ATSB Denies Request from MH370 Families for More Info

MH370 Search Director Peter Foley (left) and ATSB Chief Commissioner Greg Hood (right)

In a new article from the Australian, we learn that Greg Hood, Chief Commissioner of the ATSB, has rejected a Freedom of Information (FOI) request from the families of Chinese passengers of MH370.  (Commenter @David alerted us to the story.) According to the story, “The documents sought are the opinions of international experts, including from the US and British air crash agencies, Boeing, aerospace group Thales, and British satellite group Inmarsat, about satellite data that automatically tracked the course of MH370.” Also, we learn that  the “ATSB general manager for strategic capability Colin McNamara in February refused The Australian’s original FOI ­request, claiming release of the ­information could ’cause damage to the international relations of the commonwealth’.”

So why would the information about MH370 requested by The Australian cause damage to international relations?

With the failure of the underwater search to locate the wreckage of MH370, it is essential that we reassess what assumptions went into the analysis that was used to define search area. This will help us to understand why the search failed, and could help us to define a new search area. A true reassessment would require independent eyes to review and analyze all the available data. While certain data sets have been publicly released, and other data sets have been leaked to the public, there are still important data sets held by the official investigators that are not available to the public. Unfortunately, the recent decision by Mr Hood reaffirms the prior decision to not make these data sets publicly available.

Included below is the complete article.

ATSB shuts down details on MH370 search

by Ean Higgins

The Australian Transport Safety Bureau has invoked draconian legislation in refusing to release material about its search for missing Malaysia Airlines Flight MH370, warning that any bureau ­employee who provides such ­information to the public or a court could face two years in jail.

ATSB chief commissioner Greg Hood has used the statute to reject a plea from the families of the Chinese passengers who died on MH370 that he grant a Freedom of Information request from The Australian, with the families claiming failure to do so makes Australia complicit with a cover-up by the Malaysian government.

Some ATSB officers are having second thoughts about the agency’s official line that MH370’s ­pilots were unconscious or dead at the end of the flight.

Mr Hood has declared the Transport Safety Investigation Act covers the FOI request for critical documents the ATSB claims support its “ghost flight” and “death dive” scenario, which holds the Boeing 777 went down in an unpiloted crash.

The theory has been rejected by many commercial pilots and international air crash investi­gators who believe captain Zaharie Ahmad Shah hijacked his own aircraft and flew it to the end.

The documents sought are the opinions of international experts, including from the US and British air crash agencies, Boeing, aerospace group Thales, and British satellite group Inmarsat, about satellite data that automatically tracked the course of MH370.

The ATSB says the satellite data shows MH370 was in a rapid unpiloted dive at the end, but experts such as former US captain and crash investigator John Cox have said the data is not good enough to reach that conclusion.

ATSB general manager for strategic capability Colin McNamara in February refused The Australian’s original FOI ­request, claiming release of the ­information could “cause damage to the international relations of the commonwealth”.

The association representing the families of the 153 Chinese ­victims who died when the plane went down on March 8, 2014, ­issued a statement after The Australian reported Mr McNamara’s decision, saying “we react with extreme displeasure and ­annoyance”.

“Is avoiding offending the ­Malaysian authorities more important than discovering the truth?” the families asked in the statement.

Mr Hood, in an internal ­review of Mr McNamara’s decision, also refused to release the documents. “The activities of the ATSB with respect to assisting the Malaysian investigation are covered by the TSI Act,” Mr Hood wrote in his decision.

He advised that the act holds that if a serving or former ATSB staffer or consultant “discloses ­information to any person or to a court; and the information is ­restricted” they have breached the act, which stipulates a penalty of two years in prison.

In response to an earlier ­inquiry, Mr Hood would not say whether he would allow any ATSB staff who no longer agree with the “ghost flight” and “death dive” theory to publicly express their views.

