Zaharie Shah’s Secret Psych Evaluation

571 thoughts on “Zaharie Shah’s Secret Psych Evaluation”

1. @Gysbreght,

   The ROD implied by processing the 00:19:37 BFO in the “usual way” is certainly not well determined. I believe there probably was an accelerating descent ongoing then, but the ROD may not be nearly as large as ~14,000 fpm.

   There is a peculiarity in BFOs for the “log-on/log-off acknowledge” messages only. This is demonstrated by the odd values at 18:25:34 (+273 Hz) and at 00:19:37 (-2 Hz). To the best of my knowledge, no one at Inmarsat or ATSB has any explanation (even now). If Honeywell knew, I assume ATSB would also know. Unfortunately, it is not evident that any substantial effort has been made to figure this out or even to replicate it as an experiment. I believe that if this were done successfully, some more useful information on ROD at 00:19:37 could be a result. As far as I know, Honeywell has never gone on record as to the cause.

   Any theory for the “special processing” that is done in the SDU, for this type of message only, can be tested using the 08:25:34 BFO when the flight path is fairly well constrained by the other BTO/BFO data at 18:25-18:28. No theory of anomalous BFO I have seen, including my own, passes this test so far. It is likely that the BFO shift is aircraft speed dependent, because the shift in BFO is zero when the aircraft is stationary at the gate (16:00:13). The shift (between the BFOs for each pair of log-on/log-off messages) was +135 Hz at 18:25 and -184 Hz at 00:19. Saying it is aircraft speed dependent also implies it depends on the magnitude of the compensation term. There might be a clue in the BFO shift amounts. When the compensation term is 0 Hz, the BFO shift is also 0 Hz (at the gate at 16:00). When the compensation term is +650 Hz at 18:25, the BFO shift is +135 Hz (both positive). When the compensation term is -500 Hz (at 00:19), the shift is -184 Hz (both negative). You can’t draw a perfectly straight line through these three points, but it’s not a bad fit, either, especially when you consider that the 00:19 BFO shift is likely changing anyway to some degree because of the ongoing descent. The plot seems roughly consistent with a compensation error that results in underestimation of the correct frequency compensation magnitude by about 20%. If I take a line through the 16:00 and 18:25 points, it predicts one would get a -104 BFO shift with an unchanging ROD. That can be interpreted to imply that roughly -80 Hz of the total -184 Hz BFO shift is actually due to increasing ROD, and this implies a ROD change of roughly 4,500 fpm in addition to the 00:19:29 ROD of about 4,600 fpm for a net ROD of 9,100 fpm at 00:19:29. This result implies an average vertical acceleration downward of 0.29 G’s during the 8 second time interval. I think this method provides a better estimate than simply ignoring the peculiarity of the BFOs for that message type.

2. @all
   Fyi there was on French television tonight (France 2) a 1 hour investigation on mh370 in the “Envoye Special” programme. They normally put their programmes on the web for public viewing the day after, I believe the website is http://www.pluzz.fr
   They interviewed family members, colleagues of zaharie, military from the butterfield air base, followed blaine gibson in madagascar and the maldives etc
   Not a lot was new information. Intereeting facts/conclusions that they come up with (although I’m sure this is nothing really new to you):
   A) they have played cctv footage of the k. Lumpur airport (which they claim has never been shown before on television) showing that zaharie and his co pilot both made the metal detector ring but were only very superficially searched
   B) they claim that drift analysis shows that if the aircraft had been in the current search zone it is not normal that all debris has been found on african coasts and nothing in australia
   C) they say there is no proof at all of a radar track showing the plane go west at penang
   D) they interview people in the maldives which they claim have never been interviewed before and who say they saw a wide body flying low over the island in a place where there is never such a big plane and where there was no water plane in the air at the time of viewing
   E) they criticise the fact that no investigators are setting up a thorough search on the african coasts.

   I hope this was a decent summary, I leave it to you whether you want to have a look at the programme yourselves (it’s innFrench).

3. @DrBobbyUlich: Very good comments. Thank you.

4. @DrB

   Nice. Good food for thought.

5. Back on Oct. 1, 2014 I posted the following:
   <>
   One reply stated that as these flights approached the longitude of the Satellite the data is not very accurate in predicting the a/c location.

