew O2 bottle failure/rupture, with hull rupture, and an attempted turn-back/return to KL via UPRON (recently overflown), but, like in the CALI disaster, with UPROB being selected in error, with crew incapacitation soon after.
    https://auntypru.com/forum/showthread.php?tid=28&pid=4646#pid4646

  65. paul smithson says:

    @Tim you can put me down for that under-populated corner too.

  66. Hank McGlynn says:

    The 737 has a large manual trim wheel on both sides of the center console. My understanding is that it takes about 250 turns to move the stabilizer through its full 17 degree range. It has a pull out handle if the trim needs to be changed rapidly – smaller changes can be made by just pushing or pulling the top of the wheel. The mechanical advantage is huge, so it is easy for the pilot to just grab the wheel and stall the electric trim motor. This trips MCAS off – at least temporarily. Beeping the column electric trim also disengages the motor, at least temporarily.

    On 302 it looks as if the pilots toggled both the autopilot and manual electric trim buttons off. They are used to using the yoke buttons to trim the aircraft. When these are used the manual wheels spin like fans. It seems that the crew couldn’t move the manual trim wheel fast enough to bring the nose up and maybe didn’t pull out the handle and turn like crazy. They turned both of the electric trim toggles back on. I think if they only used the left toggle (trim button) it might have been OK. I think MCAS goes with the autopilot by way of the right toggle. But they are instructed to turn off both toggles by Boeing. They apparently needed the high speed electric trim column buttons to get the nose up and this allowed MCAS to engage and push it down at the end.

    The FDR showed that when the flaps were retracted at 05:40:00 there was a FCC (MCAS) nose down trim causing pitch down and a second FCC pitch down at 05:40:21. There was a FCC command at 05:40:42 but no pitch change because toggle may have been turned off just before. But there was no electric trim and pitch trim was flat until 05:42:49 when the toggles may have been reengaged. It is surprising that the trim was not changed manually from 05:40:45 to 05:43:21 when FCC/MCAS engaged. Maybe it was too slow, but trim never change even slightly – but 2 seconds is not a lot of time to figure it out. They never recovered after the last MCAS engagement and pitch attitude just plunged.

    I think that switching the right (FCC) toggle disconnects the FCC/MCAS input to the trim motor. The left toggle I think cuts out the yoke switches, which may be needed for a runaway trim, but maybe should have been reengaged without the A/P toggle. Maybe Boeing should have encouraged engaging the left toggle.

  67. Niels says:

    @Tim

    “And am I the only one on this site who does no support a nefarious pilot?”

    I’m among those here who prefer to leave all options open until we have more complete and accurate information available.

  68. Niels says:

    Here is a link to a pdf listing the main expressions used in my path reconstruction tool.

    https://www.dropbox.com/s/u4hwmboa1wfcxnb/WGS84PathModel4.pdf?dl=0

    Comments, questions and suggestions are welcomed. I intend to implement the model in Matlab (now calculations are carried out in Excel) as it would help to automate certain procedures needed for estimation of possible “error range”.

  69. Don Thompson says:

    @Hank,

    I wrote a comment earlier concerning the STAB TRIM cut-out switches.

    Do check your interpretation against the nomenclature on the switches on NG series aircraft vs MAX series aircraft.

    The MAX the STAB TRIM cut-out switches are labelled PRI (left switch) and B/U (right switch).

    The NG and Classic series the STAB TRIM cut-out switches are labelled MAIN ELEC (left switch) and AUTO PILOT (right switch).

    The MAX switch operation does not appear to allow the control wheel switches to command trim while the autopilot is isolated.

  70. Mick Gilbert says:

    @Tim

    And am I the only one on this site who does no support a nefarious pilot?

    I’m not a fan of this notion of ‘camps’ or ‘corners’. That sort of binary thinking generates a lot heat but does little by way of advancing the discussion. All it does is entrench thinking more firmly.

    I’ve never said that it wasn’t deliberate malicious action, it may well have been. But it may not have been deliberate malicious action. I try to maintain an open mind on the matter.

    Someone (I thought it was Norman Augustine but haven’t been able to find a proper attribution) once said something to the effect that,

    As aircraft complexity increases, so too the cause of aircraft accidents will become more complex but at no point will an accident be so complex that it couldn’t be blamed on the pilot.’

  71. Hank says:

    @Don Thompson

    I checked and you are correct. The labels are different. There was some value in being able to separately disengage the yoke trim buttons versus the FCC. I saw a wiring diagram for earlier 737 and it looked like the yoke buttons were routed through one and the A/P by the other. I suppose two buttons in series ensures cutoff. Maybe Boeing figured the A/P could be switched off and not need a switch. Not true for MCAS. I don’t suppose Boeing told NG pilots about the change?

  72. DennisW says:

    @Mick

    A saying often attributed to Mark Twain – “Climate is what you predict. Weather is what you get.”

    To the people here who insist on remaining open – don’t pursue a career as a meterologist. You won’t get by with telling me you don’t have an opinion about the weather because of the complexities.

    When my boss would ask for a quarterly revenue forecast I could not get by with – “I don’t know, a lot of things could happen”. FWIW I think I have an open mind as well, but that does not rule out having an opinion and expressing it.

  73. TBill says:

    @Andrew
    Re: O2 Masks
    Thank you…sounds like you are saying there are not even MEC Bay circuit breakers that could stop the O2 mask drop down.

    @Ventus45 @Tim
    I am not sure MH370 composite O2 cylinders have much potential for rupture, but even if it did, I presume the cylinder hold down brackets are robust enough to prevent the cylinder from becoming a projectile/missile, which is the main problem. Presumably the historic Qantas failure (which was a metal style cylinder) was really two failures in one: defective cylinder and poor design of cylinder brackets?

  74. TBill says:

    @Andrew
    Re: O2 Masks
    Thank you…sounds like you are saying there are not even MEC Bay circuit breakers that could stop the O2 mask drop down.

    @Ventus45 @Tim
    I am not sure MH370 composite O2 cylinders have much potential for rupture, but even if it did, I presume the cylinder hold down brackets are robust enough to prevent the cylinder from becoming a projectile/missile, which is the main problem. Presumably the historic Qantas failure (which was a metal style cylinder) was really two failures in one: defective cylinder and poor design of cylinder brackets?

  75. Mick Gilbert says:

    @DennisW

    When my boss would ask for a quarterly revenue forecast I could not get by with – “I don’t know, a lot of things could happen”.

    Dennis,

    I’m sure you could not, and neither could I under not dissimilar circumstances.

    The difference, of course, was that in those previous roles we generally had a reasonable sufficiency of fit-for-purpose information such that we could make those quarterly revenue forecasts or similar within the accepted bounds of accuracy. And, generally speaking, we’d be presenting such forecasts as a range with some sensitivity analysis thrown in so as to provide as complete a picture as possible. We weren’t saying ‘I don’t know‘ but we were (or should have been) saying ‘a lot of things could happenand here’s how they will impact the forecast if they do. That was certainly the generally accepted way of doing things in my world, anyway.

    I have no problem wading in and expressing my opinion when and as required (and sometimes when it’s not required). What tends to get on my goat are the sweeping unconditionally declarative statements (the ‘it must have been this‘, ‘it could only be that‘ or ‘this proves that‘ stuff) when the evidence that underpins them is conjectural. On that stuff, I will wade in. But after five years of this, frankly, my approach now is that if the discussion has no bearing on refining a search strategy, then I’m more likely to save my keystrokes.

  76. Andrew says:

    @Greg

    RE: “If I have understood correctly, these 4 events occur as part of system activation. Are there any events indicating depressurisation independent of automatic deployment? Will any events occur if only one of the electrical power sources is lost: 115 AC or 28 DC?”

    Yes, when the deployment relays energise due to either cabin alt >13,500 ft or passenger oxygen switch on, the electrical load management system (ELMS) sends a signal to the cabin services system (CSS), causing those four events. There are no other system events to alert the cabin crew about a depressurisation.

    The four system events are controlled by the CSS, which has many different interfaces and power sources for different parts of the system. I haven’t had time to check what, if anything, would be affected by a power loss, but most emergency-related systems are powered via the standby electrical system. The standby system has a huge amount of redundancy, including the main battery, which can also power a standby inverter fo AC power. The four events probably wouldn’t be affected unless there was a massive power loss. The passenger oxygen system itself is also powered by the standby electrical system and would not be affected by anything other than a massive failure.

    @TBill

    RE: “…sounds like you are saying there are not even MEC Bay circuit breakers that could stop the O2 mask drop down.”

    The functional description in the AMM/Training Manual does show breakers for the passenger oxygen system in the MEC. There are two channels (A & B) for redundancy. Each channel uses 28V DC to control its respective oxygen deployment relay and 115V AC to unlatch the passenger oxygen boxes, four circuit breakers in total.

  77. DennisW says:

    @Mick

    Well said above.

  78. Andrew says:

    RE: “Did you have anything specific in mind when you suggested that Boeing will cop some stick from EK521?”

    It seems the cause will ultimately come down to a failure to increase thrust during the go-around. HOWEVER, in my view there are significant documentation and training deficiencies related to the TO/GA inhibit function that were no fault of the crew.

    The B777 FCOM go-around procedure is a generic procedure that is meant to cover go-arounds before and after touchdown. The problem, however, is that the automation behaves quite differently in the go-around after touchdown case and the pilot MUST, amongst other things, manually advance the thrust levers to set the required thrust.

    To this day, the differences between the two go-around cases are not highlighted in the standard go-around procedure. The company where I work recognised this hazard over 15 years ago and developed a rejected landing procedure for the go-around after touchdown case. The issue was discussed with Boeing and I believe Boeing issued a notice of ‘No Technical Objection’ at the time. The rejected landing procedure is included in our FCOM and our B777 pilots receive periodic training in the simulator; however, the procedure is not included in the standard Boeing FCOM. Discussions with other airlines’ B777 pilots following the EK521 accident showed that many were not aware of the TO/GA mode inhibit and had never received training for a rejected landing.

  79. Andrew says:

    @Mick Gilbert

    My previous comment was for you!

  80. Mick Gilbert says:

    @Andrew

    Yes, got it, thank you.

    That is interesting that Boeing don’t differentiate between the two go-around cases despite the fact that their software treats the two cases differently. I’m mean it’s not as though anyone was looking to save paper when they wrote the B777 FCOM. Even more interesting that not all operators were aware of the difference. You would think that Boeing might have seen fit to share the procedure that your airline developed with other operators.

    There seems to be an element of ‘“written in stone” never to amended‘ to some of Boeing’s decision making.

  81. Andrew says:

    @Mick Gilbert

    I think it comes back to Boeing’s philosophy that automation is a tool that assists, but doesn’t replace the pilot. Consequently, Boeing’s procedural design very much relies on pilots doing what pilots are supposed to do, ie FLY THE AIRCRAFT. That’s well and good, but in the real world pilots are becoming more and more reliant on automation and when the automation does something unexpected, pilots are sometimes caught out. There is a case for pilots to be given far more manual handling practice with all the automation turned off, but some airlines are loath to allow it during line flying and insist on the maximum use of automation wherever possible.

  82. Mick Gilbert says:

    @Andrew

    Yes, it is becoming apparent that rather than there being a ‘shared’ view that the pilots and the aircraft are part of one integrated system there’s almost a dichotomous mindset at both the airlines and the manufacturer(s). There seems to be a view that the safe and effective operation of the aircraft is managed either by the application of automation or by the application of flight crew skills. The more of one, ostensibly the less is required of the other.

    And given that there are costs associated with each side of that equation that are born by either party – the manufacturers for aircraft systems and the airlines for training – there’s an underlying incentive to try and kick ‘problems’ over the fence. As Charlie Munger says ‘Show me the incentive and I’ll show you the outcome‘.

    This whole MCAS mess is perhaps as good example of this. Boeing seemed to take the approach that they could kick the issue of dealing with the deficiencies of their system over the fence to the airlines/flight crews to deal with; their approach was ‘Well, if there’s an erroneous activation then the flight crew will catch this‘. To my mind, had there have been an ‘integrated systems‘ approach you would have seen far, far better modelling and simulation of erroneous MCAS activations and far, far better testing of the ‘the flight crew will catch this‘ assumption.

    The fact that Boeing were obviously motivated to keep (publicised) differences between the MAX and the NG to a minimum clearly didn’t help. In fact, it almost guaranteed that there would be no open discourse between Boeing and the airlines on the topic. Good old Charlie
    incentive » outcome.

  83. Andrew says:

    @Mick Gilbert

    I don’t think that Boeing sees it as ‘kicking deficiencies over the fence’. Under the Boeing philosophy, the pilot is ultimately in charge, not the automation. Accordingly, if something goes wrong or doesn’t go to plan, the pilot is expected to take the appropriate action to save the day, ie FLY THE AIRCRAFT. However, that assumption relies upon pilots having a good understanding of their aircraft and knowing what to do. Those conditions are difficult to satisfy when the documentation and/or training is deficient.

  84. Trip says:

    @all
    Does anyone find Boeing’s responses to 777 and 737 Max an interesting comparison. In each case 2 planes were lost. Boeing immediately apologized for the 737 and is working to correct the software. Boeing obviously concluded that no such situation existed with the 777. It would seem Boeing would at least make a public statement. MH17 was shot down. What did Boeing conclude about the MH370? I would certainly feel more comfortable if I knew that Boeing had addressed some of the scenarios identified in various blog posts. Maybe lock or alarm the the e/e bay access. Reconfigure the auto transmitters for water landing. Extend the battery life and power for black boxes. Require location information for sat transissions. Require responses for pax background checks. Have “always on” transponders. Etc. Boeing’s response to MH370 indicates they felt whatever happened was man made and did not require any public acknowledgement. Maybe they have already made some of these changes. And maybe they already know the cause.

  85. TBill says:

    @Trip
    Yes definitely I see parallel to MH370. In the case of MH370 it is not a design flaw but error of omission letting pilot turn off transponder secretly, etc. Three 9/11 planes did that too. One wonders if NTSB is sugar coating the Atlas Air accident to avoid saying intentional grounding in that case.

    I do not think Boeing wanted the public to understand MH370, because then the public would be asking, why do we still do that? Now Boeing is going to be more vulnerable, previously they could say- our safety record puts us above even having to listen to what any critics say. And I suspect Airbus will be more responsive to MH370 cockpit security concerns.

    All of the MH370 conspiracy theorists, media mystery hypers, and hijacking naysayers have greatly aided obscuring the apparent truth of MH370 from the public. So Boeing is off the hook there. That’s why I liked last years Aussie 60 Minutes show, it told it like it should be told (though we can debate LVance content as usual). We did not get that show in the USA (and I asked USA 60 Minutes for it).

  86. Don Thompson says:

    @Andrew & @Mick

    Under the Boeing philosophy, the pilot is ultimately in charge, not the automation.

    Naturally, Boeing’s input for the ‘state of the art’ in pilot skills to their designs will be solicited from the senior, highly experienced, echelons of customer airlines. What if the training intake is not being selected for the same qualities as the senior, experienced, pilots?

  87. Neville says:

    @All

    In case anyone’s interested in checking out my ms on finding MH370 via water chemical analysis, I’ve moved it from dreary Dropbox to Google Docs. Here are the first three paras.

    Abstract: An accelerator mass spectrometer (AMS), has a sensitivity of 1/10^15 = 1g/km^3 (1). I estimate the production rate of aluminum alloy solutes for the MH370 site at the present time to be ~2.1kg/day. Deep ocean currents convey such solutes somewhat unidirectionally, and higher solute densities limit upward spreading. A typical current flow of 5cm/s will carry the solute 4.3km in one day, and reasonable assumptions indicate that during that time the solute will be confined to a volume of ~6.7 km^3, thus producing a solute density of 320g/km^3 = 320/10^15,clearly within the working range of an AMS. 5 ROVs, searching 2.1km abreast at 8km/hr, could complete a 10^5km^2 search in 48 days requiring only 4,300 readings.

    Might such sensitivity be sufficient to find the MH370 signature via upwells in the Indian Ocean Gyro? One snag here is that Al and Fe tend to have unusually high densities in gyros, mostly due to atmospheric inputs (3). A far better choice would be the I.O gyro’s aptly named garbage patch, which seems to vacuum-up samples of everything that’s going around the neighborhood. Deep searches might be best.

    The most certain choice would be the slow deep water currents driven by thermohaline circulations and thus by density gradients. Note that thermohaline currents produce sedimentary deposits called contourites and turbidites, and if any can be found near an MH370 hotspot that has been in growth recently, they might well be worth checking out. Nepheloid layers and turbidity currents might also be considered.