MH370 disappeared on a scheduled flight from Kuala Lumpur to Beijing with 239 people on board, with its radar transponder turned off and radio communication cut after about 40 minutes.

Military radar and the satellite tracking data shows the aircraft deviated back over Malaysia to the Andaman Sea, before a long track south to the southern Indian Ocean. A $200 million search directed by the ATSB based on its “unresponsive pilots” theory failed to find the aircraft’s wreckage and was suspended in January.

When last year it was revealed the FBI had discovered Zaharie had plotted a course quite close to that track on his home computer flight simulator, the ATSB joined the Malaysian government and Malaysia Airlines to hose down suggestions this pointed to the “rogue pilot” hijack theory.

Update 1 on April 18, 2017.

A group called “The MH370 China Families” is distributing a media release that reacts to the FOI denial by the ATSB and the claim by Malaysia authorities that nothing is being hidden. They state that “In the face of a failed search outcome, contradictory evidence and Malaysian authorities’ overall approach, China families reject the Malaysian transport minister’s denial.”

The group also continues to claim that the satellite data was altered to support the theory of a crash in the SIO, citing comments posted by Emil Enchev on a physics blog site. The comments have been removed, so we don’t know the basis for his claims. I have in the past asked the group to publicly release this evidence so that it could be properly reviewed. They refused my request, citing concerns about the safety of the crew and passengers of MH370, who they think could still be alive, and could harmed by their captors if Mr Enchev’s comments were released. I explained to them that their claims could not be taken seriously without supporting evidence.

Here is the media release:

Media Release: For immediate release: MH370 China families react to Malaysian minister’s denial authorities have something to hide

China family members react with displeasure at the claim by Malaysian transport minister, Liow Tiong Lai, that the authorities have nothing to hide. Instead we point to the following indicators that they are hiding.

The Australian Transport Safety Bureau, investigating MH370’s disappearance at the request of Malaysia, refused to release information under The Australian newspaper’s Freedom of Information request, claiming the release could “cause damage to the international relations of the commonwealth”.

The Southern Indian Ocean seabed search based on satellite data supplied by Inmarsat and the Malaysian government failed to find any evidence of MH370. The search was coordinated  by the ATSB October 2015 to January 2017.

China families have access to a physicists’ blog site post 28 May 2014 claiming the satellite data was altered prior to its release 26 May 2014 to support the claim by Malaysian authorities, 24 March 2014, that MH370 crashed in the Southern Indian Ocean.

The internal French report on the Réunion flaperon found that the flaperon was entirely submerged * and yet French flotation tests showed much of the flaperon above the waterline, raising serious doubts about the genuineness of the debris.

Reverse drift analyses are incompatible with the primary search zones determined by expert analysis of the purported satellite data.

Malaysian authorities are unwilling to explore the area of 25,000 sq kms, after which, according the the “First Principles Review”, November 2016, “prospective areas for locating the aircraft wreckage, based on all the analysis to date, would be exhausted.”

In the face of a failed search outcome, contradictory evidence and Malaysian authorities’ overall approach, China families reject the Malaysian transport minister’s denial.

* The report, by Pierre Daniel, 8 February 2016, obtained by China families, states «La présence de crustacés, du genre Lepas, des deux cotés du flaperon suggèrent une ligne de flottaison différente abc une pièce qui serait totalement immergée.»

“The presence of crustaceans, gender Lepas, on both sides of the flaperon suggest a different abc waterline a piece that would be completely submerged.”

260 Responses to “ATSB Denies Request from MH370 Families for More Info”

  1. Brock McEwen says:

    Well said, Victor. In my most recent email to Inmarsat (cc ATSB) seeking details on the uncontroversial log data purportedly used to derive their 4600 micro-second offset differential for arcs 1 and 7 – a request both groups deftly deflected – I put it this way: it is one thing to search for three years, not find anything, and suspend the search indefinitely. It is quite another to search for three years, not find anything, suspend the search indefinitely, and yet continue to hoard key data and analysis from others who haven’t given up.