   Also, I think it was DennisW who pointed out that the BTO & BFO biases differed from flight to flight. If true, why in the world would someone assume an unchanged bias after a reboot? Maybe this wasn’t an intentional spoof but a variation of the bias after each reboot placing MH370 outside of the searched area?

6. @Matt Moriarty,

   You said: “We also know from Honeywell that EOR LNAV is HDG HOLD in either MAG or TRUE depending on what is selected at the time of EOR.”

   That is exactly what Honeywell has told others, and it may be correct, but I have my doubts. I am doubting this for 3 reasons:

   (1) Victor’s PC flight simulator results show True Heading is maintained. Where would they get that if it were not correct?

   (2) Victor’s experiments also demonstrated the difficulty in simulator testing in order to get a reliable result. You must continue the route until significant magnetic variation occurs, and this can be lengthy. For example, your test in the B777 simulator was not long enough to distinguish between Magnetic Heading and True Heading being maintained. Possibly Honeywell did the same thing you did (and that Victor did initially) and misinterpreted the results.

   (3) There is no Constant Magnetic Heading route that matches BTO/BFO/Wind/PDA. There is one Constant True Heading route that does so.

   Just because the display shows MAG HDG after EOR does not mean that is what is being controlled to stay constant. Of course the number “displayed” changes when you change the NORM/TRUE switch. It should always do that, but to say which one is “controlled” is determined by that switch setting seems very odd to me.

7. Oops here is the text of the Oct. 1, 2014 post:

   “Per the ATSB report, Inmarsat used the March 7, 2014, BFO data to predict the flight paths of MH021 and MH009. Presumably, they were able to calibrate their BFO model to the known actual location of each of these flights. At the point where these flights crossed the 7th Arc, their predictions were off by 145nm and 200nm, respectively.”

8. At second engine flame out, there would be no more pressurization air from the pacs.

   Cabin altitude would begin to bleed up.

   In a “normal” cabin, with outflow valves working, they would run closed and the cabin would only slowly climb, initially, but would eventually get to alarm level.
   When the alarm goes off, the pilot knows there is no way of maintaining cabin pressure, so he has to go for his mask, and would probably commence an emergency descent anyway.

   What if the pilot “overdid it” a bit and pitched down too much, and it “got away from him” a bit, leading to excessive speed, and excessive rod, which he may (or may not) have eventually recovered from.

9. @Mph
   We they suggested no proof of west after Penang, do you mean no proof of radar path up Malacca Straights? Others have said this.

   @Lauren
   Have you previously given your path? I would be interested.

10. @tbill
    You are correct they were suggesting there is no proof of a radar path up the Malacca straights.
    The more so as they claim the plane was not picked up by indonesian radar

11. It’s sad seeing journalists get rolled like this. The same thing happened to Florence de Changy. They get in over their heads and they don’t stand a chance.

12. @mph

    Did they give any reasoned suggestion as to where it went if it didn’t go up the Malacca straights?

13. @crobbie
    No not really, they just seem to make the suggestion if I understood correctly that it is not at all unlikely that the plane ended up east of diego garcia, 5000 km north of tye current search area, and that that location would be consistent with all the debris washing up on the african shores and with what the eyewitnesses in the Maldives saw.
    I do not have the expertise many of you have, but this hypothesis was already debated and discarded isn’t it?

14. @Mph

    “this hypothesis was already debated and discarded isn’t it?”

    Yes, along with a couple of posters overly passionate relative to that view. You are in Andre Milne territory with that one. Not at all a good place to be.

15. @Ventus45
    “What if the pilot “overdid it” a bit and pitched down too much, and it “got away from him” a bit, leading to excessive speed, and excessive rod, which he may (or may not) have eventually recovered from.”

    Well, that’s one thing I like about Victor’s 180S Mag Heading cases with the constant descent, it would be below critical altitude by the time it ran out of fuel. I was going to say the pilot could control the dive, but you are saying after outta fuel (FS9 simple 777 model probably allows a pull out).

16. @DennisW, @TBill, FAA regulations stipulate (!4CFR 121 Subpart R) that flights exceding 8 hours require 2 pilots plus 1 additional flight crew member, unless MAS had a Fatigue Risk Management System in place that would fulfill a list of requirements. Depending on number of PAX and weight, an average flight to Jeddah from KL on a 777-200ER would be 9.15-9.30 hours and return about 8.45hours. This could mean that flights to Jeddah included an additional flight crew member. I was not able to determine if MAS had a FRMS in place at the time to do the haul with just 2 pilots. MattM can confirm whether the time pilots need to report in before a flight is included in the 8 hour restriction. If the answer is yes, it would be safe to asume that the Jeddah flights included 2 pilots and 1 additional flight crew member.