    MH370–IF THE FISH CAN SMELL IT SO CAN WE.
    https://docs.google.com/document/d/1Gug9HrPhAgEvXeQUcPRuZeLGk1D-m8mv2_HEdZtnSa8/edit

  88. Andrew says:

    @Don Thompson

    That’s definitely an issue and is likely become increasingly problematic if current forecasts of a global pilot shortage come to fruition. That said, some time ago, during a discussion about FCOM wind shear procedures, it was put to our chief pilot by Boeing that they had to cater for the ‘lowest common denominator’ during the development of FCOM procedures and recommendations.

    I suspect the MCAS fiasco came about partly because its development and certification was rushed and failed to adequately consider the human factors aspects associated with a system failure. An investigation into those aspects is probably beyond the scope of the two MCAS accident investigations, but more details will hopefully emerge from the various Senate, Transportation Department nd grand jury investigations underway in the US.

  89. Don Thompson says:

    @Hank,

    737 FCOM procedures state that runaway/uncommanded STAB TRIM issues are resolved by selecting both STAB TRIM switches to CUT OUT. The Nov 6th 2018 Boeing FCOM Bulletin reiterates this.

    If a pilot familiar with the NG, but flying a MAX, decides to attempt restoration of the left switch to the active state and expects to restore only Main Electric Trim for the control wheel STABILIZER TRIM switches, that pilot will be bewildered/confused/surprised when uncommanded STAB TRIM follows.

    The MAX switches appear to imply left (PRI): STAB TRIM driven by FCC-A or control wheel STABILIZER TRIM switches; or right (B/U) STAB TRIM driven by FCC-B or control wheel STABILIZER TRIM switches.

    I’m concerned that the Ethiopian AIB Preliminary Report has not indicated the state of the STAB TRIM cut-out switches. Determining whether both switches are, or only the left switch is, restored to normal before the final two short manual bursts and the ultimately unrecoverable automatic AND command is critically important factual information.

  90. airlandseaman says:

    Tbill: Re: ” In the case of MH370 it is not a design flaw but error of omission letting pilot turn off transponder secretly, etc.”

    “…error of omission…”: What utter nonsense. Every transponder ever installed in an aircraft has essentially the same control switch positions as the B777, including STBY and OFF. Pilots need control over the Transponder Mode for many reasons. Certainly not an omission by Boeing or anyone else. SOP.

  91. Andrew says:

    @Trip
    @TBill

    A number of changes have occurred in the five year period since the disappearance of MH370. I don’t know if Boeing had any involvement, but the internal MEC access hatches in our B777s are now locked and the MEC is inaccessible from the cabin. I know that some other operators have taken the same action. Some of the other issues you mentioned are beyond Boeing’s purview, because the relevant standards and regulatory requirements are set by ICAO and other civil aviation regulators. Nevertheless, new standards have emerged to cater for some of the deficiencies that were identified following the disappearances of AF447 and MH370. For example, there are new standards for aircraft tracking, 25 hour CVR recording, deployable recorders, recorder location, etc.

  92. Trip says:

    What do we know about ELTs? At one point people seemed to be saying that they activate inconsistently. How long do they transmit? How deep did underwater can they transmit? How far do they transmit? Were there any tests done with them? The lack of a signal is also a clue.

  93. Brian Anderson says:

    @Trip,

    ELTs have been covered ad infinitum.

    They do not operate “inconsistently”. They may or may not activate, depending on the type of crash. i.e. They, or more likely the antenna feed or the antenna itself might be destroyed in the crash. Once triggered they will continue to operate for days, or longer.

    ELTs cannot transmit if the antenna is under water. 406 MHz ELTs are detected by satellite. Range is not an issue.

    The specification for 406 ELTs provides for a self check procedure before any emergency signal is transmitted, The delay is of the order of 20-30 seconds . . long enough for the antenna to be under water in the case of MH370.

    Lack of a signal is a strong clue that the aircraft did not do a soft water landing. Rather the impact was severe and teh fuselage was under water within a few seconds.

  94. Mick Gilbert says:

    @Don Thompson

    Don,

    Re: ‘The MAX switches appear to imply left (PRI): STAB TRIM driven by FCC-A or control wheel STABILIZER TRIM switches; or right (B/U) STAB TRIM driven by FCC-B or control wheel STABILIZER TRIM switches.

    I see what you are getting at with that explanation but I think that the operation may be different. The following is from an NG FCOM:

    Stabilizer Trim

    ‘Stabilizer trim switches on each control wheel actuate the electric trim motor through the main electric stabilizer trim circuit when the airplane is flown manually. With the autopilot engaged, stabilizer trim is accomplished through the autopilot stabilizer trim circuit. The main electric and autopilot stabilizer trim have two speed modes: high speed with flaps extended and low speed with flaps retracted. If the autopilot is engaged, actuating either pair of stabilizer trim switches automatically disengages the autopilot. The stabilizer trim wheels rotate whenever electric stabilizer trim is actuated.

    ‘The STAB TRIM MAIN ELECT cutout switch and the STAB TRIM AUTOPILOT cutout switch, located on the control stand, are provided to allow the autopilot or main electric trim inputs to be disconnected from the stabilizer trim motor.

    For the NG, at least, that suggests both FCCs operate the stab trim motor through one circuit when autopilot is engaged (the autopilot stabilizer trim circuit) and the other when in manual flight (main electric stabilizer trim circuit). That suggests that the Speed Trim System (STS) commands the stab trim motor through the main electric circuit.

    On the basis that using the control wheel stabilizer trim switches will disengage the autopilot if it is engaged, it’s not 100 per cent clear (at least not to me at this point) whether the control wheel stabilizer trim switches use only the main electric stabilizer trim circuit or both circuits.

    Under normal circumstances if Boeing changed the basic structure of the main electric and autopilot stabilizer trim circuits you would expect that to be covered in the Differences Training. With the MAX, however, that rule may not apply. What the MAX differences training does include is a slide noting the nomenclature change for the cutout switches but suggesting that the circuits controlled have not changed;

    viz NG ‘MAIN ELEC’ = MAX ‘PRI’ and NG ‘AUTO PILOT’ = MAX ‘B/U’.

    The circuit that MCAS uses is almost certainly the ‘old’ main electric stabilizer trim circuit (the same as the STS), referred to as the PRI(mary) circuit on the MAX.

  95. Mick Gilbert says:

    @Trip
    @Brian Anderson

    Since fixed emergency locator transmitters are back into the discussion let me share what I have learned on that topic. Some time back I had a series of rather fruitful exchanges with two extremely helpful fellows from each of COSPAS-SARSAT and ECA Aerospace (parent group of ELTA, manufacturers of MRO’s ELTs) and some very fortuitous timing – COSPAS-SARSAT were at the time running their annual ELT Expert Working Group workshop in Montreal so both the COSPAS-SARSAT Technical Manager and the ECA Group ELT Product Manager were in the same place at the same time.

    The key findings are as follows:

    – The ELTA Emergency Locator Transmitter Model ADT406² AF/AP PN: 01N65900 fitted to 9M-MRO was certified by COSPAS-SARSAT to the technical requirements as defined in COSPAS-SARSAT document Specification for COSPAS-SARSAT 406 MHz Distress Beacons C/S T.001 Issue 3 Revision 3 October 1999 under the test conditions as defined in defined in COSPAS-SARSAT document COSPAS-SARSAT 406 MHz Distress Beacon Type Approval Standard C/S T.007 Issue 3 Revision 7 October 2000.

    – The flight deck remote control unit fitted to 9M-MRO was an ELTA component (P/N 93N6035); it features a two position ARMED/ON switch that is spring loaded in the ARMED position and guarded in the ON position together with a separate push button TEST/RESET switch. According to ELTA the flight deck remote control unit fitted to Malaysia Airlines B777s 9M-MRP and 9M-MRQ is a Boeing sourced component; it features one three position RESET/ARMED/ON switch.

    – The TEST/RESET switch on the ELT (controlled by the flight deck remote control unit) activates one of two functions depending on the status of the ELT unit. If the unit is in its default standby mode, the TEST/RESET switch will activate the self-test function. If the unit is in self-test mode or has been activated (preparing to or actively transmitting a signal), the TEST/RESET switch will reset it to standby mode.

    – To initiate a self-test the TEST/RESET switch must be depressed for two seconds. Once the self-test sequence starts the TEST/RESET switch must be released to reset the switch; holding the TEST/RESET switch in the depressed position serves no further purpose once the self-test sequence starts.

    – There is a 50 second delay between activation of the test function and the transmission of the test signal; the delay consists of a 10 second internal self-test sequence followed by a 10 second period when the self-test result is displayed on the ELT unit via an LED followed by a 30 second ‘waiting condition’ delay.

    – The test signal consists of one burst only with a specific frame synchronization pattern such that it can be distinguished from a distress signal. Once the test signal has been transmitted the unit returns to standby mode.

    – There is a 50 second delay between manual activation (by moving the ON/ARMED switch from the guarded ARMED position to ON) and the transmission of the first distress signal; the delay consists of a 10 second self-test sequence followed by a 10 second period when the self-test result is displayed followed by a 30 second delay.

    – There is a 30 second delay between automatic activation via the ELT’s inbuilt accelerometer and the transmission of the first distress signal; when the unit is activated automatically by the accelerometer switch the self-test and associated results display sequences are by-passed.

    – Holding down the TEST/RESET switch cannot disable the device. The ELT Product Manager rejected the assertion to that effect made in the ICAO Brief on the SAR Response to MH370 dated 29 January 2015. He made it very clear that holding down the TEST/RESET switch would simply activate the self-test sequence. If the switch is held down continuously subsequent to the self-test sequence starting it fulfils no function; the self-test sequence will run and then the unit will return to standby mode.

    COSPAS-SARSAT has ELT detection equipment on the INSAT-3A satellite which sits above the equator at longitude 93.5° East (virtually due north of the likely range of crash sites) so it would have been ideally positioned to detect a distress transmission. As we know, it did not.

    Additionally, one of the COSPAS-SARSAT low earth orbit satellites, MetOps-A, was tracking south-south-west over central Australia at the time most of us suspect that MH370 impacted the water. At 0019 UTC MetOps-A was 826 km above 13.57°S 135.34°E (about 490 km ESE of Darwin) and 10 minutes later it was 840 km above 48.54°S 125.04°E (about 1,850 km south of the Great Australian Bight). That placed it in a not unreasonable position to detect ELT transmissions from locations ranging from around 20°S to 39°S along the 7th arc for impacts around 0025 – 0030 UTC.

  96. Tim says:

    @Andrew

    Re—A number of changes have occurred in the five year period since the disappearance of MH370.

    At my outfit there has been a change in they servicing regime of the O2 bottles. These are now only serviced at base, not down route. Are you aware how yours are serviced these days?

  97. Andrew says:

    @Tim

    Our O2 bottles haven’t been serviced down route for many years, long before MH370. I believe there were too many quality control issues that can be better handled at our home base.

  98. Don Thompson says:

    @Mick,

    Thanks for engaging on the subject of 737 stab trim.

    You wrote… suggests that the Speed Trim System (STS) commands the stab trim motor through the main electric circuit.

    The 737-800 (an NG model) FCOM, Flight Controls – Systems Description for Stabilizer, in the sub-section for Speed Trim System states that “The STS monitors inputs of stabilizer position, thrust lever position, airspeed and vertical speed and then trims the stabilizer using the autopilot stabilizer trim.” That is: the autopilot stabilizer trim.

    The STS function, like MCAS, is also computed in the FCCs and commands only if autopilot is not engaged.

    I’m confident that the NG series STAB TRIM AUTO PILOT cut-out switch affects any command originating in the AFDS FCCs, whether A/P engaged or STS. Ergo, STAB TRIM MAIN ELEC affects only control wheel Stabilizer Trim switch commands. Can’t be emphatic as I have only NG series FCOMs as source docs.

    The Ethiopian AIB Preliminary Report quotes ETH’s MAX series FCOM, “The STAB TRIM PRI cutout switch and the STAB TRIM B/U cutout switch are located on the control stand. If either switch is positioned to CUTOUT, both the autopilot and main electric trim inputs are disconnected from the stabilizer trim motor.

    Again, I can’t be emphatic, but the MAX switches appear to imply left (PRI): STAB TRIM commanded by FCC-A or control wheel STABILIZER TRIM switches; or right (B/U) STAB TRIM commanded by FCC-B or control wheel STABILIZER TRIM switches.

    An FCC originated command may result from normal A/P flight guidance, be STS originated, or be MCAS originated.

    Besides clear explanation of MCAS implementation and the modified STAB TRIM cut-out function, I want to see NTSC-ID and AIB-ET describe the state of each STAB TRIM cut-out switch throughout each accident flight. At some point prior to 05:43:09 the ETH crew restored the ability to command STAB TRIM from the control wheel switches but that action also allowed MCAS to command STAB TRIM. The NTSC-ID only describes the crew operation of the STAB TRIM switches during PK-LQP’s flight prior to the accident flight.

    I am not seeking to deprecate the significance of MCAS’ contribution to these accidents, rather highlight that a change in the STAB TRIM cut-out function contributed to the inability of the crews to safely manage the situation.

  99. Trip says:

    @Mick
    Thanks for your very thorough response. A couple of questions. So any attempt to switch the ELT function will generate a signal? Do the units have their own battery power source or do they rely on power from the plane? Did the engineers have any idea how the plane could make hard water contact and not generate a signal? Is it possible to disable the units themselves? Did mh17 or the 737 Max’s generate a signal? I assume a destructive land crash would not generate a signal. Do we have examples of situations where signals were generated?

  100. Don Thompson says:

    @Trip

    I can add that the Netherlands OVV described that 9M-MRD’s fixed ELT did activate and its signal was received by five ground stations.

    It is noted that the fixed Emergency Locator Transmitter first transmitted a signal at around 13.20:36 (15.20:36CET).

  101. TBill says:

    @Andrew
    What tool do you have next to the fire extinguisher in the B777 cockpit, and has that changed?

  102. Andrew says:

    @TBill

    I assume you’re referring to the crash axe? We don’t carry them on our aircraft and have not done so for many years. We do carry an insulated crow bar.

  103. Andrew says:

    Andrew

    @TBill

    Further to that, US-registered aircraft are still required to carry a crash axe in an area inaccessible to passengers.

  104. Edward Baker says:

    Victor,

    Outstanding analysis, helping to understand that this simply was not a “it stopped working” event. The TCAS would have been in the TA/RA position by normal SOP. Often, selecting TA Only would be driven by checklist associated with some kind of failure that would impact responding to the RA portion of a TCAS event. It is common for the QRH to require RA Only with a single-engine approach, for example.

    The “flyby” waypoint such as IGARI would be a natural and expected flightpath under LNAV flightpath control (either flown manually following the Flight Director, or under autopilot.

  105. Victor Iannello says:

    @Ed: Thank you. Your comments are always welcome here.

  106. Mick Gilbert says:

    @Trip

    So any attempt to switch the ELT function will generate a signal?

    Not exactly. Moving the two position ARMED/ON switch from ARMED to ON initiates manual activation. There is a 50 second delay before the ELT transmits. Pressing the TEST/RESET button before transmission will reset the ELT and suppress the transmission.

    Do the units have their own battery power source or do they rely on power from the plane?

    Battery (LiMNO2) with a five year service life. The battery in 9M-MRO’s fixed ELT was good till November 2014.

    Did the engineers have any idea how the plane could make hard water contact and not generate a signal?

    The ELTA ELT Product Manager was of the view that the aircraft most likely impacted the water at high speed and that the antenna was submerged before the transmission could be made.

    Is it possible to disable the units themselves?

    If you could get to it the ELT unit itself it has an OFF switch.

    The fixed unit is mounted to the inside of the top of the fuselage at STA 1880. That places it right at the rear of the passenger compartment around row 41 and about 3.5 metres above the cabin floor. The ceiling fittings (ceiling, lighting and overhead compartments) are between the cabin and the unit.

    Did mh17 or the 737 Max’s generate a signal?

    As Don has pointed out MH17 did. No mention of ELT activations for either JT610 or ET302 but ELT activation is not included as part of a standard accident report.

    I assume a destructive land crash would not generate a signal.

    Short answer; it depends. The 2012 Asiana Airlines flight 214 (B777) crash at San Francisco had a successful ELT activation/detection as did the 2016 LaMia flight 2933 (Avro RJ85) crash south of Rionegro/Medellín Airport, Colombia. The 2016 Flydubai flight 981 (B737) crash at Rostov-on-Don Airport, Russia did not have a successful ELT activation/detection.