    I appreciate your staunch support of the idea that all countries must now turn out their pockets. Perhaps it’s the Easter spirit of hope reborn coursing through my veins at the moment, but I feel hopeful enough about this nonpartisan posting of yours to admit that I’ve been a tad easy on the Malaysian government over the years. To at least partly remedy this: I hereby publicly underscore my support for Victor’s calls for more fulsome disclosure of Malaysian radar data.

    Speaking of which: does anyone know where the radars on Diglipur (North extremity of Andaman & Nicobar Islands) and Great Nicobar (South extremity) are located, specifically? I have good data on Port Blair and Car Nicobar (which, it turns out, had upgraded from Rohini prior to March 8, 2014), but really need the other two for completeness. Google Earth tells me there are some nice high mountains for the Diglipur base to utilize.

    Also: if folks could provide all they know about US radar coverage of the area that night (via ground, ship, air, or sat), I’d be much obliged.

    Also Cambodia (e.g. Koh Tang), Myanmar (e.g. Zadetkyi), Viet Nam (e.g. Tho Chau), and any other assets which had a chance to see something. We need to go back and carefully re-interview everyone re: what they had in position, and what those positions reported.

    Profuse thanks in advance.

  2. Oleksandr says:

    @Victor,

    In my understanding Mr Hood’s original mission was to terminate the search operation:

    “As one of his first duties, Mr Hood will have to make the difficult decision whether or not to continue the search for the missing aircraft.”

    http://www.theaustralian.com.au/business/aviation/new-atsb-chief-greg-hood-faces-tough-choice/news-story/e4d067d263387a9c5d499abbe19c21f5

    Nothing surprising about his recent decision. Typical bureaucrat.

  3. Andrew says:

    @Oleksandr:

    The decision to cancel the search was not made by Greg Hood. It was a tri-partite decision, made at ministerial level between the governments of Malaysia, China & Australia. Hood had no choice but to comply with the decision and associated directives from the Australian government.

  4. Oleksandr says:

    @Andrew,

    Indeed. Greg Hood serves the people, who appointed him. Obviously, if his agency is not paid, he cannot do anything. But he has a freedom to agree or disagree with such a decision, select a proper strategy for a given budget, propose a new path forward, or at least avoid threatening his staff. It is not his responsibility to worry or comment on the international relationships. I don’t know the Australian legislation, but it looks weird to me that he warns his staff not to release any information to a court.

  5. David says:

    @Victor. I note that the decisions by the ATSB and Commissioner were not about data but opinions.

    You ask, “So why would the information about MH370 requested by The Australian cause damage to international relations?” Well I do not think that those foreign governments, businesses and agencies would be too impressed if their manufacturers’ and agencies’ opinions were released this way since confidentiality is crucial to attaining frank and full information flow. Apart from anything else these opinions can evolve as you know, new information and thinking coming to light. Besides, during an investigation I do not think it would be the practice anywhere to disclose such opinions as those sought, even afterwards.

    Possibly there would be scope for the families of the Chinese victims to appeal this FOI decision but that has been overtaken now by the Commissioner’s edict, evidently on quite different grounds, the TSI Act.

    Presumably what the Commissioner had in mind is that act prevented himself and others covered by that legislation from such disclosure regardless of the FOI reason.

    Higgins conflates this with whether ATSB staff should be able to disclose their views as to there being an active pilot active at the end, apparently proscribed earlier by the Commissioner invoking the same act.

    I question the applicability of the act here: https://www.dropbox.com/s/v0ugt6wlcuzm9yw/TSI%20Act.docx?dl=0

    In summary, the Commissioner has rejected two appeals for information about professional opinions on the same grounds. Neither sought satellite data though unfortunately it could well be that an application for that would get a similar reception from him and the principal accident investigator unless the data owners were willing to release it. If not and the TSI Act were adjudged inapplicable, the FOI reason would be there still, though other grounds would need to be found to inhibit ATSB staff and others from disclosing their individual views.