17. @Johan, The National Defence Univerity of Malaysia (NDUM) has centres that focus on 1. niche critical defence to fulfil defence policy needs. All products from research are used for the Malaysian Air Force (MAF) and 2. Centre for Defence and International Security focusses on research of defence and security needs for the Ministry of Defence (MINDEF). The centre works with, amongst others, South East Asia counter terrorism (SEACCT) and has close relations with ASEAN defence and security institute (NADI). Anyone wanting to do their homework as it relates to Military Radar and systems available to them, security in the area in general and Indonesia and Thailand in particular would perhaps be able to obtain it from them? The number of calls in and out, stick out IMO.

18. @VictorI,

    You said: “Not to mention that you are now saying it is acceptable to consider a loiter in the flight path.”

    Not I didn’t say that. I said it is conceivable that there could have been an aborted or unsuccessful attempt to enter a Holding pattern at ANOKO. I don’t think one occurred, but if my CTH route is correct, the best holding speed was set near ANOKO. I know it is possible to do this using the MCP, but I was searching for a rationale for the FMC to do it, and it may not exist. Still, I will try to figure it out.

    Another item that mystifies me is the ~12NM SLOPs (Strategic Lateral Offset Procedures) to the right of N571. The normal ICAO standards are 1 NM or 2 NM to the right. Why 12 NM? Does anyone have any ideas? You can enter 1 to 99 NM and Left or Right.
    There is also an interesting part of the Honeywell FMS Pilot’s Guide that says the following:

    “The offset entry propagates through the remaining flight plan up to any of the following:
    – The end–of–route waypoint
    – A discontinuity
    – The start of a published STAR transition or STAR or approach transition or approach procedure
    – A DME arc
    – A heading leg
    – A holding pattern
    – A course change of 135 degrees or greater.”

    I interpret this to mean if ANOKO was an EOR, the offset would stop when ANOKO was reached. The next question is whether the N571 lateral offset propagated to the route to ANOKO from N571 during the FMT. The turn was less than 135 degrees, so that would not end the offset. That route won’t fit through a point 12 NM west of ANOKO, so I don’t think the offset continued after the FMT. Perhaps it was simply set to zero NM as shown in the Pilot’s Guide to disable it.

    But there is a second interesting note in the Pilot’s Guide:

    “NOTE: The aircraft does not enter a pre-planned holding pattern if it is on an offset path.”

    This seems clear enough in indicating the holding pattern is not flown, but it does not offer any information on what the FMS does instead. I presume it would continue using the LEGS page, but if ANOKO is the last entry, wouldn’t that cause an EOR error when it reached ANOKO?

    I am confused by the apparent inconsistency of those two statements: (1) The offset propagates up to a . . . holding pattern, and (2) The aircraft does not enter a pre-planned holding pattern if it is on an offset path. How can you ever get #2 if #1 does what it literally says? Sounds like Catch-22 to me.

    So here is a straw-man scenario for an aborted Holding pattern that might accomplish both the best Holding speed and the passage to ANOKO:

    1. Flying along N571 just past MEKAR at FL350 and LRC
    2. At about NILAM a 12 NM Right SLOP is performed
    3. About 12 minutes later, a holding pattern is set that uses ANOKO as the fix
    4. The other parameters to define the Holding pattern are not material for this scenario
    5. The speed is set to the best holding (258 KIAS) (I’m not sure whether this speed change happens right away or when the fix is reached; does anyone know?)
    6. The lateral offset is continued on the course to ANOKO; it is not zeroed
    7. At ANOKO the holding pattern is aborted because of the lateral offset
    8. An End of Route error also occurs at ANOKO, setting the LNAV to maintain heading
    The FMS may not actually work this way because of something I have missed or misunderstood. Perhaps you or others who are more knowledgeable about the FMS can point out the flaws.