    Do we have examples of situations where signals were generated?

    Yes, plenty. Cospas-Sarsat publishes an annual Report on System Status and Operations (C/S R.007). Annex C to those reports is a List of SAR Events Assisted by Cospas-Sarsat. It includes all ELT activations/detections for the calendar year prior to the report year (eg Annex C to the most recent 2018 report lists all activations/detections for 2017.

  107. Mick Gilbert says:

    @Don Thompson

    Don,

    Yes, I see what you are getting at and I think you could be right. The following comparison is interesting:

    From NG FCOM:

    The STAB TRIM MAIN ELECT cutout switch and the STAB TRIM AUTOPILOT cutout switch, located on the control stand, are provided to allow the autopilot or main electric trim inputs to be disconnected from the stabilizer trim motor.

    Probably from MAX FCOM (taken from ET302 Preliminary Report, p.18):

    The STAB TRIM PRI cutout switch and the STAB TRIM B/U cutout switch are located on the control stand. If either switch is positioned to CUTOUT, both the autopilot and main electric trim inputs are disconnected from the stabilizer trim motor.

    That does look like there has been a change in the function of those switches. And it appears to be a change that may not be fully explained in the Differences Training – this is a screenshot of one page of the computer-based MAX differences training for the MAX dealing with the switch nomenclature change.

    If this turns out to be another change between the NG and the MAX that Boeing thought it best not to burden pilots with that’s probably not going to play all that well.

  108. TBill says:

    @Andrew
    Yes that is what I was wondering. I know some airlines have replaced the axe with crow bar. FSX PMDG777 flight sim shows the axe.

  109. Julia Farrington says:

    @Ed @Victor

    IGARI may well be a commonly used waypoint on flight paths and the ATSB told me years ago that flying via IGARI saved fuel. My argument was and still is that we know Zaharie requested to ATC a change in the flight path and first waypoint, to IGARI while MH370 was on the tarmac ready to take off. OK so it might be a frequent request from pilots, but I was suspicious at Zarahie’s request because he would have known IGARI was out of civilian radar at some point and I THINK you have said that the loss of radar allowed him to turn unnoticed (except possibly by military radar). So I think this points or could point to a significant initial step in a pre formed plan by Zaharie to eventually sabotage the flight. Tell me please if you think I’m flogging a dead horse!

  110. Andrew says:

    @Don Thompson
    @Mick Gilbert

    RE: “If a pilot familiar with the NG, but flying a MAX, decides to attempt restoration of the left switch to the active state and expects to restore only Main Electric Trim for the control wheel STABILIZER TRIM switches, that pilot will be bewildered/confused/surprised when uncommanded STAB TRIM follows.”

    The Preliminary Report states that if either STAB TRIM cutout switch is selected to CUTOUT, both the autopilot and main electric trim input are disconnected. Ergo, both switches would have to be selected to NORMAL for any stabiliser trim (autopilot or main electric) to operate.

    Technical info about the 737 MAX is somewhat sparse, but the following image purportedly shows a 737-7/-8 Training Manual page that states:
    “If there is a stabilizer runaway condition, the pilots move the STAB TRIM PRI (primary) switch to the CUTOUT position. This removes power to the STAB TRIM B/U switch and these…”
    http://www.b737.org.uk/images/mcas-mtm.jpg

    The runaway stabiliser checklist actually calls for both STAB TRIM switches to be moved to CUTOUT, but that might just be for commonality with other versions of the aircraft. The statement from the Training Manual implies that the STAB TRIM B/U switch is simply a backup switch for removing power from the stabiliser trim.

  111. Andrew says:

    @Julia Farrington

    Pardon me for butting in, but the Malaysian safety investigation reports clearly show that MH370 was flight planned via IGARI. The aircraft was initially cleared via the PIBOS A departure while on the ground and after take-off was cleared direct to IGARI. What change in flight path did Zaharie initiate?

  112. TBill says:

    @Julia Farrington
    I agree with Andrew, my understanding MH370 was given clearance straight to IGARI which saves a slight amount of time going direct straight path. But IGARI is the normal path.

    However I feel this new data is important. As pointed out the by @Victor and @sk999, turning off the transponder seems to happen directly abeam IGARI. That is the exact moment when the pilot’s screen would show the aircraft’s heading changing from IGARI to the new waypoint, BITOD. So that implies the pilot is sitting there, watching the screen, and waiting for that exact moment to turn off the Transponder, and conduct the U-turn.

    I can think of no reason why a 3rd party hijacker or fire or something would happen at that moment. Clearly we have apparently have the pilot making this manuever at the exact time necessary to remain in Malaysian airspace (and more to the point, military radar coverage).

    Obviously we do not have the black box to confirm everything, but if we have to propose the preliminary cause of this accident, let’s face it, it looks like the pilot diverted the aircraft for what appears to be his own reasons.

    Also the fact that the altitude data is missing, we seem have evidence the Transponder selector switch was being toggled, in other words we do not have a mechanical event or fire cutting out the Transponder, as Ed Baker points out today.

  113. Paul Smithson says:

    @tbill. Unless you posit a very funky turn, which the military radar did NOT have the resolution to discern, then the turnback commenced at 172200. This is demonstrable from my earlier “trombone” paper. You can’t model a turnback that starts significantly earlier than that but comes back at the right time on the right trajectory. So I maintain that the observables strongly support the inference that the plane continued with its turn and on toward BITOD for 90 seconds aftwr transponder off. Subtract a few seconds, if you like, to roll into the turn. That still makes 80+ seconds before turnback commenced. Why??

  114. Julia Farrington says:

    @Andrew @TBill
    I hesitate to write things on Victor’s blog for fear of appearing technically unhelpful and unknowledgeable.
    I can only reiterate what I’ve already said that it is my understanding from watching a BBC Horizon documentary (which I’ll attempt to find and post the link for), that Zaharie requested a change of flight path/plan from the one filed to one which went via IGARI.

  115. Victor Iannello says:

    @Julia: Documentaries are written by people that are not necessarily experts in this saga. The filed flight plan included IGARI as a waypoint. ATC directed Zaharie to cancel the standard departure route and fly directly to IGARI, which bypassed some waypoints and shortened the distance a bit.

  116. Julia Farrington says:

    https://topdocumentaryfilms.com/where-flight-mh370/

    This is one of the Horizon documentaries but am unsure if it’s the one I can remember. Earylish on in this documentary it says that ATC requested a change to IGARI But my recollection of a documentary I watched said that Zaharie requested it.
    I’ll keep looking

  117. Greg says:

    @Victor

    Can we be absolutely sure that the initial LNAV turn on IGARI was to BITOD (058°) and not, for example, to BIBAN (042°)?

  118. Victor Iannello says:

    @Greg: The ADS-B data shows the plane was on a track to IGARI at 25.5 deg. Around 17:20:12, before reaching IGARI, it began its turn towards BITOD. The track for the final point was 42.6 deg, which is around midway to the IGARI-BITOD track of 58.2 deg. The path followed almost exactly what you would expect for a 15-deg banked turn from IGARI to BITOD in LNAV mode, where the turn is half-way completed at the flyby waypoint.

  119. Greg says:

    @Victor: Thanks, SIR data suggested a delay in starting the turn and caused doubts about the target WPT. Hence the question. I think the matter is definitely clarified.

  120. Greg says:

    @victor: This is of course true provided that there is no time offset in the ADS-B data

  121. TBill says:

    @PaulS
    I do not know when the U-turn started, so I am just saying transpomder off at IGARI in preparation. Right now, I assume the Entry Waypoint and Exit Waypoint are approximately correct, but they could be offset a little bit.

    Below is what FlightSim seems to suggest close-up view of IGARI turn looks like, with a little overshoot. Not sure if real B777 is similar. I am thinking maybe MH370 kept on the curve shown, and then A/P came off so now MH370 is a little above the line to BITOD, and then the turn starts to the Entry point. When I get a chance I will try that.

    https://docs.google.com/document/d/1nvenFN-EO5jnTwwObDtZfIlLIzdhYyBPxlrqj6JPwBE/edit?usp=sharing

  122. TBill says:

    @PaulS
    PS- I am thinking the pilot had a lot to do between IGARI and start of U_Turn…basically I am thinking next step is Left Bus Off, Right Bus off, Right Xfer bus off, Right Tie off, maybe Left Tie Off, Left Xfer Bus off, so now I am in a secondary flight mode to do the U_Turn manually, and I might be ascending too.

  123. Mick Gilbert says:

    @TBill

    Re: ‘Also the fact that the altitude data is missing, we seem have evidence the Transponder selector switch was being toggled, in other words we do not have a mechanical event or fire cutting out the Transponder, as Ed Baker points out today.

    Sorry Bill but that’s all highly conjectural.

    This ‘missing altitude data equals manual deactivation‘ speculation requires that the perpetrator took more than one second to rotate the Transponder Mode Selector knob through about 120°. That task would typically take less than one third of a second to accomplish. For a theory that incorporates precision of timing and ostensibly an initial high workload doesn’t that strike you as an oddly and incongruously casual approach to that task?

    Short of having a proper understanding of how an electrical failure would manifest itself the absent altitude data are interesting. They are perhaps evidence of a very causal manual manipulation of Transponder Mode Selector but they are most assuredly not evidence that there wasn’t a mechanical or electrical event or fire.

  124. Victor Iannello says:

    @Mick Gilbert: Assuming the timestamps are precise, we have two readings with no altitude that are spaced by 0.4s. That then becomes the minimum dwell time in the ALT OFF position. That doesn’t seem like a very long time to me at all.

  125. Trip says:

    @ Mick
    Thanks again for your ELT responses. It seems that the lack of an ELT signal reduces the chance of a soft water landing or a hard earth landing so a missing signal does give some information.

  126. CapFranz says:

    Is it possible that MH370 did not make a sharp turn after IGARI?
    Was that turn made with the Autopilot ON?
    —————————————————————
    “Air Traffic Control/Mode S Transponder System”
    The aircraft transponder was operating satisfactorily up to the time it was lost on the ATC radar screen at 1720.36 UTC, 07 March 2014 (0120:36 MYT, 08 March 2014) as it is stated in page 374 of “The Malaysian ICAO Annex 13 Safety Investigation Team for MH370” ATSB ICAO from July 02, 2018.

    The first LEFT turn that MH370 made after disappearing from ACC Radar was direct to EGAMO heading 273° or WEST bound as it is stated in page 50 of “The Malaysian ICAO Annex 13 Safety Investigation Team for MH370” ATSB ICAO from July 02, 2018. At 3.2 nm after passing IGARI, SSR radar position symbol of MH370 dropped off from radar display.

    That left turn took 2 minutes and 12 seconds to complete the 127° turn from 040° starting at 3.2 NM after IGARI to EGAMO on a heading of 273°. (Airway M765 from IGARI to Kota Bharu is established on Radial 238°-058° from VKB VOR)

    IMAGE: https://alcione.org/INCIDENTE_MALAYSIA_MH370/__________IGARI-HDG-040-EGAMO-HDG-273a.jpg

    (My theory) While the B777 was turning the pilot started the APU and selected the APU GEN ON (Pilot was wearing his quick done O2 mask) Meanwhile ISOLATED the RIGHT AC by disconnecting the LEFT ENGINE IDG, LEFT BACK UP GEN, RIGHT BACK UP GEN, RIGHT ENGINE IDG and ISOLATED THE LEFT AC BUS TIE leaving POWERED only the RIGHT electrical system.

    The pilot also depressurized the B777 by OPENING the FWD and AFT OUTFLOW VALVES and then turned off THE LEFT and RIGHT ENGINE BLEED and both L&R AIR PACKS.

    VIDEO: https://www.youtube.com/watch?v=992PYZRkAXs

  127. Mick Gilbert says:

    @Victor Iannello

    0.4 seconds may not seem like a long time but compared to 0.3 seconds, the time that it would normally take to turn the Transponder Mode Selector knob through about 120° to the STBY position, it most certainly is.

    In moving the selector knob from the normal TA/RA position through to the STBY position it would pass through three other positions. Only one of those, ALT OFF, the penultimate position to STBY, would cause altitude reporting to be disabled.

    If the timing of the manipulation of the selector was perfectly coordinated with the ADS-B transmissions so as to yield the minimum required dwell time (quite the feat) and the manipulation was executed in a uniform fashion you’d be looking at something like 1.3-1.5 seconds to complete a task that would routinely be completed in 0.3 seconds.

    Again, that strikes me as an oddly and incongruously casual approach to that task? Now, none of that is to say that it didn’t happen that way; it may have. But it strikes me as unlikely.

    Just by the bye, do we know anything about a transponder’s ability to ride through power interruptions? Or the source and path for the altitude data?

  128. Mick Gilbert says:

    @Trip

    Re: ‘It seems that the lack of an ELT signal reduces the chance of a soft water landing or a hard earth landing so a missing signal does give some information.

    I would argue that the lack of an ELT transmission is a serious problem for the ‘controlled ditching’ brigade.

  129. Victor Iannello says:

    @Mick Gilbert: On a very consistent basis, your concept of “unlikely” is much, much different than mine. Now you believe it is “unlikely” that there was dwell in the ALT OFF position for a fraction of a second because that would be “incongruously casual”. I strongly disagree, but I won’t try to persuade you otherwise.

    The “altitude missing” data combined with the timing of the event that almost exactly coincides with passing IGARI is yet more evidence that points towards a deliberately diversion by the crew. We cannot make that conclusion with certainty, but evidence continues stacking up in that direction.

  130. Victor Iannello says:

    @CapFranz: If the turn did not begin until reaching the “entry waypoint” as presented in the SIR, then to reach the position and timing of the “exit waypoint”, a bank angle of 35° is required, which is not possible in autopilot. However, if the turn began before reaching the entry waypoint, a turn using HDG SEL or TRK SEL is possible, especially if the speed was turned down to something closer to Minimum Maneuvering Speed (MMS) before the turn.

    So, from my vantage point, it is uncertain whether the turn was flown manually or in autopilot.

    Why did you choose an electrical configuration in which the right IDG and the right backup generator were isolated? If the right IDG continued to supply power to the right bus, there would be no need for the APU. Does the APU serve another use? Was it to maximize thrust from the engines?

  131. Mick Gilbert says:

    @Victor Iannello

    Victor, if you want to believe that it would not be ‘incongruously casual‘ for a trained flight crew member to take four to five times as long as usual to perform a task that was ostensibly pivotal to their malicious plan, a plan that apparently relates to the precision of timing of passing a waypoint, then I certainly won’t try to persuade you otherwise.

    For something that is potentially more instructive in eliminating search area perhaps we might look at the following from the SIR;

    At 1801:59 UTC [0201:59 MYT] the data showed the “blip” on a heading of 022°, speed of 492 kt and altitude at 4,800 ft. This is supported by the “blip” detected by Military radar in the area of Pulau Perak at altitude 4,800 ft at 1801:59 UTC [0201:59 MYT]. At 1803:09 UTC [0203:09 MYT] the “blip” disappeared, only to reappear at 1815:25 UTC [0215:25 MYT] until 1822:12 UTC [0222:12 MYT], about 195 nm from Butterworth, on a heading of 285°, speed of 516 kt and at an altitude of 29,500 ft.

    What are we to make of the report that the aircraft descended to 4,800 feet near Pulau Perak?

    Interestingly, the description from the SIR roughly matches our good old mate, the Lido Hotel slide. The descent to 4,800 feet followed by a subsequent ascent back to 29,500 feet would account for the lacuna in the Lido slide.

    Some time back, well before the SIR was released, I was exploring a variety of possible scenarios to explain the Lido lacuna. One of those scenarios was that the Lido captures had come from the RMAF radar at Western Hill and that the lacuna was caused by the target passing below the radar horizon at about 5,000 feet and then climbing back through it at about 11,000 feet before being finally lost below the radar horizon at about 30,000 feet. I rejected that scenario at the time because of the high average speed required, a ground speed of roughly 500 knots, exceeding Vmo.

    Of late I’ve had cause to re-examine that scenario. The SIR essentially describes the vertical flight profile that would produce the Lido captures albeit with some timing errors.

    If the aircraft were flown in secondary or direct mode, absent the envelope protections, would those speeds (circa up to 450 KIAS) have been possible? Those speeds are certainly well beyond Vmo of 330 KIAS; there is no disputing that. The question is whether it was possible. And if it was possible and did happen, that clearly has some very significant implications for fuel flows and terminal range.

  132. DennisW says:

    @Mick

    The gap in the radar data (I resist “lacuna” for no good reason) is indeed odd. Certainly altitude is a logical explanation. Another explanation is the radar data did not come from Western Hill. That has never been clarified to my knowledge.

  133. Mick Gilbert says:

    @DennisW

    Dennis,

    For a long time I have been of the opinion that the Lido captures were from a Thai military radar. As I mentioned sometime ago I tried looking at all possible scenarios; single radars, composite view from multiple radars, altitude variations impinging the radar horizon, range gating, cycling, etc.

    The SIR had me dust that work off and revisit the Western Hill altitude variation scenario. It broadly matches the SIR description and it explains the Lido captures.

    Separately and more recently, discussing the Ethiopian Airlines crash with some pilots had me rethink my initial rejection of the scenario based on the very high speeds involved, well in excess of Vmo. I’m now of a view that that combination of vertical profile and speed would have been possible but I’m happy to be disabused of the notion by a factually reasoned argument.

    If we’re going to disregard that element of the military radar captures as ‘impossible’ then I would like that to be demonstrated. Otherwise it could well be a variation of ‘When you have eliminated the impossible, whatever remains, however improbable, must be the truth.

    If the aircraft did fly that vertical profile at those speeds it provides some insight into the primary flight controls system mode and, perhaps more importantly, the fuel load at loss of contact. If the aircraft did fly that very high speed descent and subsequent climb west of Penang then you can essentially rule out quite a lot of the southern band of the search area. That, I would think, would be helpful.

  134. Andrew says:

    @Mick Gilbert

    RE: “If the aircraft were flown in secondary or direct mode, absent the envelope protections, would those speeds (circa up to 450 KIAS) have been possible? Those speeds are certainly well beyond Vmo of 330 KIAS; there is no disputing that. The question is whether it was possible. “

    I don’t know the exact figure, but the B777’s design dive speed is somewhere around 360-380 KIAS. Speeds faster than that would probably cause structural damage.

  135. Victor Iannello says:

    @Mick Gilbert: The radar altitudes and speeds in the Malacca Strait are impossible for a B777. Either the altitude capability of the radar was not functioning, or what was captured was not MH370.

    The radar data shows a drop from 58200 ft to 4800 ft in 1 minute. Assuming the starting vertical speed was zero (notice the peak), that corresponds to an average downward vertical acceleration of 0.92g, and a final downward vertical speed of 106,812 fpm = 1055 kn. When the downward velocity is combined with the measured groundspeed of 492 kn, the total speed is 1164 kn, or Mach 1.76 at ISA+10K temperature. (Wind would not change those numbers very much.) There is absolutely no possibility that a B777 that reaches that speed, if even possible, could continue for another 6 hours of flight.

  136. Victor Iannello says:

    @All: Recently I had a discussion with an individual who was in Malaysia in the weeks after the disappearance and was using his expertise to help to interpret the radar data. His observation was the Malaysian military radar system was not properly maintained, and all the altitude data measured by the military radar system should be dismissed. He also believed the Indonesian military radar system was not operational, as the radar stations were powered by diesel generators which were turned off at night to save fuel. The Malaysians and Indonesians were reluctant to admit to any of this.

  137. Don Thompson says:

    @Mick,

    The ATC/Mode S Transponder System control panel is microprocessor based unit connected connected, via low speed ARINC 429 bus, to both ATC/Mode S Transponder units.

    The panel switches will be sampled with some finite periodicity, and the panel settings communicated to the transponder units, also with some periodicity.

    The period during which the control panel mode switch dwelled in the ALT RPTG OFF state is moot. If the panel’s processor sampled the switch as it was in the ALT RPTG OFF position, the panel would signal that state to the transponder units as being active for an period equal to the sampling and/or panel to transponder status update rate. I’d expect the samping rate for a HMI element, such as this switch, to be in the order of 2 to 5Hz, 2Hz fits this case as the two Extended Squitter messages are timed 400ms apart. The final reply at WMKC’s SSR showed a valid altitude, however, its interrogations are made at an interval of 3.8sec. The ADS-B ES messages are timed as received within the interval between the final SSR reply and the next expected reply.

    The data path to the transponder from other avionics sources is common for all data to be stored in the transponders BDS registers: dual, simultaneously active, busses. One bus from each AIMS cabinet. Further, there are two independent sources, a master and a shadow DCGF hosted on separate CPMs, for each bus within each AIMS cabinet.

    RADAR

    We may never get a conclusion to the quandary for the track depicted over the Str of Malacca and the ‘hole in the middle’/lacuna. However, it should be noted that the RAT-31DL is limited in the number of pencil beam formed on the transmit side – only 4. Ergo, it cannot scan through its full gamut of elevation, across the entire range spectrum, without alternating the beam elevation (the Martello, as a contrary example, beam forms only in the receive path).

    I suspect it is possible that the RAT-31DL was surveilling in a mode that provided maximum range (beams at very low elevation) yet covered the practical gamut of elevation closer to Western Hill (beams at higher elevation). It is then possible that such a scanning technique could result in a ‘gap’ between the two scan volumes.

    It has also been suggested that the track was acquired by a naval vessel stationed in the Str of Malacca, that is also possible.

    But classifying the performance or operation of these systems as state secrets is a nonsense.

  138. TBill says:

    @Victor
    Good info from the radar expert. I assume the early JIT team knew all or most, only us public are in the dark. Somebody throwing you a bone (more data) on the data for this blog article too. Thank you for sharing.

    @Mick
    Re: Lido/SIR-
    I am also trying to understand if low altitude is possible. I am confused if there exists an eyewitness report from Pulau Perak island, but if so the person ought to be able to tell us if the aircraft was high altitude or skimming the top of the 800-ft high little island.

    @DonT
    How about Singapore airborne radar as source of Lido track?

    @CaptFranz
    Interesting ideas. I am thinking right side elec all off too, I am thinking that would cut off DFDR.

  139. Marijan says:

    @Victor

    Both facts sound bizarre enough that they are almost certainly true. That can explain well why Malaysia was so uncertain about radar data and waited so long before releasing them, i.e. maybe extra time was required for extra postprocessing.

  140. Greg says:

    @Mick
    @Victor

    „The radar data shows a drop from 58200 ft to 4800 ft in 1 minute.”

    Of these two altitudes, 58200 ft seems rather unlikely.
    If we reject this point as incorrect, we will get a descent from approximately 36000 ft to 4800 ft in 7 min. This looks completely different.

  141. Victor Iannello says:

    @Greg: Yes, it is impossible that a B777 was at 58200 ft. It’s also impossible that with the meteorological conditions on that day it flew at 589 knots. It’s also unlikely it flew at 492 knots at 4,800 ft without damaging the aircraft. (The military data also says the heading was 022° at this time, which differs from the path that has appeared in numerous reports.) So, we are left with a lot of data that doesn’t add up for a B777. Either the speed and altitude data is very inaccurate, or the targets were not MH370.

    All this highlights the need for Malaysia to release the raw military data so that we can do independent analyses instead of guessing.

  142. airlandseaman says:

    The military radar provided reasonably accurate range and azimuth (Lat/Lon) information, but ZERO accurate altitude information according to 2 experts I have communicated with in Malaysia, both familiar with the radar systems. This is consistent with what was reported in the MH370 Safety Investigation Report.

    1.1.3 (1)
    “It became very apparent, however, that the recorded altitude and speed change “blip” to “blip” were well beyond the capability of the aircraft. It was highlighted to the Team that the altitude and speed extracted from the data are subjected to inherent error. The only useful information obtained from the Military radar was the latitude and longitude position of the aircraft as this data is reasonably accurate.”

    Note that speed is also reasonably accurate if averaged over several “blips”.

  143. TBill says:

    @Victor
    Isn’t another circumstantial evidence tidbit against 3rd party take over at IGARI, is that the pilot would have been able to use the Transponder keypad to send the Hijack code? What are the steps for that?

  144. Victor Iannello says:

    @TBill: I am not aware that the pilots of the hijacked planes on 9-11 were able to configure the transponder to the hijack code before control of the aircraft was lost to the hijackers. I wouldn’t put much weight on this for MH370.

  145. Greg says:

    @Airlandseaman
    @Victor

    I do not expect that the release of MIL RADAR raw data will explain anything about altitude, speed and heading. The basic issue is to verify by an independent party if every “blip” is actually MH370. I agree with Victor that this is the most important thing right now.

  146. airlandseaman says:

    Greg: Re: “The basic issue is to verify by an independent party if every “blip” is actually MH370”. How do you propose to do that? How could anyone do that? Every single blip? How would you verify the source of PSR obs?

    What we have is a unique PSR trace from 17:30 to 18:02 (or 18:22 if you count the hotel radar data) that matches the timing of the diversion at IGARI on one end, and the 1st ARC on the other. There are not many PSR obs in the whole database we have, and none of them look anything like a consistent track to anywhere, except the track that is attributed to MH370. I’m sure there are minor errors in the data, but they do not rise to the level of raising doubt about basic path and speeds.

  147. Mick Gilbert says:

    @Victor
    @Andrew
    @airlandseaman
    @Don Thompson
    @Greg
    et al

    Victor,

    Thank you. Yes, I understand that achieving an altitude of 58,200 ft, in of itself, would be impossible. Consequently, the flight dynamics of a descent from such an impossible altitude are similarly impossible. As Mike said the SIR itself indirectly acknowledges the impossibility of some of the data. We can therefore safely strike out the 58,200 ft.

    That done, what still warrants both attention and consideration are those aspects of the reporting that are not manifestly impossible. As Greg has mentioned, if you discard the impossible 58,200 ft and consider the other reported data – 44,700 ft at 1752:31 UTC, 36,700 ft at 1754:52 UTC and 4,800 ft at 1801:59 – you get rates of descent in the order of 3,400 fpm – 4,480 fpm. That’s a vertical speed of 34-44 knots. When added to the measured groundspeed of 492 kn, you get a total speed of around 531 kn. None of that strikes me as being beyond the realm of possibility.

    One of the other aspects of the reporting worth consideration is the statement ‘This is supported by the “blip” detected by Military radar in the area of Pulau Perak at altitude 4,800 ft at 1801:59 UTC [0201:59 MYT]. At 1803:09 UTC [0203:09 MYT] the “blip” disappeared …

    The first thing to note is that that implies that the target was tracked at 4,800 ft for 70 seconds (and the SIR Figure 1.1B supports that interpretation). This altitude measurement is therefore less likely to be an erroneous one-off like the 58,200 ft.

    More to the point, how does a military radar that has ostensibly successfully tracked a target from 220 nm out lose it when it is only 120-odd nautical miles away over water? One answer – in fact, probably the simplest answer – is that the target passed below its radar horizon. And 4,800 ft out past Pulau Perak fits the ‘below its radar horizon’ scenario.

    So, ‘This is supported by the “blip” detected by Military radar in the area of Pulau Perak at altitude 4,800 ft at 1801:59 UTC [0201:59 MYT]. At 1803:09 UTC [0203:09 MYT] the “blip” disappeared, … ‘ are statements that are consistent with a target passing under the radar horizon of the Western Hill radar. And that is what you would expect for a target at that altitude out past Pulau Perak. And it is the sort of narrative that fits the Lido slide, albeit with a timing offset (I seem to recall that someone else posited that possibility).

    It is one thing to query the accuracy of the altitude measurements, it is something entirely different to explain the complete loss of contact. And where one (altitude) is in keeping with the other (loss of contact), it’s a little more difficult to be dismissive of either.

    As an aside, with regards to the altitude measurements, I note that the report that ‘Between 1724:57 UTC [0124:57 MYT] to 1737:35 UTC [0137:35 MYT] … Military data also recorded a significant height variation from 31,150 to 39,116 ft.‘ in fact looks like being within ±1,500 feet of what the independent assessment of the Kota Bharu civilian PSR data suggests. That is there or thereabouts accuracy-wise for 3D air defence radars so maybe the altitude measurements weren’t all that bad (at least at longer ranges) after all.

    As Andrew has pointed out exceeding Vd increases the possibility of structural damage. Were it to have occured though, not all structural damage is an immediate impediment to continued flight. There are a number of examples of airliners suffering not insignificant structural damage and continuing to fly. Moreover, you might consider whether an aircraft that had been subjected to an earlier overspeed event might be more likely to subsequently shed trailing edge components in a later high speed uncontrolled dive? or whether it might be more likely to subsequently suffer a more significant structural failure, something like a wing break, in a later high speed uncontrolled dive?

    I’m most assuredly not wedded to the notion of a high speed low level run up part of the Strait but I’m struck by the fact there is enough by way of a consistent narrative (as well as being broadly consistent with the Lido slide) to mean that it shouldn’t be simply dismissed on the basis of that single impossible altitude datum. I also note that if it did happen then it shifts the southernmost bound of possible termini much further north. On the basis that it potentially refines the possible search area, I think that it is worth some consideration.

  148. DennisW says:

    @Mick

    More to the point, how does a military radar that has ostensibly successfully tracked a target from 220 nm out lose it when it is only 120-odd nautical miles away over water? One answer – in fact, probably the simplest answer – is that the target passed below its radar horizon. And 4,800 ft out past Pulau Perak fits the ‘below its radar horizon’ scenario.

    I don’t agree. The ISAT data at 18:25 supports an airspeed of ~500knots at a heading of 296. That is totally incompatible with an altitude of of 5000′ at Palau Perak. Something is very wrong.

  149. Mick Gilbert says:

    @DennisW

    Re: The ISAT data at 18:25 supports an airspeed of ~500knots at a heading of 296. That is totally incompatible with an altitude of of 5000′ at Palau Perak.

    Actually, I don’t think that it is totally incompatible, Dennis.

    While it is clearly outside the normal operating envelope for the aircraft, I think that it is possible to be at 5,000 feet with a groundspeed of around 500 kn at 1803:09 UTC when you disappear under the Western Hill radar horizon west of Pulau Perak and subsequently, twenty-two minutes later, to be on the other side of MEKAR at around 30,000 feet at around 500 kn. You’ve got to average a groundspeed of ~500 kn through the ascent but the rate of climb required is not particularly aggressive.

  150. Greg says:

    @airlandseaman
    @Mick
    @Victor
    @Andrew
    @Don Thompson

    SIR (1.1.3.1) states: „The Military radar data provided more extensive details of what was termed as “Air Turn Back”. Where are these details? SIR describes 1 “radar return” and 7 “blips” specifying the altitude, speed and heading for 6 of them. Simultaneously SIR reserves “… the altitude and speed extracted from the data are subjected to inherent error”. It is clear that most of the “blips” is not a single “blip” but rather a radar contact with a target consisting of a sequence of “blips”. These “blips” sequences with „the latitude and longitude position of the aircraft as this data is reasonably accurate” are all I would like to see.

  151. Victor Iannello says:

    @Greg: The Malaysians provided the ATSB with military radar data after the turnback at a spacing of 10 seconds. Although I suspect some of that data points were interpolated to fill in gaps, there were clearly more points than isolated blips. We need to see this data so we can better understand it. My suspicion is that the accident investigation team did not try to analyze the data in any significant way. Mostly, the data was used to confirm that MH370 turned back over Malaysia.

  152. paul smithson says:

    @Victor. Re military radar analysis my guess is the same as yours.

    One of interesting revelations from the recently radar and ADS-B data is the magnitude of the time-offsets that have been observed between different data sources. Over 20s in one instance, ~250s in another. In this light, it is not so difficult to imagine that the ~100 second difference between last known position (from Lido) and the first BTO “crossings” might, after all, be explicable by timing offset rather than a lateral offset maneuver. I think this was suggested at the time by Dr B as a possibility, but it looked like too big a timebase inconsistency to be conceivable!

  153. TBill says:

    @Victor
    Another “let’s face it” if Malaysia really thought we had a 3rd party hijacking by non-Malaysian expert pilot perpetrators, there would be more openness on the radar data over Malaysia. I also do not think the CIA etc. would let Malaysia handle the investigation unless the JIT felt what happened was home-grown Malaysian domestic issue. To quote airline safety consultant George Bibel, the industry believes MH370 was an apparent “crime” (ie; not an international terrorism epidode).

  154. Victor Iannello says:

    @TBill: I suspect Malaysia’s initial reluctance to share the military data was mostly because the deficiencies in the radar system would be embarrassing and those weaknesses could possibly be exploited. (The malfunctioning altitude measurements demonstrate this.) Those deficiencies existed five years ago and under a different administration. There is little justification for not releasing the data today.