    The key questions I have are:

    1. Why a 12 NM SLOP?
    2. If a holding pattern is set up with a fix, does the holding speed become effective when the fix is reached or when the holding pattern entry command is executed?
    3. Will the lateral offset on N571 continue on the path to ANOKO?
    4. If the offset continues to apply to the path to ANOKO, is the holding pattern executed or aborted when ANOKO is reached?

19. @VictorI, I realise that the lateral offset – continue WNW – FMT is the preferred solution to achieve BFO and BTO fit. However, there is a much simpler solution if we set aside BFO; namely an earlier FMT.

    When you look at the (3rd order) polynomial fit of points 1941-0019 and project back to 1825 to 1827 you find that the 1825-27 prediction is about 100 microseconds higher than observed.

    When you look at 1822 you find it is “too close” to 1825 ring for aircraft to have continued WNW. And the same is true for 1802 if steady speed is maintained (whereas 1822 is consistent with 1802 for both speed and direction).

    Now, bearing in mind that BFO 1825-27 might not be reliable a) possibly we are seeing a themal-dependent oscillator pattern – the pattern certainly bears strong resemblance https://www.dropbox.com/s/u2oy8gwfxvqfhpr/BFO%20vs%20time%20at%20logon%201825.xlsx?dl=0
    b) a change in BFO offset for new log-on of the sort described by Bobby U above.

    So, now considering BTO alone, what can we glean 1825-27? The answer is that it bears closer resemblance to a southerly path (expect ~ -0.2 microseconds per second) than a WNW path (expect ~ -0.9 microseconds per second). By my calculations, if we believe all BTO points inclusive R600 with its offset, then it CANNOT be a continued WNW path 1825-1828 with confidence limits on BTO of +/- 40 microseconds. Predicted delta BTO 182527-182815 = -150; observed delta = -40 (or -120 if first and last points BOTH out by 40 microseconds in opposite directions).

    So rather than moving the aircraft “further north” with a lateral offset to hit the 1825-1828 BTOs, we could match BTO prediction instead by FMT occurring just before 1825.

    This way:
    – 1822 radar position can be correct, both radial and range
    – 1825-1828 BTOs can be correct level/timing
    – delta BTO slope 1825-28 in line with FMT having already occurred.

20. edit: scratch the first statement:
    “When you look at the (3rd order) polynomial fit of points 1941-0019 and project back to 1825 to 1827 you find that the 1825-27 prediction is about 100 microseconds higher than observed.”

    That rather argues in favour a turn ~200 seconds after ~1827

21. @KarenK:
    The numbers do stick out. But I would assume it has to do with pilot education, wouldn’t you? Either they are using their simulators or teachers or courses or simply their facilities or examination hall — and administrator. Or Z was giving lectures or cooperated with a teacher concerning education. If he’d be talking to an administrator about dates and times and payable hours for supervising students, then you have it right there. It is most likely unrelated in its detail in any case.

22. @RetiredF4

    From Factual Information
    “The Mode S symbol of MH370 dropped off from radar display at 1720:36 UTC, and the last secondary radar position symbol of MH370 was recorded at 1721:13 UTC.”

    So from above, the Mode S portion failed 37 seconds prior to total failure of the transponder.
    Now compare this to the modes transmitted for each position of the Transponder Mode Selector.

    STBY (standby) – transponder not active

    ALT RPTG (altitude reporting) OFF –
    transponder enabled in Modes A and S
    altitude reporting disabled

    XPNDR (transponder) –
    transponder enabled in Modes A, C, and S
    in flight, altitude reporting enabled.

    As we can see, there isn’t a selection possible on the transponder which would disable Mode S but still leave the transponder transmitting for 37 seconds.

    Therefore we can conclude that the transponder was not switched off by the Transponder Mode Selector in the cockpit. The failure had to occur in the E&E Bay.

    And what happened in the E&E Bay prior to departure? The crew oxygen bottles were topped up.

23. @OXY

    Several have noted this, there is a long discussion in the thread below, and perhaps this is interesting:

    http://www.airliners.net/forum/viewtopic.php?f=3&t=585987&start=200#p9009935

24. @DennisW:
    Dan M. Kahan is on a Swedish lunch television forum right this minute. I believe they are talking about the role of state TV and internet in “post-fact society.” One issue is if search engines and their dynamic will contribute to a politically polarized society or not and what should be done about it. And the role of fake news, fake information, quasi information, conspiracy theories and the like. Sane and sound British nutrition-myths fighter Michael Mosley is there too. It is “surprisingly” interesting to watch.

    But they are all fighting ghosts with clubs, spiked clubs and chains, as one would expect.

25. @DrBobbyUlich:
    1) The 12 nm offset might have been to create further distance from Indonesian airspace, which looks to me was an objective. In your scenario of a possible hold at ANOKO, it makes no sense.
    2) The wording you cited regarded offsets and holding patterns is confusing. However, notice though that in Statement 2, there are the words “pre-planned”, i.e, part of the existing route. Also, understand that a holding pattern may be entered at any time by choosing PPOS (present position) as the fix. So if this is done, the offset of the route is removed, per Statement 1. On the other hand, if the holding fix is already part of the route, the hold is not entered when flying an offset, per Statement 2. I can’t be sure of this without trying it in FSX, but I suspect this is the proper way to interpret the words to avoid your “Catch-22”.
    3) You asked when the speed would be changed to the best holding speed. If flying in SPD mode, the speed would have to be manually selected via the MCP. If flying in VNAV mode, the autopilot would automatically try to achieve whatever target speed (and altitude) information is included in the route.

    Again, if a landing at Banda Aceh (WITT) is contemplated, I see no rational reason for remaining at cruise altitude.

26. @Paul Smithson said, “When you look at 1822 you find it is “too close” to 1825 ring for aircraft to have continued WNW. And the same is true for 1802 if steady speed is maintained (whereas 1822 is consistent with 1802 for both speed and direction).”

    That’s not consistent with my calculations. I believe that if you extend the radar-determined position at 18:22 on a track of 296 deg and groundspeed of 495 kn, you come close to the 18:25 BTO. This is described in a slide I put together in April 2015 when I first proposed the lateral offset manoeuver as way to reconcile the BTO and BFO. (In that slide, I tried to explain the spike in BFO as a short, steep climb. That probably did not occur.)

    https://www.dropbox.com/s/t0w3mrfgin4qsv3/Lateral%20Offset.jpg?dl=0

    The fact that the first and last BFO values in the 18:25 to 18:28 interval agree leads me believe that those values are valid, and are consistent with a track parallel to N571 at a groundspeed of about 495 kt.

    But if you throw out the BFO values in that interval, then you can certainly have an FMT after 18:25 and before 18:40 that matches the BTO values.

27. @Oxy: Do we know for sure that if the Mode S response from the transponder omits the altitude data that the Mode S symbol remains on the ATC radar screen?

28. @VictorI. Having just constructed time-specific ping rings (using aqqa ping model but PAR5 sat ECEF) I concur that the 182527 ping ring is not (as per my previous post and as DSTG asserts) “too close”.

    By my reckoning, 182527 projected position from 182212 produces a BTO of 12560 – 40 microseconds too high but definitely in the ballpark.

    However, if you continue on a course of ~290 at 500kts you will find that your 182815 BTO cannot be within limits no matter what. You will end up on ~12420 instead of the 12480 observed.

    I’m therefore of the view that BTOs on their own provide strong indication of a turn taking place ~1826 (not earlier than 1825 and not later than 1827).

    It is interesting to note that this would place FMT
    a) corresponding with beginning a Banda Aceh approach 2-Charlie via ANSAX.
    b) implies a system power-up immediately preceding or at the same time as turning to commence an attempted approach to Banda Aceh.

    ps: I’m aware that we cannot be perfectly confident in 1822 radar position (or other uncertainties on ping ring calculus/assumptions). However, whichever way you cut it you cannot cross the ping rings 1825 to 1828 at ~500kt GS and course of ~290 without busting BTO confidence limits on one side or the other.

29. @DrBobbyUlich: I looked at your proposed scenario on SkyVector. If I am understanding it correctly, you continue on N571, with offset, until almost IGOGU, then turn towards ANOKO. First, if in LANV, it would be much more likely that IGOGU was chosen as the waypoint, in which case the final track would be 180T from IGOGU to ANOKO. Second, you talk about an ANOKO STAR to Banda Aceh (WITT). However, at NILAM, somebody contemplating a landing at WITT would track towards ANSAX, not ANOKO. It makes no sense to continue on N571 towards IGOGU. Not to mention continue at cruise altitude.

    I know you proposed the scenario as a straw man, and you are determined to find a rationale for this scenario, but I really don’t see one.