  155. DennisW says:

    @Victor

    The specifications of the Western Hill radar are in the public domain. It is literally a commercial off-the-shelf piece of equipment that you or I could probably buy. State secret is total nonsense.

    The only semi-valid reason not to publish the raw data is if the data came from a site other than Western Hill.

  156. Victor Iannello says:

    @DennisW: The operational condition for the radar site is not in the public domain. Just look at the false altitude data. That performance deficiency demonstrates a vulnerability. I’m told that the radar sites had not been properly maintained in some very embarrassing ways.

  157. airlandseaman says:

    paul smithson: Re: “One of [the] interesting revelations from the recently [released] radar and ADS-B data is the magnitude of the time-offsets that have been observed between different data sources.”

    Actually, some of us do understand the differences in time stamps fairly well (now). Radar head time, vs. LAN Capture time vs. GPS time. The LAN Capture times were based on a clock running in the laptop used to capture the radar data on the LAN. That time base was delayed by ~206 seconds relative to the radar head time stamps (probably loosely coupled to GPS). That laptop time base is also relatively noisy due to the fact that it runs asynchronously wrt the radar head clock, and it is subject to interrupts, etc. If the Civil PSR radar head time stamps are used, and the CAP times are ignored, the data fits reasonably well with other clocks (<20 seconds).

    That said, we do not know the Lido radar time-base, so your point about the 18:22:12 timestamp is a good one. It certainly could be off by 10s of seconds. That's one more reason why the Malaysians need to release the relevant military radar data sets so all the radar and ADS-B data can be adjusted to a common clock.

  158. DennisW says:

    @Victor

    No intelligent adversary would try to exploit a transient condition (the operational status of poorly maintained equipment). That could be corrected at any time by writing a check to the manufacturer. It is the inherent capabilty of the radar that is the crux of any vulnerability.

    As far as embarrassment is concerned, the Malaysians already endured that by publishing the Lido slide with the “licuna”, and allowing a conclusion that the data came from Western Hill.

    Of course, the Mick Gilbert altitude hypotheses could certainly be correct.

  159. DennisW says:

    @all

    In other news, BA stock has taken a hit this week. A discounted cash flow value of ~$600 is doing little to stop the slide. No way to know where the bottom is.

    https://photos.app.goo.gl/kjri1BKQJPCzNCXk7

  160. Victor Iannello says:

    @DennisW: The fact that the Lido Hotel radar data has never been publicly released is evidence that soon after it was shown, there was an attempt to bury it. So much so that the Australians (who were spending tens of millions of dollars on a search in the SIO) were never given the data from that slide: the military data that was shared with the Australians had no data after 18:02 except for a single point at 18:22.

  161. DennisW says:

    @Victor

    Good points.

  162. CanisMagnusRufus says:

    @VictorI
    Loooong time lurker, first time poster here.

    Question: I think now the evidence points to the pilot ZS as the person most likely to have turned off the transponder. However, does that necessarily mean he also initiated the turn back?

    You mentioned in this latest post the turnback didn’t have to involve handflying the aircraft, meaning someone with access to the flight computer from the E/E bay could have initiated the turnback.

    Couldn’t someone hiding in the E/E bay have waited for ZS to turn off the transponder, then taken over the aircraft from the E/E bay?

    BTW: Love your blog

  163. Victor Iannello says:

    @Canis: Thank you, and welcome.

    I think it is theoretically possible to commandeer the plane from the E/E bay, including disabling the controls in the cockpit. I lean towards the simpler scenarios, but I wouldn’t completely rule out this out.

  164. CanisMagnusRufus says:

    @ VictorI

    So ZS could have intended to turn off the transponder at IGARI, continue flying over the South China Sea, and when approaching mainland China turn the transponder back on again?

  165. DennisW says:

    @all

    Sorry my link to the BA stock price seems to have gotten messed up. Reposted below (old age??). At my age you need to get a driver license renewal every four years. I struggled with my recent renewal for Class C (regular driver license), Class M (for motorcycle extension), and Class A extension (to drive heavy equipment like water tenders for the fire department)…Cali recently imposed the Class A requirement due to the number of civilians being squashed like beer cans by volunteer drivers of class A vehicles. Spent half a day at the DMV.

    Ami cheered me up with her comment that it is probably the last driver license I will ever need.

    https://photos.app.goo.gl/JP42Rm7u7p3KZbxu5

  166. Victor Iannello says:

    @Canis: If you are proposing that occurred, that would contradict the satellite, radar, and debris data.

  167. Andrew says:

    Aviation Week:
    Ethiopian Crash Data Analysis Points To Vane Detachment

    LOS ANGELES—As the investigation continues into the causes of the Mar. 10 Ethiopian Airlines Boeing 737 MAX accident, sources close to the probe say flight data recorder (FDR) data firmly supports the supposition that the aircraft’s left angle-of-attack (AOA) sensor vane detached seconds after take-off and that, contrary to statements from the airline, suggests the crew did not follow all the steps for the correct procedure for a runaway stabilizer.

    Detailed analysis of the FDR trace data shows that approximately six seconds after liftoff was signaled by the weight-on-wheels switch data, the data indicate the divergence in angle-of-attack (AOA) and the onset of the captain’s stick-shaker, or stall warning. Almost simultaneously, data shows the AOA sensor vane pivoted to an extreme nose-high position.

    This, says one source, is a clear indication that the AOA’s external vane was sheared off—most likely by a bird impact. The vane is counter-balanced by a weight located inside the AOA sensor mounting unit, and without aerodynamic forces acting on the vane, the counterweight drops down. The AOA sensor, however, interpreted the position of the alpha vane balance as being at an extreme nose-high angle-of-attack.

    With the stick shaker active, the trace indicates the crew pushed forward on the column to counteract what they believed were indications of potential approach to stall. The aircraft, now in level flight, also accelerated rapidly as its power setting remained at 94% N1 thrust used for take-off. This was followed by some manual trim inputs using the thumb switches on the control column.

    Seconds after speed advisories were heard, the crew raised the flaps. With the autopilot turned off, flaps up and erroneous AOA data being fed to the flight control computer (FCC), the stage was set for the MAX’s maneuvering characteristics augmentation system (MCAS) to activate. This is indicated by approximately 8-sec of nose-down stabilizer movement, which was followed by the use of manual trim on the control column. However, with the MCAS having moved the stabilizer trim by 2.5 units, the amount of manual nose-up trim applied to counteract the movement was around 0.5 units, or roughly only 20% of the amount required to correctly re-trim the aircraft.

    Because of the way the aircraft’s flight control computer P11.1 software worked, the use of manual trim also reset the MCAS timer, and 5 sec. later, its logic having not sensed any correction to an appropriate AOA, the MCAS activated again. The second input was enough to put in the full nose-down trim amount. The crew again manually counteracted with nose-up trim, this time offsetting the full amount of mis-trim applied by the latest MCAS activation.

    By then, some 80% of the initial MCAS-applied nose down trim was still in place, leaving the aircraft incorrectly trimmed. The crew then activated the stabilizer trim cutoff switches, a fact the flight data recorder indicates by showing that, despite the MCAS issuing a further command, there was no corresponding stabilizer motion. The aircraft was flying at about 2,000 ft. above ground level, and climbing.

    The crew apparently attempted to manually trim the aircraft, using the center-console mounted control trim wheels, but could not. The cut-out switches were then turned back on, and manual trim briefly applied twice in quick succession. This reset the MCAS and resulted in the triggering of a third nose-down trim activation lasting around 6 sec.

    The source says the residual forces from the mis-trim would be locked into the control system when the stabilizer cut-off switches were thrown. This would have resulted in column forces of up to around 50 lb. when the system was switched back on.

    Although this could have been reduced by manually trimming the aircraft, this did not occur, and the third MCAS activation placed the aircraft in a steep nose-down attitude. This occurred with the aircraft near its peak altitude on the flight—about 6,000 ft. The engines remained at full take-off power throughout the flight, imposing high aerodynamic loads on the elevators as the crew attempted to pull back on the columns.

    Vertical acceleration data also indicates momentary negative g during which the AOA sensor on the left side unwinds. This is seen as further validation of the theory that the external part of the alpha vane was detached as the apparent change in angle indication could only be explained by the effect of negative g on the counterbalance weight, forcing it to float up inside the sensor housing. In addition, the captain’s stick shaker also comes off twice in this final phase, further reinforcing the severed vane notion.

    The source indicates the crew appeared to be overwhelmed and, in a high workload environment, may not have followed the recommended procedures for re-trimming. Boeing’s stabilizer runaway checklist’s second step directs pilots to “control aircraft pitch attitude manually with control column and main electric trim as needed,” according to one U.S. airline’s manual reviewed by Aviation Week. If the runaway condition persists, the cut-out switches should be toggled, the checklist says.

  168. CanisMagnusRufus says:

    @ VictorI
    I’m suggesting ZS was instructed pre-flight by someone higher up to turn off the transponder after IGARI and turn it back on when nearing the coast of China.
    However, a third party took over the plane and turned back and flew to the Indian Ocean.

  169. TBill says:

    @CanisMagnusRufus
    It is hard to imagine how perpetrators in the MEC Bay would be able to conduct the U-Turn at IGARI and stay in Malaysian military radar coverage, and also if you go back a few blogs to Victor’s Penang fly by article, even a professional pilot says it looks like ZS making an experienced pilot approach to the airport waypoints.

    So if you say ZS is innocent, I think you have to go with 3rd party in the cockpit forcing the pilot to fly on thru Penang and possibly the 3rd party takes over after Penang. But the current blog article does not support that idea very well, because why would ZS wait for BITOD to show on the screen and then make a turn? You could imagine 3rd party hijacker with Malaysian military and pilot expertise, but that’s a stretch, and also it still implies MH370 was an episode of Malaysian domestic intruigue, probably including the pilot ZS in some capacity.

  170. Don Thompson says:

    @Andrew,

    I find it odd that Aviation Week has quoted its source as describing that the ETH crew “may not have followed the recommended procedures for re-trimming” but not questioned its source as to why the crew then appears not to, or is unable to, follow the recommended procedure for manual trimming.

    The FDR trace appears to show that the procedure was followed to the point where the STAB TRIM cut-out switches were toggled. An interval of 2m15s then followed with no stabilizer position change. I can’t imagine the crew considered they’d solved the problem during that interval, that they simply did not make any attempt to manually crank the trim wheel, and then toggled the STAB TRIM cut-out switches back to NORMAL so as to demand NU trim using main electric trim. After toggling the switches to NORMAL, the FCC made a successful final AND demand.

    It is vitally important that a full disclosure is made for the ETH crew actions during that interval, including the time at which one or both STAB TRIM switches were toggled back to NORMAL.

  171. CanisMagnusRufus says:

    @TBill
    I believe ZS had nothing to do with the turnback, but he was responsible for turning off the transponder on the instructions of MAS/Mod.
    There may be information on the military radar which would reveal what happened at IGARI, which is the real reason why Malaysia military is reluctant to release it. This could also be why it’s a state secret, nothing to do with the capabilities or lack thereof of the military radar.

  172. TBill says:

    @CanisMagnusRufus
    I agree with you, Malaysia is keeping a lid on exactly what happened. Essentially Malaysia is taking the 5th amendment, excising its right as a sovereign country to not incriminate itself or its people.

  173. Andrew says:

    @Don Thompson

    In saying the crew ”may not have followed the recommended procedures for re-trimming”, I think Aviation Week is referring to the crew’s actions before the STAB TRIM CUTOUT switches were selected to CUTOUT. The runaway stabiliser checklist calls for the crew to ”control airplane pitch attitude manually with control column and main electric trim as needed” before activating the cutout switches. The FDR data shows the crew applied two nose up trim commands with the electric trim, the first of which was relatively short. Those nose up trim commands were not sufficient to re-trim the aircraft and consequently it was still trimmed nose down when the cutout switches were activated. The pilot was left holding a significant amount of back pressure on the control column to stop the aircraft descending, which then made manual trimming via the trim wheels difficult, if not impossible. If the aircraft had been trimmed properly before the cutout switches were activated, the crew might not have been tempted to restore the cutout switches to NORMAL, which was the pre-cursor for the final MCAS input.

  174. Andrew says:

    BTW, I don’t blame the crew for what happened. The Boeing FCOM Bulletin stated that “Electric stabiliser trim can be used to neutralise control column pitch forces before moving the STAB TRIM CUTOUT switches to CUTOUT”. The bulletin failed to point out that electric trim MUST be used to neutralise the control column forces before activating the cutout switches. It also failed to point out that manual trimming might not be possible if the aircraft is not trimmed properly before the cutout switches are activated.

  175. CapFranz says:

    @Victor

    You wrote:
    –“Why did you choose an electrical configuration in which the right IDG and the right backup generator were isolated? If the right IDG continued to supply power to the right bus, there would be no need for the APU. Does the APU serve another use? Was it to maximize thrust from the engines?”

    Yes, it was to maximize thrust from the engines also by depressurizing the airplane. That would give an extra thrust power to fly much farther away.

  176. Hank says:

    @Andrew @Don Thompson

    Re ET302. It is strange that there has not been much media discussion of the change to the Stab Trim panel between NG and MAX and what should be done now. It seems it is OK for NG and MAX to operate differently? Boeing should change back to the NG panel and wiring. More than software is required for MAX.

    It makes sense for one toggle switch to allow disconnect of the yoke buttons in case a runaway is caused by a stuck trim button. It also makes sense to allow the other toggle to lock out the autopilot/MCAS. This would allow the pilots to re-engage yoke electric trim if a stuck button was not the cause of a hard over trim command.

    I could not understand why there was no pitch trim change in the two minutes after the toggle switches were throw. I assumed they would have pulled out the handles on the trim wheels and spun them to zero out the elevator. But I didn’t consider that loads could lock up the wheels.

  177. Andrew says:

    @Hank

    Very little information about the 737 MAX’s revised STAB TRIM CUTOUT switches has surfaced. My understanding is that the switches now serve the same purpose, ie in the CUTOUT position, either switch will deactivate ALL electric trim input, regardless of the source. I don’t know the logic behind that change, but there must have been a reason.

    I don’t believe the change in switch architecture had any impact on the outcome of these accidents. The runaway stabiliser checklist has always called for both switches to be selected to CUTOUT in the event of a runaway. There is no procedure that requires only one switch to be selected to CUTOUT. In a runaway scenario, pilots don’t have time to analyse the cause, nor do they have the ability to determine which system was causing the fault once both switches are in CUTOUT.

    RE: “I assumed they would have pulled out the handles on the trim wheels and spun them to zero out the elevator. But I didn’t consider that loads could lock up the wheels.”

    The Preliminary Report (p.11) shows that the FO did try to manually trim the aircraft, but according to the CVR he stated that “…it is not working”, probably due to the excessive air loads on the stabiliser. That characteristic of the B737 has been known for a long time, but is currently only described by a single paragraph in the FCTM:

    “Excessive airloads on the stabilizer may require effort by both pilots to correct the mis-trim. In extreme cases it may be necessary to aerodynamically relieve the airloads to allow manual trimming. Accelerate or decelerate towards the in-trim speed while attempting to trim manually.”

    The outcome of the Ethiopian MCAS runaway might have been very different if the FCOM Bulletin that was issued after the first accident had mentioned the potential problems with manual trim, and reinforced the requirement to make sure the aircraft is properly trimmed before selecting the CUTOUT switches.

  178. Andrew says:

    @Hank

    RE: “In extreme cases it may be necessary to aerodynamically relieve the airloads to allow manual trimming. Accelerate or decelerate towards the in-trim speed while attempting to trim manually.”

    Further to my previous post, in the Ethiopian case the pilots would have needed to relax the back pressure on the control column and accelerate to relieve the air loads on the stabiliser. However, given the speed was already at or above Vmo, they did not have have that option available to them. The only option they had was to restore the electric trim via the CUTOUT switches and apply a significant amount of nose up trim to properly trim the aircraft. After properly trimming the aircraft they would have needed to immediately re-select the CUTOUT switches to CUTOUT. Mind you, that’s easily said in hindsight and from the comfort of my desk. The Ethiopian pilots were faced with a very difficult situation and were more than likely panicked. They did restore the electric trim, but they failed to apply sufficient nose up trim and did not re-select the CUTOUT switches to CUTOUT before MCAS commanded the terminal nose down input.

  179. ST says:

    Thanks Victor and fellow IG members for another very fascinating analysis. The graphical representation helps readers like me understand the complex analysis in a lot more easy to fathom way.