30. @OXY
    The EE Bay O2 cylinders are filled all the time in many aircraft. I am not aware of any history of failure of the synthetic (Kevlar) type gas cylinders in MH370. I know the Qantas historic failure but I believe it had a metal cylinder.

    I do consider the EE Bay entry before the flight to fill O2 as a potential sabotage activity, if there was some kind of conspiracy. One could argue prior flight drained O2 to less than 1200 psig to trigger need for refill. Maybe some circuit breakers switched at the time.

31. @Johan

    “One issue is if search engines and their dynamic will contribute to a politically polarized society or not and what should be done about it.”

    Don’t really have an opinion about that, but I do know that AI, and AMZN AI in particular, has gotten incredibly good. Hardly a week goes by that I don’t one click something from AMZN because their algorithms have me so perfectly profiled.

    What I do know for sure is that Trump’s tweets have cost me a lot of money over the last month or so. Surprisingly, despite being generally well-educated, my friends are mostly anti-vaxxers, and seem to be enjoying the experience.

    Definitely a new world out there in terms of how information flows and is interpreted.

    Back on topic i.e. @PaulS/VictorI:

    I am confident relative to the 18:25 aircraft position, speed, and track. All the data, ISAT, radar, phone registration,… points to a location of 6.8N 95.9E at a track of 296 degrees with a GS of 510 knots. IMO, this is the last time we can be very sure of where the aircraft was, the GS, and track.

    Additionally, the accuracy of the BFO data at this time virtually rules out the possibility that a decompression event occurred before 18:25.

32. Dennis W. Are you equally confident that 1828 is still on the same track / speed. I don’t see how it can possibly gel with BTO confidence limits.

33. @VictorI
    Do you still favor lateral offset move? and what other route plans could help avoid Indonesia radar? Seems to me approach or FMT near Banda Aceh gets too close.

34. @Paul

    “Dennis W. Are you equally confident that 1828 is still on the same track / speed. I don’t see how it can possibly gel with BTO confidence limits.

    18:28 BTO’s are clearly flawed, IMO.

35. @TBill: The lateral offset could have been added to the route to keep MH370 far from other traffic, especially since TCAS was not operating. If Indonesian airspace was avoided, a turn that encroaches on it makes no sense. The route was probably chosen to minimize a response from Indonesia, not necessarily to stay out of radar range. All supposition, of course.

36. @Oxy
    “The Mode S symbol of MH370 dropped off from radar display at 1720:36 UTC, and the last secondary radar position symbol of MH370 was recorded at 1721:13 UTC.”

    While a failure mode in the transponder system can not be excluded, there are other explanations as well. The timeline might be an inaccuracy in the FI report, and we know there had been a few.

    On Page 40 of the FI it says:
    “The transponder of the occurrence flight was operating satisfactorily up to the time it was lost on the ATC radar screen at 1721:13 UTC, 07 March 2014 . There was no message received from the aircraft to report a system failure.”

    We may conclude, that the drop off the mode S symbol is a normal occurance, which sometimes happens, that they got the timeline wrong or that the happening at 1720:36 was not the drop from mode S, but the drop of the altitude information. Another explanation might be that the the time gap between 1720:36 and 1721:13 was due to coasting effect in the ground system.

    I´ve copied the following link some time ago, and I can not verify its genuine origin at the moment. Anyway here it is within a dropbox available, and I do not know the holder of that dropbox.

    https://www.dropbox.com/s/9n85w9rvfqxpwza/mh370.PNG

    It shows all mode S-data until 1721:03. Data which would not have been available on mode A alone.

    @VictorI
    “Do we know for sure that if the Mode S response from the transponder omits the altitude data that the Mode S symbol remains on the ATC radar screen?”

    I would guess so. But it would bear some logic, that the loss of altitude data would be displayed by a change of the symbol.

37. @DW. Now you have got me puzzled.

    Why on earth believe 1828 flawed while believing in the integrity of the BTO as a whole?

    I’d be more sympathetic if you said 1825 was dodgy since it is immediately after logon and on a channel (R600) for which we have to make an offset assumption.

    Moreover we have a bunch of adjacent, mutually compatible BTOs in the 1827-1828 time window. Why on earth would you think that 1828 is flawed??

38. As long as he stayed outside of the Indonesian FIR boundery he would be save. Also if Indonesian radar tracked him crossing the MY peninsula and after. They would not interfere for it was in Malaysian airspace not flying towards Indonesia.