    Something Andrew mentioned on one of the posts stuck with me as a reader. He said that pilots don’t keep worrying about the passengers when they are flying an aircraft – their primary focus is in keeping the aircraft safe and in effect the passengers safe. The analysis in this post seems to draw us more closer to the intent as regards MH370 being the opposite of above no matter how much benefit of doubt we want to give…

  180. Greg says:

    @all – Question:
    Has anyone analyzed skills and training level on B777 for co-pilot MH370 (FH)?
    SIR provides some puzzling information related to it.

    1.5.10 Psychological and Social Events
    „… the FO’s ability and professional approach to work was reported to be good. This was evident with the rapid fleet promotion within 3 years as a professional pilot.”

    3.1.7 Organisational and Management Information 2) Malaysia Airlines
    „vii) there were no training records available for the FO from the beginning of his simulator training and initial operating experience (IOE) to his present fleet where he was still under training. All the training reports were with him in his personal training file on board the flight.”

    4.2.3 Safety Recommendation #16
    „A document back-up system should be implemented on every training sorties, simulator trainings, and flight trainings completed by a trainee should have their original form submitted to the Training Department and a copy retained by the trainee in his personal training file.”

    RMP provides raw data on the FH training schedule, not only for B777 but also for the A330. It’s probably not a coincidence. For some reasons, they were considered important. The data are included in the sheets “Idividual Crew Roster Report”.

    A short summary
    A330 training:
    4 SIM sessions completed with a Certificate of Test Qalification (A330COFT)
    6 sectors of observation flights (OBS/OBA)
    6 sectors Line Training for FO (LTF1 – with safety pilot)
    6 sectors Line Training Functional for FO (LTFF – without safety pilot)
    2 sectors Final Line Check for FO (FLCF)

    A330 fleet experiences:
    1 year and 2 days in the service, 50 completed sectors, around 300FH

    B777 training:
    15 SIM sessions completed, no information about a Certificate of Test Qalification
    4 sectors Line Training for FO (LTF1 – with safety pilot)
    2 sectors Line Training Functional for FO (LTFF – without safety pilot)
    2 sectors Line Training for FO (LTF1 – with safety pilot again)
    2 planned (MH370) sectors Line Training Functional for FO (LTFF – without safety pilot)

    SIR: 1.17.2 Malaysia Airlines, 3) Operations, a) Flight Operations, xii) MAS B777 Training and Standards
    „On the B777 a pilot under training normally would require to operate a certain number of minimum sectors before he could be certified to be fully functional as a line operational pilot (end-of-training). Depending on theprevious aircraft flown, the minimum and maximum number of required training sectors were, as follows:
    • Last aircraft flown B737: Minimum 10 sectors, Maximum 14 sectors; and
    • Last aircraft flown A330: Minimum 8 sectors, Maximum 14 sectors.”

    I would be grateful for the impressions and opinions of professional pilots, but also for other people present on this blog.

  181. Don Thompson says:

    Greg,

    I have long been curious why FH underwent type conversion to the A330, then within 18mths was doing the same for the B772. That is an expensive process.

    Also, short absences became common during 2013, as recorded in his Individual Crew Roster Report (code MC1 or EL). No Individual Crew Roster Reports are included for Sep 2013 thru Nov 2013.

    Shoddy record keeping and reporting but doesn’t bring us anything specific for the events of 7-8 Mar 2014.

  182. Andrew says:

    @Greg

    RE: “SIR provides some puzzling information related to it.”

    What, specifically, do you find puzzling?

  183. DennisW says:

    @Andrew

    The Leeham News link creates more questions than answers. Why a new satellite system? Inmarsat is a satellite system that can relay position over most of the earth. The Iridium satellite system (my preference using the Garmin inReach device) uses it. Ami uses SPOT (a device tethered by the Globalstar satellite system). We use both the devices to track each other. Me for tracking Ami between our ranch and beach places. I use inReach so Ami can track me on my off road motorcycle adventures. Both capabilities have been around for years, and are very inexpensive and reliable.

    The real issue, IMO, is whether the system can be turned off in the air. Both SPOT and inReach have a long battery life, and there would be no need to power either one during a flight. An invention is not needed. The technology is already here (and commercially available).

    Example track of Ami going from ranch to beach.

    https://photos.app.goo.gl/awn1aBM7UZYkQjoK9

  184. CanisMagnusRufus says:

    @all

    QUESTION: does anybody know if a normal MH370 flight from KL to Beijing appears at any point on Vietnamese radar?
    We know that MH370 was asked to check in with HCM ATC but at what point would they actually appear on Vietnamese radar?
    Is there a radar ‘blind spot’ between Malaysia and Vietnam?

  185. DennisW says:

    @Andrew

    BTW, in my working career I would routinely “no bid” FAA solitications. They called me in to ask why. I told them they were a “pain in the ass” to work with, and there were much easier ways to make money. Dealing with people sucking on a government teat is usually unpleasant.

  186. Andrew says:

    @DennisW

    RE: “Why a new satellite system? Inmarsat is a satellite system that can relay position over most of the earth. The Iridium satellite system (my preference using the Garmin inReach device) uses it.”

    Aireon’s space-based ADS-B system uses ADS-B receivers hosted on the Iridium NEXT satellite constellation. I guess the advantage is that ADS-B transmits far more than just position data and that Aireon’s system can feed that data into the existing ground-based ADS-B systems. I agree that a ‘tamper-proof’ system is still needed, but this new system has huge advantages for ATC purposes in that controllers will be able to reduce the separation between aircraft in many areas where that is not currently possible.

    A ‘tamper-proof’ autonomous distress tracking (ADT) system is still several years away and will initially only apply to newly-built aircraft, although there are incentives for operators to modify older aircraft. The ADT system will transmit the aircraft’s position at least once per minute when certain events are detected by the aircraft’s systems. Aircraft operators will also be able to activate the system remotely any time there is a concern about the aircraft or its position.

    RE: “BTW, in my working career I would routinely “no bid” FAA solitications.”

    I don’t blame you. I hate dealing with bureaucrats.

  187. Don Thompson says:

    @DennisW

    Aireon’s network is a straightforward relay of an existing aircraft output, the 1090MHz extended squitter message from the transponder. It benefits from the FAA & EASA mandates for ADS-B equippage but, in itself, Aireon’s performance does not place any interoperability requirements on aircraft.

    The ICAO set out performance requirements for GADSS, the Global Aviation Distress and Safety System. Aireon enables the first phase of GADSS.

    Autonomous Distress Tracking is the (future) element of GADSS intended to provide continued tracking when an aircraft ceases to cooperate in routine communication, navigation & surveillance.

    Orolia/Kannad, an existing aircraft ELT manufacturer, working within the EU Helios project (involved with the Galileo GNSS) has developed an ELT for the Distress Tracking scenario. The Orolia/Kannad ELT-DT is compliant with existing regulations for ELT’s but adds Galileo’s Return Link Service to the device, ergo, the ability to initiate the ELT from the ground.

    I have seen discussions of other initiatives that suggest modifying the performance requirements for the ATC Transponder to render that device tamperproof. I guess that makes sense for Honeywell.

    ADT inevitably requires flight plan conformance or route adherence monitoring of tracked aircraft where exceptions to that monitoring would provide a trigger to the ELT-DT. I find the ELT-DT solution an elegant one, and one that implies the least resistance to adoption by regulators.

  188. Greg says:

    @Don Thomson

    “…doesn’t bring us anything specific for the events of 7-8 Mar 2014.”

    It’s really nothing specific. But in the end there were only two people in the cockpit of departing MH370. CRM was still there. It probably had some impact on the events.That’s the whole reason for my interest.

  189. Greg says:

    @Andrew

    RE: “SIR provides some puzzling information related to it.”

    Specifically: I am surprised by the statement that a rapid fleet promotion is only evidence of professional skills as if there were no other evidence. I can not believe that there were no training records available for the FO. How he was released to a passenger flight without a safety pilot. The recommendation is as obvious as “fly safely”.
    These are just my personal feelings.

  190. Greg says:

    Co-pilot training and skills – cont.

    A comparison of Individual Crew Roster Report sheets for ZS and FH indicates that they were scheduled on the same day, Jan 16, 2014 for a simulator session marked ZFT (Zero Flight Time) for FH and TRE (Type Rating Examiner) for ZS. Usually, such a session takes place at the end of the training and includes an exam for COFT (Certificate of Test Qalification as in the case of A330, probably it’s the same as proficiency check). Finally, FH passed ZFT session (RMP) and proficiency check (SIR) on Jan 26 probably with another TRE (no trace in ZS data). Did ZS and FH really never meet in the simulator?

  191. Mick Gilbert says:

    @Greg

    Re: A comparison of Individual Crew Roster Report sheets for ZS and FH indicates that they were scheduled on the same day, Jan 16, 2014 for a simulator session marked ZFT (Zero Flight Time) for FH and TRE (Type Rating Examiner) for ZS.

    If, you have a good look at those rosters you’ll see that the Captain’s and FO’s duty times only overlap by 90 minutes out of their respective 5:30 hour shifts. More importantly, their sim times don’t overlap at all; the Captain’s sim session finishes at the same time that the FO’s starts, 16:00 hrs 1/16/2014 UTC.

    At best they might have past each other on the steps.

  192. Brian Anderson says:

    Just happened across this youtube video today. Don’t know if it has been around for a while.
    Probably the most factual and balanced summary that I have seen.

    https://youtu.be/kd2KEHvK-q8

  193. Andrew says:

    @Greg

    RE: ” I am surprised by the statement that a rapid fleet promotion is only evidence of professional skills as if there were no other evidence.”

    Section 1.5.10 of the SIR provides a cursory psychological assessment of the Capt and FO; it’s not an appraisal of their professional skills. I don’t know about Malaysian Airlines, but pilot promotion is normally based on seniority. Provided a pilot meets the required standard and does not have any disciplinary issues, he or she will normally be promoted according to time in the company. I believe the statement about rapid fleet promotion was only intended to show that the FO was performing well and had a professional approach to his work.

    RE: ” I can not believe that there were no training records available for the FO.”

    It seems that training records were kept in a personal training file that was carried by the trainee during his or her training. That practice was clearly deficient, as noted by Safety Recommendation #16, which states:

    ” A document back-up system should be implemented on every training sorties, simulator trainings, and flight trainings completed by a trainee should have their original form submitted to the Training Department and a copy retained by the trainee in his personal training file. “

    RE: ”How he was released to a passenger flight without a safety pilot.”

    The Malaysia Airlines Training Manual stipulates that a safety pilot is only required for the first six sectors of line training/initial operating experience. I don’t think that’s unusual. Where I work, we normally only carry a safety pilot for the first four sectors following a zero flight time conversion course.

    RE: ” A comparison of Individual Crew Roster Report sheets for ZS and FH indicates that they were scheduled on the same day, Jan 16, 2014 for a simulator session marked ZFT (Zero Flight Time) for FH and TRE (Type Rating Examiner) for ZS. “

    The way I read it, FH was rostered for the first of two ETOPS simulator sessions on 16 Jan. He wasn’t rostered for the ZFT session until 26 Jan. ZS was also rostered for an ETOPS session as TRE on 16 Jan, but as Mick said, he was finishing his simulator session as FH’s session started. Clearly they were not in the simulator at the same time that day.

  194. ArthurC says:

    Sorry, I can’t help but ask: did @oddball lose interest or did he get banned for using up too much bandwidth? 😀

  195. Victor Iannello says:

    @ArthurC: @oddball is neither banned, nor do his comments require my approval before appearing.

  196. Hank says:

    @Andrew

    Thanks for your response regarding ET302 toggles. I agree with everything that you said. I agree that the memory item checklist required both switches to be used for NG and MAX.

    I suspect it is because of MCAS that Boeing changed the STAB TRIM panel and associated wiring, but it would be interesting to learn their rationale since it was the unchanged from the 737-100/200 to the classic to the NG series.

    While the poor MAX design caused the significant trim upset, the underlying inability of the manual trim system to operate during certain aerodynamic loads is a major issue. I have read about how experienced 737 pilots porpoise the airplane dropping elevator to trim and then applying elevator to get the stabilizer where they want it. Seems like this should be a regular part of simulator training.

  197. Wall says:

    Hey guys,

    I was wondering how long it could take before Ocean Infinity is going to look for this aircraft again. Are they currently busy with other projects? Do you think it will happen this year?

    Wall

  198. Victor Iannello says:

    @Wall: My guess is there will be no search until at least November due to seasonal constraints. Even then, I don’t think there will be another search until a manageable area is defined and justified, and the Malaysian government agrees.

  199. George.G says:

    @Wall,
    Victor’s summary seems very apt.
    A review of Victor’s blogs will show that there is no real consensus or even tendency for agreement on where the aircraft actually lies.
    And that is from a team of concerned and interested parties who do their combined best to glean any relevant information from the body of knowledge gathered.
    But, even here, the temptation to speculate, even fantasise, can sometimes rear it’s ugly head occasionally concerning MH370.

    The only argument with any weight would seem to be to search along the seventh arc from 25S north to 20S to at least “put that one to bed” thus ensuring coverage of any even likely position along the arc.

    After that one could argue ad-infinitum.
    If it was an “armchair exercise”, and not a real-world tragedy and massive undertaking, then perhaps (and “my best guess”) would to “trawl” around the whole searched area making larger and larger circuits until a larger area had been covered. This would be just in case local winds or other condition caused the descending aircraft to traverse further from the arc than previously considered likely. Of course, this course of action is extremely unlikely to ever to be undertaken.

    Pity all involved, if sometime in the distant future when we are all passed it is found in a previously searched area.

    There, that is a plan of attack, in three separate steps.
    Unlikely that even the first step will be actioned in the near future.

  200. ventus45 says:

    Ocean Infinity has said that they would not search without approval from the Malaysian Government. That seems to be the primary roadblock.

  201. Victor Iannello says:

    There is a new drift study completed at the University of Miami. The study concludes the most probable point of impact along the 7th arc is 25S latitude.

    I don’t understand the conclusions of this paper, which were based on a drift models constructed from undrogued drifter data and the timing and location of recovered debris.

    Figure 4 shows the posterior distribution along the 7th arc for each of the debris pieces that were analyzed. The distributions generally fall into two groups that peak at either 38S or 25S, with the piece (“Roy”) found at Mossel Bay in South Africa an outlier. (The paper says 36S rather than 38S, but clearly the more southern peak in Figure 4 is south of 36S.) In general, the debris found at the more southern latitudes in East Africa are predicted to originate from the more northern latitudes along the 7th arc, and vice versa.

    Of the two peaks at 36S (38S) or 25S, the 25S impact location is considered more probable because drifter 56568 crossed the 7th arc near 25S in March 2014 and drifted near Reunion Island in July 2015, as did the flaperon. However, there is surprisingly no mention in the paper that the flaperon is expected to drift at a higher speed than undrogued drifters due to the extra leeway that has been experimentally measured using replicas.

    My interpretation of their results is the most likely impact location would be between the two peaks, i.e., somewhere between 38S and 25S.

  202. Victor Iannello says:

    And yet another drift paper from GEOMAR.

  203. airlandseaman says:

    The new GEOMAR drift study (Fig 6) suggests a most likely POI at ~S34.1 degrees.

  204. DennisW says:

    @alsm

    The original Geomar study, based solely on the flaperon finding as I recall, had a POI much farther North – up around Christmas Island.

    https://photos.app.goo.gl/6nhB8M9wgLEQo5m68

    That location was supported by the ocean disturbance analysis done by the Russian, Mikhail P., (who never responded to any of my questions).

    https://photos.app.goo.gl/v8KP5ih8ZnMjvZua7

  205. Brian Anderson says:

    Sadly GEOMAR continues to perpetuate the myth that the aircraft’s engines are somehow involved in the handshakes . . . “Detailed analysis of satellite communications, provided in the form of handshakes between the aircraft’s engines and satellites . . . . “

  206. DennisW says:

    @Wall

    I was wondering how long it could take before Ocean Infinity is going to look for this aircraft again. Are they currently busy with other projects? Do you think it will happen this year?

    There is no compelling reason to do an underwater search anywhere. I could wave my arms around and argue for 25S to 20S near the last arc, but it truly would be an arm waving argument. We are at the point of a “cold case”, and I don’t sense that Malaysia has any enthusiasm to renew a search.

    I doubt an underwater search will ever be undertaken again. The flight path is as obscure as the motive for it.