    As I remember well they only declared (Indonesia): ‘we did not see the plane in OUR airspace’. That’s all they said. Never heard a statement they did not see it anywhere else.
    Missed something?

39. @VictorI,

    Thanks for responding to my comments on the FMT route.

    You said: “First, if in LANV, it would be much more likely that IGOGU was chosen as the waypoint, in which case the final track would be 180T from IGOGU to ANOKO.”

    Much more likely? I disagree. Why would you pick a waypoint very close by that has no rationale (IGOGU) versus one that is a STAR point (ANOKO)?

    You also said: “However, at NILAM, somebody contemplating a landing at WITT would track towards ANSAX, not ANOKO. It makes no sense to continue on N571 towards IGOGU.”

    I agree with your conclusion but not your assumption. I don’t think that decision to make a southward turn was made at the time the SLOP was made (near NILAM). If so, why do the SLOP? The rationale for the southward turn appears to be driven by a separate intention that came about later in time.

    Here is a map which shows the lateral offset maneuver:

    https://drive.google.com/file/d/0BzOIIFNlx2aUODYxY252ZTZrcGs/view?usp=sharing

    At the time of the end of the SLOP (18:28:10) the distance to ANSAX is 53 NM and to ANOKO is 96 NM. If one wanted to land at WITT then, certainly the ANSAX2C STAR would be the shortest. However, at the time of the start of the southward turn (18:37:38) the distance to ANSAX is increased to 113 NM and to ANOKO it is reduced to only 33 NM. So by that time ANOKO is much closer than ANSAX and an ANOKO2C STAR would be the shortest route to WITT.

40. @DennisW:
    Ah yes the pharmaceutical stocks are falling due to Trump coming to office. Well isn’t that like Hitler coming down on the munitions industry? They will come to terms eventually, be sure of that. What will be lost on vaccinations will be earned back from band aids. 🙂

41. @Ge Rijn

    I agree. Also worth noting:

    “another military radar suggestion said that it was once detected in the Andaman Islands. So, it could very likely have cleared Sumatra island in the north before making another turn to the south until it was 2,500 kilometers from Perth.”

    http://www.thejakartapost.com/news/2014/03/26/govt-insists-mh370-did-not-fly-indonesian-airspace.html

    To my mind, this could amount to the “hook around Sumatra” scenario, entertained in the very first ATSB report, in which the person in control of the a/c flew manually on headings (for want of suitable waypoints), monitoring the distance to Sabang and decreasing altitude during the 18:40 phone call to reduce line of sight for the southern turn).

42. @Paul Smithson,
    @DennisW,
    @VictorI,

    Here are plots of BTO and BFO from 18:22 to 19:00 showing the lateral offset maneuver:

    https://drive.google.com/file/d/0BzOIIFNlx2aUa2h4YjU1SnVXMTg/view?usp=sharing

    All BTOs and BFOs are shown here except for 18:25:34 (i.e., the 273 Hz).

    The agreement is quite good. I don’t see any inconsistency at 18:28. I use a turn rate that produces a 25 degree maximum bank angle at the current speed (LRC at FL350).

43. @Dr Bobby – if you use confidence intervals of +/- 70 microseconds then sure, there’s no problem

44. @VictorI,

    I appreciate your looking at my straw-man scenario.

    In response you said: “On the other hand, if the holding fix is already part of the route, the hold is not entered when flying an offset, per Statement 2. I can’t be sure of this without trying it in FSX, but I suspect this is the proper way to interpret the words to avoid your “Catch-22”.”

    It would be very useful if you would put a “pre-planned” holding pattern in the route and approach it with a lateral offset to see what FSX does, and whether it flies the holding pattern or has an EOR error at the holding fix.

    You also said: “You asked when the speed would be changed to the best holding speed. If flying in SPD mode, the speed would have to be manually selected via the MCP. If flying in VNAV mode, the autopilot would automatically try to achieve whatever target speed (and altitude) information is included in the route.”

    I’m not sure how to interpret this in terms of the timing. My question was would the holding speed get applied (using a “pre-planned holding pattern) before the holding pattern fix is reached (on the leg approaching it), or would the slow-down occur at the fix itself? If the holding pattern turns out to be “aborted” when approached with a lateral offset in place, does the holding speed get set anyway?