  207. Wall says:

    @airlandseaman

    That comes closes to the 35S area.
    An area I believe is still the most likely one. I told Victor a couple of months ago that Copernicus data shows it could lie west of the previous search zone (35/34 S). But we will only know when the plane is found. Until then, it’s guessing, and hoping it will be found soon.

  208. Victor Iannello says:

    @DennisW said: There is no compelling reason to do an underwater search anywhere.

    Ocean Infinity has the desire and the resources to conduct another search. There are without a doubt some headwinds to starting the search, such as the reluctance of Malaysia, and the challenge in defining a manageable area that has a reasonable probability of success, but it is too early to conclude a search will not occur in the near future.

  209. Niels says:

    @alsm, Victor
    At first sight, and focusing on the blue dots in fig. 6 that are intersected by the 7th arc the new Geomar paper seems to indicate roughly S33 – S36, which matches well with the area of interest indicated by the “First Principles Review”. It also corresponds to a possible POI indicated by the method which I recently reported, which is encouraging (but in the current status not more than that).

    http://mh370.radiantphysics.com/2019/03/03/mh370-family-member-give-us-the-truth/#comment-22264
    http://mh370.radiantphysics.com/2019/03/03/mh370-family-member-give-us-the-truth/#comment-22323
    http://mh370.radiantphysics.com/2019/04/03/insights-from-new-mh370-tracking-data/#comment-22825

    Clearly, we should be aware of possible coincidences, before drawing any conclusion. For all these studies there can be significant errors (systematic and/or random in nature), so a first question that arises is how the “Geomar” authors validated their method. Did they for example compare model results with drift paths of undrogued drifters? That’s at least what David Griffin did (and it revealed problems with the circulation model, hence the need for the de-biasing “fix”). I will study the new paper having this in mind and where necessary contact the authors.
    Regarding my own work: I’m in the process of writing it out in detail and may be able to share a paper somewhere in May. Estimating possible error ranges will take several months more.

  210. airlandseaman says:

    Re 610 and 302, here is an informative interview with Peter Lemme:
    https://airinsight.com/boeings-max-podcast-on-where-are-we-and-what-do-we-know/

  211. Peter Norton says:

    > Brian Anderson says:
    > Sadly GEOMAR continues to perpetuate the myth that the aircraft’s engines
    > are somehow involved in the handshakes . . . “Detailed analysis of
    > satellite communications, provided in the form of handshakes between the
    > aircraft’s engines and satellites . . . . “

    Somehow this keeps popping up regularly. Is this an entirely baseless myth or is there anything whatsoever that the engines contribute to the satellite communication ?

  212. Victor Iannello says:

    @Wall: There is some interest in further exploring whether the BEDAX-SouthPole great circle path, which crosses the 7th arc near 34.3S altitude, has special statistical significance beyond the already-demonstrated match to the BTO and BFO data. However, this path implies the debris field was missed, or the aircraft glided beyond the area along the arc that was previously searched. On a positive note, if we assume the plane was programmed to fly the BEDAX-SouthPole path (due south along a line of constant longitude), the size of the search area would be quite manageable.

  213. Victor Iannello says:

    @Peter Norton: Journalists don’t seem to understand the distinction between SATCOM, ACARS, and EHM (engine health monitoring), and use them interchangeably since the disappearance. I don’t think that will change.

  214. DennisW says:

    @Victor

    There is some interest in further exploring whether the BEDAX-SouthPole great circle path, which crosses the 7th arc near 34.3S altitude, has special statistical significance beyond the already-demonstrated match to the BTO and BFO data.

    Statistical significance relative to what?

  215. Victor Iannello says:

    @DennisW: To be clear, I am not the primary contributor, and I prefer to not say more about the underlying statistics. I expect at some time the new analysis will be presented here for comments. It’s a work in progress, and it’s not conclusive that there will be a statistically preferred path, although I can say that BEDAX-SouthPole is getting a lot of attention.

  216. Don Thompson says:

    @LNI610 & ETH302 followers

    Reuters reporting that the Canadian view is that MAX differences should involve simulator training time.

    Peter Lemme’s discussion at AirInsight covers the MAX issues well, an important contribution.

  217. Andrew says:

    @Don Thompson

    The FAA’s draft revision to the FSB report for the B737 is open for comment until 30 April. It will be interesting to see if the revision is amended if there’s pressure from Canada and possibly others.

    https://www.faa.gov/aircraft/draft_docs/fsb/

  218. Andrew says:

    An interesting article on pilot training from Aviation Week (registration required):

    Why Pilots Need More Than Simulator Training

  219. Andrew says:

    RE: B737 Manual Trimming

    For those interested, the following video shows that manual trimming is almost impossible if the aircraft is severely out of trim at high speed. It also explains how pilots can recover from such a scenario. The ‘rollercoaster’ technique requires a lot of height to recover from a nose down trim condition. Further, the manoeuvre is no longer described in the manuals and it’s unlikely that airlines train their pilots in its use as a recovery technique. Given ET302’s relatively low height above the ground, the pilots are unlikely to have been able to accomplish a recovery using the ‘rollercoaster’ technique, even if they had been trained in its use.

    https://youtu.be/aoNOVlxJmow?t=605

  220. DennisW says:

    @Victor

    I can say that BEDAX-SouthPole is getting a lot of attention.

    That area is interesting for a number of reasons. In my case I am a big fan of integrated Doppler since it is “quieter” than snap shot position fixes (BTO). The 33S-35S region (original Inmarsat analytics) shows particularly good agreeement between the LNAV BTO/BFO “conventional” methodology and integrated Doppler. I almost stuck a pin in the map there myself.

  221. Brian Anderson says:

    @Victor,

    Re: BEDAX-SouthPole

    Seems to me that there are a couple of inconsistencies with this track though.

    1. I don’t know how accurately GE plots the arcs, but this track seems to miss the 19:41 arc, rather than cut it [twice], as a polynomial through the LOS distances to the satellite would indicate.

    2. The speed would have to be significantly lower between 19:41 and 20:41. Perhaps about 410 knots.

    3. The speed to the next arc, at 21:41, would have to be significantly greater. Perhaps about 516 knots.

  222. Victor Iannello says:

    @Brian Anderson: I assure you that there are great circle paths with a constant Mach speed, on a track of 180°T (i.e., of constant longitude), that pass near BEDAX, and which satisfy the BTO error criteria (RMS error and average error) as well as the fuel constraints. At this point, many have checked this. It looks as though you are assuming there is no BTO error. In fact, the standard deviation of the BTO error for the R1200 channel is around 29 μs.

  223. TBill says:

    @Victor
    Re: MH370 Search
    I’d say the big news here is that the proposal to next search 20-25 South has apparently fallen by the wayside. Although I am not highly optimistic of any search area anymore, we perhaps need to decide if the search is process of elimination, or going after presumed hot spots.

    Implications of discounting 20-25 South are:
    (1) redouble focus on passive paths previously suggested
    (2) discount of +/- 25 nm range from Arc7 (going for wider glide below 25 South)
    (3) I was told (on Reddit discussion) some apparent lack of IG confidence in orig Fugro search completeness, whereas the submarine was found by re-searching older searched area.
    (4) No wonder OI is not getting going searching if we have no consensus where to search.

    But the 34 South has been serached quite well. If I recall, Fugro did the “meat” in middle of Arc7 whereas OI sandwiched the outside areas (whole wheat bread).

    I don’t see why searhcing 20-25 South +/-25 NM is considered too big…seems like OI could have knocked that off in a few weeks if they had kept going. Although, it does get deeper water, which I always think the searchers prefer searchable areas. But not sure the MH370 Capt had easy search area in mind.

    I am an old fan of 180 South path, I do consider that the base case (first guess) flight path. But my current guess is ative pilot deviating off it.

  224. Brian Anderson says:

    @Victor,

    I agree that “near” BEDAX would work. And yes, I understand about the BTO Std Dev. I don’t have tolerances on the arcs plotted in GE though.

    If I provide for a little tolerance on the 19:41 BTO then it looks like a very good fit.

    The simplicity of a 180deg track, or a target end point of 90degS, 90degE is quite appealing.

  225. Niels says:

    @Victor, Brian Anderson

    I think Brian raises an important point. In the “best fit” path I shared back in March I see a transition towards 180 degrees track between 19:41 and 20:41, coming from around 185 degrees at 19:41, see last page:
    https://www.dropbox.com/s/ralni6o1r2izoo7/MH370_TTpaths.pdf?dl=0

    So for those looking for a statistical “optimum” (perhaps by looking at zero mean BTO error in addition to min. RMS BTO error) it could be interesting to see what happens if one takes 20:41 as starting time in stead of 19:41.

  226. Victor Iannello says:

    @Niels: If there are acceptable automated paths starting at 19:41, what is the logic of starting at 20:41?

  227. Niels says:

    @Victor
    It is possible that the 19:41 point “contaminates” the statistical analysis. I would do both and see what it brings.

  228. Victor Iannello says:

    @Niels: It’s possible that both 19:41 and 20:41 “contaminate” the statistical analysis. Where do you draw the line?

  229. Victor Iannello says:

    @TBill: At this point, there is no consensus about where to search next. There are reasons to be interested in the due south path (such as the simplicity of the BEDAX-SouthPole path), but many of those reasons are “hunches” rather than objective criteria. That path also has some drawbacks, e.g., the 7th arc in that vicinity has been searched out past 25 NM, and it requires a loiter north of Sumatra before traveling south. The area along the 7th arc north of 25S latitude also has some drawbacks.

    At some point, perhaps the case for searching the due south path can be made, and contributors here can weigh-in. People can of course continue to make the case for other areas to search. In my opinion, a persuasive case has not yet been presented for any new area.

  230. Niels says:

    @Victor
    Intuitively I can imagine that the “steepness” of the optimum may tell us something and that it will vary depending on including the 1941 point or not.
    I will try to find a specialist here on statistics of small number of data points to see how to approach this best.

  231. TBill says:

    @Victor
    “In my opinion, a persuasive case has not yet been presented for any new area.”

    I agree.
    To quote John Nance from an old MH370 TV news clip, finding MH370 is like eating an elephant, one bite at a time. So if Malaysia/OI says that they need to know what part of the elephant contains the aircraft, we are done searching.

    But John Nance did not say we do not search for aircraft unless we know what part of the elephant to search.

  232. DennisW says:

    @TBill/Niels/Victor

    The question of where to search next really cannot be determined from parametric statistics IMO. Parametric in this context means being able to be derived from some assumed probablity distribution. A good example is asking what the probabilty that the 100th person measured will be the tallest person in a 100 person sample. I think most people here would compute the mean and standard deviation of the previous 99 samples. Pick the largest of these samples to determine the probabilty of the last sample exceeding that value. We have a collective love affair with parametric statistics.

    The non-parametric approach is simpler and more robust. The probabilty of any one sample being the largest is 1/100 regardless of how the human height is distributed.

    In my view we can use Richard’s drift study and the 00:19 BFO values to say that the probability of the aircraft being +/- 25nm from the last arc between 20S and 38S is very high. Searching 25S to 20S is clearly preferred over the small probability that the wreckage was missed.

    I do not think additional analytics will change that view.

  233. paul smithson says:

    If you are going to bring back searched sections of arc why would you not put 38S at the top of the pile, except that we have spent so long convincing ourselves that the first best estimate was wrong?

  234. DennisW says:

    “If your experiment needs a statistician, you need a better experiment.”

    ― Ernest Rutherford

  235. Niels says:

    Rutherford must have been a great experimentalist. I came across papers from his lab when looking for relative simple and small scale methods to perform nuclear fusion. We’ve already collected a 50 kV source needed, and now looking for a suitable Friday afternoon 🙂

    More seriously: in the case we are looking at the dataset is what it is, and I think we all agree we are trying to squeeze every last bit of information out of it.
    It looks (already for a while) that you think the limit has been reached. Perhaps (practically speaking) you are right. Let’s critically judge and discuss each of the papers that hopefully will be shared here sometime soon.

  236. Niels says:

    (@DennisW)

  237. Niels says:

    @TBill
    If I understand the 3 Oct 2017 ATSB report correctly:
    32.8 – 35 degrees south, close to 7th arc, was searched by Go Phoenix (“underwater search area 1”, fig. 47)

  238. Victor Iannello says:

    @Niels said: Intuitively I can imagine that the “steepness” of the optimum may tell us something and that it will vary depending on including the 1941 point or not.

    We know that reducing the number of constraints can only broaden the peak (dip) of the function that is maximized (minimized).

  239. Victor Iannello says:

    @Paul Smithson: The impact at 38S is contradicted by fuel consumption models and drift analyses, but I agree that if we are considering a crossing of the 7th arc at 34S, it would be useful to re-examine 38S.

  240. Victor Iannello says:

    @DennisW: Without a doubt, the final BFO values, indicative of an increasingly steep descent, and the failure of OI’s past search, both contradict a search near 34S. Unfortunately, all remaining areas have contradicting evidence.

  241. Ventus45 says:

    @Niels,

    Have you considered trying to optimise the “steepness” of your function, by initially using only the mid flight arcs (3rd 4th & 5th) and then extending outwards to the 2nd & 6th respectively, and in the final step, out to the 1st and 7th ? Would such an approach be helpful ?

  242. Niels says:

    @Victor
    To illustrate the “contamination” by a data point which is “off”, and the fitting procedure I had in mind when making the comment, I created the following example.
    I have two series of points. In the first series I have 4 points aligned and one point “off”. In the second series I omit the point which is “off”:
    1: (0,0.2), (2,0), (4,0), (6,0), (8,0)
    2: (2,0), (4,0), (6,0), (8,0)
    For both series I fit a straight line ax+b such that the mean error with the points is zero (by varying b), for a number of different values of a. I calculate the RMS error for each such “fit” ax+b. The result is plotted. In the second case the minimum in the RMS error as a function of a is deeper.

    https://www.dropbox.com/s/nvjkmj2auozc188/RMS1.pdf?dl=0

    @Ventus45
    It is an interesting suggestion; however Victor could be right that if you remove too many constraints (and if the data points left are relative close together) a possible optimum might become less pronounced.
    Intitial calculations indicate that there is a possible issue with path “straightness” in the 19:41 – 20:41 interval. Therefore, for the moment, my suggestion to compare the case including all arcs from 19:41 onwards to the case including all arcs from 20:41 onwards.

  243. Victor Iannello says:

    @Niels: Your example is not really what we have. We know that each BTO measurement has an uncertainty (standard deviation) of around 29 μs. In your example, 4 of the points have zero error, and the first point has error of 0.2.

    In our case, we are searching for one or more paths that are statistically “not improbable” based on a small sample of independent measurements, knowing what the standard deviation and mean are for a large population of independent measurements. That sample of measurements should satisfy certain criteria regarding standard deviation, mean, and “randomness”. If those criteria are met for a sample of measurements that includes 19:41, I don’t see why the value at 19:41 should be excluded.

    In any event, I look forward to seeing your work.

  244. DennisW says:

    @Niels

    Measurement “editing” was a contentious topic in my working career. One of my staff scientists, Dr. L, was always messing around with it. In the context of GPS, where you often have an over-determined solution (more than four satellites), it makes sense to compute a position, and then check the residual errors on each of the ranges. A large range error can then be discarded. The technique was very useful in urban canyons where there is often no line of sight to a given satellite, and the only signal is due to a reflection off a building. Dr. L even devised an “altitude hold” approach to use an over-determined technique when only four satellite signals were present. Altitude hold is reasonably justified in cities.

    Being old (and simple-minded), I regarded the whole endeavor as cute but unnecessary given the Kalman filtering using odometry and gyros in a blended automotive navigation system. At the end of the day I gave up arguing about it. So it goes in this case.

  245. TBill says:

    @ALSM
    You have brought SATCOM_Guru to my attention re: 737MAX as an industry expert. So I’d be curious of any prior position he had on MH370?

  246. airlandseaman says:

    TBill: I asked Peter about your inquiry.

    Hi Peter. A colleague asked if you have followed the MH370 story. Do you have any current opinions on what happened or where to look next?

    Peter Lemme replied:
    I have no idea. I know the Inmarsat person that had the data recorded from AF447 review by good practice, but I have not followed anything closely at all. I figure whatever happened, the flight deck had no one at the control, the CVR will show silence and then alarms as the flameout, DFDR more or less normal, until flameout. A hijack seems most likely since multiple systems were disabled.