45. While the detection of 9M-MRO by Indonesian military long-range air defence surveillance at Lhokseumawe or Sabah is possible, the consequence of surveillance is zip.

    The TNI-AU certainly did not have an intercept capability in Mar 2014. Their nearest base of capable aircraft (F-16s) was Pekanbaru, too far for an unaided (by AAR) intercept of an aircraft flying away from ID territory,

    Any consideration for a landing at Sultan Iskandar Muda International Airport, Banda Aceh, is contradicted by the simple fact that ID’s aerodrome information publications state that the airfield is unmanned & navaids are unavailable during non-operational hours (during hours of darkness).

46. @Nederland
    Good article, if I recall some observers believe it was the Singapore AWACS plane that might have made that Andamans siting. I wonder if the Singapore AWACS plane would stay on standard flight paths like N571.

    Has anyone found a way to include Andamans in a ping-compliant path? Makes perfect sense to me to fake north and go 180S at around 92 East.

47. @TBill

    I think Indonesia was referring to their Sabang radar:

    “Indonesian Defense Minister Purnomo Yusgiantoro said the Indonesian military radar placed in the country’s western-most city of Sabang did not detect an airplane flying over Indonesian territory.”
    http://www.antaranews.com/en/news/93270/indonesian-military-radar-did-not-detect-missing-airplane

    This seems to be echoing the words used in that other article. It would be highly misleading if that radar was turned off at the time.

    Sabang radar is in Maimun Saleh (waypoint WITN)
    http://airport-flights.net/airport/maimun-saleh-airport/

    Campbell Bay, on the other hand, is a small outpost with no fighter jets and reasonably little radar surveillance, Port Blair too far north to matter.
    http://thediplomat.com/2014/03/andaman-and-nicobar-islands-indias-strategic-outpost/

    That route is also fairly similar to the path found on the flight simulator.

    The southern route could be close to the flight paths suggested in the first ATSB report (“northern hook around the tip of Sumatra”, p. 26)

    http://www.atsb.gov.au/media/5243942/ae-2014-054_mh370_-_definition_of_underwater_search_areas_18aug2014.pdf

    So, presumably it is possible to come up with a valid flight path.

48. The Indonesian statements on their radar were very convoluted.

    As I remember it, (and I may be wrong) they did say that they tracked the aircraft on it’s way to Igari, (info from MEDAN Radar) but they never said ANYTHING about tracking it westbound.

    They were very specific, they said (essentially) that “they did not see it where the Malaysians said it was”.

    They never actually said that “we did not see it westbound AT ALL”.

    I think that the “riddle” of the Indonesian statements is the big elephant in the room, that no one has ever solved.

49. @ventus45

    My interpretation of Indonesian radar statements is the following.

    Medan radar is the only one mentioned in the FI. Medan tracked MH370 up to IGARI with its transponder on, but not thereafter with its transponder off, according to the FI.

    Aceh primary radar was either switched off or on short range only and did not detect MH370.

    Sabang primary radar was active and monitored, but a week later the operator could not ascertain whether or not he did detect MH370 outside of Indonesian territory (which is none of his business), although he could recall perhaps having it spotted close to Campbell Bay, but he could confirm he did not see it in own airspace. The data were not recorded.

    The deduction is that “the most likely route taken by the plane was via the northern Indian Ocean, west of Sumatra, then heading south to a location Najib indicated was the plane’s crash point.” (outside of Sabang radar coverage)
    http://www.thejakartapost.com/news/2014/03/26/govt-insists-mh370-did-not-fly-indonesian-airspace.html

    This would amount to a crash further north on the arc, which is also consistend with drift modelling.

50. @RetiredF4

    Mode S does not depend on what information has been transmitted, it depends on the way it was transmitted. If the received signal is in Mode S format it will be displayed.
    There are SSR Transponders with Mode S which have neither aircraft identification nor pressure altitude capability.

    As far as coasting effect (radar track prediction), the Factual Information report is silent on this.
    Imagine if they released the actual playback of the Malaysian ATC SSR radar and it shows a Mode S coasting period to 1720:36, then followed by another coasting period to 1721:13.
    That would mean a hijack is extremely remote and that the maintenance of the oxygen bottle is the prime suspect.
    And who would be responsible?
    And who is not releasing the playback?

Comments are closed.