  247. Niels says:

    @VictorI, DennisW
    Re 19:41 point
    I think we are in closer agreement than it may seem. I’m not in favor of arbitrarily throwing out data points, and you can make the case that my slight distrust of the 19:41 data point at this moment is subjective: it is based on the path curvature and lower groundspeed that I find around 19:41 with the “best fit” procedure, suggesting that some maneuvering was still ongoing around that time (and backwards compatible with a turn south just before IGOGU). However, with a different polynomial fit (still within error margins) this picture may change. Trying different polynomial fits is on the agenda (but a lot of work to do properly). From an “old and simple minded” perspective you can also make the case that using both datasets (19:41,….) and (20:41,…) will easily end up in a “cherry picking” scenario. I’m aware of that. On the other hand a “try and see where it brings us” approach, as unscientific as it may look, does sometimes lead to unexpected insights.

  248. TBill says:

    @ALSM
    Thank you for making the contact!

  249. DennisW says:

    @Niels

    I truly do support your (and anyone else’s) efforts. That is how, as you pointed out earlier, progress is made. I echo Victor’s comment – look forward to your work.

  250. David says:

    A 737 Max liability assessment below. I notice that the apparently corrupt left AoA data sourced from two successive LionAir indicators gets no coverage in this.

    https://seekingalpha.com/article/4255952-boeing-737-max-determining-costs-culpability-groundings

  251. Peter Norton says:

    Victor Iannello: “all remaining areas have contradicting evidence”

    What evidence contradicts 20-25S ?

  252. Victor Iannello says:

    @Peter Norton: I believe that the a poor fit to the BFO data and/or complicated paths significantly reduce the probability that these paths are correct. Perhaps others have other reasons. Of course if you are willing to accept multiple turns and changes of speed after 19:41 and changes in altitude timed with handshakes, there are acceptable paths north of 25S.

  253. TimR says:

    @Victor “Of course if you are willing to accept multiple turns and changes of speed after 19:41 and changes in altitude timed with handshakes, there are acceptable paths north of 25S.”

    There is another possibility that I have raised previously.
    A possible scenario was that MH370 flown by Captain Zaharie Shah proceeded out into the Andaman Sea then came back and around to BEDAX and on to a heading due South and slowed to 370 knots.
    Using waypoints at ISBIX, Cocos Islands and Christmas Island crossed arcs 2, 3, 4, 5, maintaining a steady 370 knots.
    After Christmas Island slowed and crossed the 6th arc on a heading to Yogyakarta airfield.
    The last signal, the 7th arc, placed MH370 some 160nm from the airfield and only 80nm from land.
    Nothing that I have been told or seen since 2014 is convincing enough to say that the Captain deviated from this plan.
    However for some unknown reason he ran out of fuel.

  254. Mick Rooney says:

    I’m always grateful for the interchanges on this site. I’m pragmatic, and I understand rabbit holes. So many. We can analyse until the cows come home. Where next? Some realities. Ocean Infinity (or any other exploration company), now or in 2020, is not going to entertain a non-consensus. But that is where we are and in many ways the forming of consensus is actually what brought about the last Ocean Infinity search.

    We don’t have that now. Continue and push north? Widen the arc east and west? That may be the next search step. It isn’t going to happen without two search vessels over several months. I don’t see any way forward beyond that.

    And, once again, that may require locking the right partners and independent minds in a room, over a weekend, wherever (London, Melbourne, Paris, etc), throwing away the key until they can produce a cohesive search plan OI or any other company can take to the Malaysia authorities.

    Granted, the past year has been all about distraction. particularly since November 2018. But we’ve all got used to that now. March 2019 helped a little, but no more than a reminder that consensus is not where it needs to be.

    Victor, I’m wondering how that can be achieved and what more needs to be done to create a legitimate search in 2020. Your thoughts?

  255. DennisW says:

    @Victor

    I believe that the a poor fit to the BFO data and/or complicated paths significantly reduce the probability that these paths are correct.

    I strongly disagree. I have long held the opinion that the usefullness of the BFO data is limited to infering the aircraft went South at the FMT, and lost altitude very rapidly at 00:19. Figure 5.4 of the DSTG book supports this view.

    https://photos.app.goo.gl/zUy8R3ACpu25A7ms9

    The fact that the area South of 25S has been searched to +/- 25nm without success speaks volumes. I don’t understand how you or anyone else) can prefer the area South of 25S to the continuation of the underwater search of the area North of 25S. I think it is misleading to categorize the BFO values for flight paths North of 25S as “contradicting evidence”.

  256. Victor Iannello says:

    @TimR: Yes, your path is a possibility. However, I place it in the “hunch” category along with many other possibilities until additional evidence surfaces that raises its relative probability. I don’t think that’s likely to occur at this point.

    @DennisW: Yes, the fixed frequency bias profile that was presented in the DSTG report shows that bias drift of the measured BFO is possible. However, significant bias drift was not observed for flight MH371, which immediately preceded MH370, nor was it observed by Inmarsat for the Kuala Lumpur to Amsterdam flight that was contemporaneous to MH370, as presented in the JON paper by Ashton, et al. If there was bias drift on MH370, it occurred in such a way to resemble a straight flight of constant Mach number that terminated well south of 25S. I agree that it doesn’t completely rule out a terminus further north, but I’d say the BFO data suggests more southern paths are more likely.

  257. Victor Iannello says:

    @Mick Rooney: Either we reach a consensus and make a recommendation to Ocean Infinity, or we don’t. My estimate is that we’ll know before September (2019). I think in the next several weeks you’ll see at least one new analysis surface, and that analysis will getting a thorough review on this blog.

    As for distractions, they are only significant if we allow them to be. I don’t think anybody here is distracted by the numerous stories in the UK tabloids or the silliness that appears on Facebook and Twitter. Nor do I believe that Ocean Infinity is distracted. I for one have deleted my Facebook account and refuse to participate in the absolute stupidity and drama that I have seen in the VeritasMH370 and Quest groups. As for Twitter, I primarily use it to raise awareness of some of the material presented on this blog.

  258. Niels says:

    @Mick Rooney
    “And, once again, that may require locking the right partners and independent minds in a room, over a weekend, wherever (London, Melbourne, Paris, etc), throwing away the key until they can produce a cohesive search plan OI or any other company can take to the Malaysia authorities.”

    I have been thinking in similar direction for a while now. My feeling is that through a focused/well prepared, few day workshop, bringing together some of the long term independent contributors, already important steps could be made, at least in identifying key bottlenecks in reaching consensus, and how to efficiently address those.
    I see a few “loose ends” coming back on the table repeatedly, and as a collective we’re currently not very efficient in dealing with those.

  259. TBill says:

    @Victor
    I have previously floated the idea of a smaller MH370 workshop in the Wash DC/NoVA area. That was a couple years years ago, maybe before your blog got going. Of course Australia seems to have more participants. Not clear to me workshop helps right now but perhaps.

  260. Victor Iannello says:

    @Niels: What do you see as the “loose ends” that could be addressed in a group setting?

  261. Warren Platts says:

    Re: final ADSB points, the bulls-eye, turn at IGARI, possibly to MIDAN

    I just now saw this post, and looked at my GE to compare it, and I have a point marked 17:21:03 at 6.97N 103.63E. I honestly can’t remember exactly where that came from, but I am thinking it must be an ADSB position.

    I interpolated the position of the “Bulls-eye” to the best of my ability, and got a position of 7.02N 103.68E. Further, the Bulls-eye lands smack on my projected path, i.e., it forms a 3-point straight line from 17:20, 17:21:03, 17:22.

    The ground speed from the 17:21:03 position to the “bulls-eye” is only 257 knots. That seems awful slow, but if the aircraft was in a steep climb, such a slow speed could be expected.

    The heading of this line is ~044+, such that if you project it out it runs straight into the waypoint MIDAN (8.0N 104.633333333E). Coincidence?

    (BTW, if anybody wants my waypoint database, it is kml file that is pretty much comprehensive for the area of the world we are interested in: https://www.dropbox.com/s/b7jbhoh8dp3eace/waypoints.kml?dl=0 )

    These facts taken together would indicate that the “bulls-eye” position is consistent with what we know, other than the apparent cutoff of the AD-SB data abeam of IGARI at 17:20:34:55. It could perhaps be that list of the AD-SB data is cut off for some reason?

    If the “bulls-eye” is legitimate, that would indicate the transponder was cut off after 17:22. As for mode of flight, if MIDAN was programmed into the FMC, it could perhaps still have been in LNAV mode. Alternatively, if LNAV was shut off at IGARI when the heading was 044, the plane would maintain the same course until further inputs, I think.

    (Also, it seems that the spreadsheet with the AD-SB data has the times offset by 4 minutes? Am I reading that right?)

  262. Warren Platts says:

    Re: “loose ends”

    One that bugs me is the FMT. Can we say for sure that a path that proceeded directly down airway P627 from NILAM or SANOB until, say, POVUS or BEDAX must be excluded from the analysis (as a opposed to a turn more in the vicinity of IGOGU that then overflies the northern tip of Sumatra)?

    If not, that would considerably relieve fuel constraints for a far western crash site.

  263. Niels says:

    @Victor
    I wouldn’t focus only on typical “loose ends” to guide formulating aims and agenda for such a workshop/meeting, if that’s how you read my comment. They are perhaps more “symptomatic” for the way we operate, that’s why I mentioned this.

    Enhancing the “collective” approach of the problem we are addressing should be leading. My experience with international collaborations is that you can exchange as many emails as you want, add Skype meetings to the communication etc., usually what really enhances a collaboration and the communication effectively, is to meet face-to-face at least once and preferably periodically.

    Besides that: of course the content of such meeting should be well planned: what to share (tools, models, data..), what to study together more in detail (e.g. through inviting specialists), identifying key bottlenecks, which strategies to align, formulating action points and how to implement those, aimed level of conclusions at the end of the meeting etc.

  264. Victor Iannello says:

    @Warren Platts: The ADS-B presented above was from an ATC receiver at Terengganu. The only other ADS-B data we have when MH370 was near IGARI is what FlightRadar24 (FR24) has provided. If you examine FR24’s data and compare this to the ATC data, you see that only some of the records are archived, some of those records are interpolated, and some are extrapolated. (Look at the low resolution of the latitude and longitude of many of the records.) As stated in the post, the Safety Investigation Report says that the Mode S symbol dropped off the radar display at 17:20:36, which is close in timing to the last ADS-B point. It is very unlikely that any ADS-B data was transmitted after 17:20:34.55.

    As for trying to conclude what path MH370 followed after 18:22, I’d say that with the data we have in hand right now, that’s almost impossible, except in the eyes of some that refuse to believe any scenario but their own.

  265. TBill says:

    @Warren Platts
    If I recall correctly, there were some approx. 1-min time differences between Vietnam and Malaysia radar quotes, so 17:22 might be approximately correct depending on which radar source is being quoted. Obviously to get accurate speeds we would need to somehow normalize the different radar sources to try to come to a unified picture. Whereas the military and/or other country radar data appears to not be forthcoming, I think we are probably stuck with the relatively accurate civil primary radar data. Given the uncertainty in the U-turn exact shape and exact time, we are somewhat guessing whether it was a slow turn with ascent or normal speed.

    I guess Victor et al were still trying to see if anything more could be learned from new data discussed above.

  266. Victor Iannello says:

    @Niels: I am not opposed to a meeting where there could be a productive exchange of ideas and a discussion of how to move forward. Mike, Don, and I participated in a meeting of this type organized by OI in London in December 2017, and the group collectively did reach an agreement about where to search next.

    For many reasons, what you are proposing at this point in time is much more difficult. That said, I am open to any reasonable proposal. Unfortunately, in order to make a search recommendation with any reasonable chance of success, what we really need is additional data that either does not exist or is unlikely to be released.

  267. Niels says:

    @Victor
    Yes, I think I understand and share some of your concerns. The expectations on direct results of such meeting should not be too high. Which does not mean it will not be useful; the impact on slightly longer term could be considerable.

  268. Benjamin says:

    Might it be possible that there was an exchange of planes around IGARI, so that it was not 9M-MRO that flew (back) to Malaysia? – Consider 9M-MRO diving below radar coverage and the second plane, the decoy, climbing above radar coverage. The plane exchange might have happened either without any turn of 9M-MRO, just as passing each other. Or after the 9M-MRO-turn when there is the radar coverage gap (~103.25/7.0 to ~103.15/6.85). Three reasons that make me think this way: First the assumably weak military radar coverage of this (too sharp) turn, including the shifts, might have been able to confuse two planes for one. Maybe also a reason for not publishing the data, if Malaysia was involved in the crime. Second the coverage gap. Third the Combodia story as the Airlines first response theat prevented urgend actions; maybe it was correct data and not a missleading “projection” only, but with 9M-MRO flying low altitude, avoiding radar coverage. In this case the Airline would not have been part of the crime. Most weak part of the story is still the factual non-response from at least six bodies accountable for air traffic over several hours. This might explain this at least a bit.

  269. Victor Iannello says:

    @Benjamin: There’s some evidence that suggests that the airframe that was captured by primary radar was indeed 9M-MRO. The evidence includes:

    1) The registration of the First Officer’s cell phone on a tower on Penang Island.
    2) The fixed frequency bias (FFB) of the BFO data during the log-on at 18:25 is consistent with 9M-MRO.
    3) Some of the recovered debris, such as the flaperon, was identified as part of 9M-MRO.

  270. Peter Norton says:

    @Victor Iannello:
    What if 9M-MRO passed Penang, continued on that heading and what we interpret as the FMT was actually a decoy aircraft logging on at 18:25 ?

    This explains (1).

    As for (2), could the FFB match by chance? If not, could it have been timed to match on purpose?

    The debris (3) could have been dumped into the ocean somewhere along the 7th arc. (If I create a fake SatCom trace into the SIO, I obviously need debris there too.)

    Granted, it’s far-fetched (which is probably true for every remaining scenario at this point), but is it technically possible ?

    At least it’s one of the few scenarios that provides a good reason for the re-logon.

  271. Victor Iannello says:

    @Peter Norton: With sufficient resources and knowledge, every piece of evidence (e.g., radar, SATCOM, debris, cell phone registration, sim data) can be faked. In the extremely unlikely event that it was, there’s no hope of finding the plane with the evidence we have.

  272. Benjamin says:

    @Victor Iannello: I am aware that these three parts indicate that 9M-MRO took the way as officially assumed so far (Malaca Street, SIO). But they may be part of the deceit and could have been falsified. My question was more intended to question the following: Whether the data you have newly analysed around IGARI could look the same if there would have been a “plane exchange”? – Why such far-fetched ideas? Well, to me it becomes more and more likely that the whole crime might have been of a much larger scale (people involved on board and on ground) and with several elements of deliberate deceit, assuring a long lasting or even forever coverage of the truth. And with that in mind, I consider mainly two obvious intensions: First to kill someone or destroy something that was on the plane. This would fairly go with a remote but unknown SIO destination. Or Second, to kidnap someone or to steal something from the plane. This would require a safe landing (or less likely as quite risky, a mid-air drop). And with a safe landing it would be of high priority to make it look the opposite (look like remote undknown SIO destination). We have to ask ourselfs the “WHY?”: Why was somone so passionate, and furthermore so successful, in covering this crime?

  273. Victor Iannello says:

    @Benjamin: As I said to @Peter Norton, every piece of evidence we have could have been faked. The challenge then becomes to assemble a reasonable scenario that includes the who, why, and how. I haven’t seen one.

  274. TBill says:

    @Benjamin
    It is conceivable that MH370 was larger scale than one person. @TimR’s extortion plot rumor holds that the known-to-some-people purpose of the diversion was to negotiate terms with PM Razak (re: new elections or Anwar release etc.). Personally I would say it looks probable that ZS was involved in diverting the aircraft for clandestine reasons, but I do not know exactly the intent (I have some guesses, of course).

  275. DennisW says:

    @Victor/TBill/…

    The extent of ZS’s political involvement has consistently been down played by the Malaysian investigations. “Democracy is Dead”, the submarine t-shirt, the Pardi involvement… It goes on and on. Shah was a patriot who wanted political change. The embezzlement of 100’s of millions of government funds by Najib and his buddy Jho Low was well known by the citizens of Shah’s class.

    The diversion was not a suicide – nothing to point in that direction. It was not a kidnapping or hijacking of cargo. There has not been an incident (that I could find) in the history of commercial aviation involving kidnapping or cargo hijacking. It is much simpler to obtain a person or an item on ground.

    My long held opinion was that the diversion involved the recovery of embezzled funds via an electronic transfer. Quick and easily verifiable, and virtually impossible to reverse without a lengthy legal process involving courts outside Malaysia.

Leave a Reply