  6. Victor Iannello says:

    @David: I agree that opinions and interim results should not be released for the reasons you mention. However, we also know that there is a lot of evidence not released that was used to derive results presented in reports. Off the top of my head, this includes satellite data from previous flights, radar data, Boeing simulator data, and Thales SATCOM functionality. Do we really think that releasing the technical details surrounding the 4600 μs offset at a log-on is going to cause an international incident?

    We also know that the Factual Information released by Malaysia in March 2015 is missing important evidence. For instance, the connection with the cell tower when MH370 flew by Penang Island is important for establishing the flight path after the turn back at IGARI. I believe the simulator data found on the captain’s computer is also relevant. We only know about these data sets because of unauthorized releases to French media.

    Considering the failure of the official investigation to find the plane, evidence not normally released should be. It is likely that this evidence will be used by journalists to write sensational stories that are misleading, and by families to pursue merit-less law suits. I think that is a fair price to pay for the possibility that we uncover more facts that will lead us to solving this mystery.

  7. ventus45 says:

    @Oleksandr

    Not meaning to steal your thunder Oleksandr, but there have been some amendments since the date of the document you linked to on the ATSB website (which is not entirely surprising).

    This website presents the most authoritative and up to date status of the act.
    https://www.legislation.gov.au/Series/C2004A01102
    Specifically, the latest compy (inclusive of amendments to the Statute Law Revision Act (No. 1) 2016 ( https://www.legislation.gov.au/Details/C2016A00004 )

    So far as I can determine, the most current TSI legislation reference (dated 09/Jun/2016) is here:-
    https://www.legislation.gov.au/Details/C2016C00617.
    It can be downloaded in two formats as follows:
    PDF: = https://www.legislation.gov.au/Details/C2016C00617/54523ae0-ec4f-4a04-b676-e20009e62e80
    DOC: = https://www.legislation.gov.au/Details/C2016C00617/22402e62-7467-4ab9-bb1a-88a7cbf89b77
    OR as a ZIP: = https://www.legislation.gov.au/Details/C2016C00617/39c4ed3b-f193-4207-b69d-cdcdff9c0b57

  8. Andrew says:

    @Oleksandr:

    “Greg Hood serves the people, who appointed him.”

    ”It is not his responsibility to worry or comment on the international relationships. I don’t know the Australian legislation, but it looks weird to me that he warns his staff not to release any information to a court.

    Well I guess that’s debatable – he’s actually appointed by the government and is bound by the Australian Transport Safety Investigation Act. That Act, amongst other things, makes it an offence, punishable by two years imprisonment, for ‘Commissioners, staff members and consultants’ to disclose restricted information to ‘any person or to a court’. The reasoning behind the limitation on disclosure of information to a court lies in the role of the ATSB, as defined under the Act, which specifically states that it is not the ATSB’s role to ‘apportion blame for transport safety matters’, or ‘to provide the means to determine the liability of any person in respect of a transport safety matter’, in accordance with ICAO Annex 13.

    Don’t get me wrong, I’d like to see the information released, but it seems to me that the Malaysian government, for whatever reason, has blocked the release of most information relating to MH370 by other countries. Greg Hood has simply invoked Sect 33 of the Australian Freedom of Information Act, which exempts documents from disclosure in cases where disclosure ‘would, or could reasonably be expected to, cause damage to…the international relations of the Commonwealth’. Like it or not, that judgement is part of his role as Chief Commissioner of the ATSB.

  9. ventus45 says:

    Can anybody seriously continue to support the contention that MH370 was “a bona-fide accident”, and not “a deliberate act”, thus being a “crime” ?

    A “bona-fide accident” should be investigated, “as an accident”, and thus the provisions of Annex 13 and the relevant enabling legislation in the various countrys involved are in play.

    BUT!!

    It is becoming increasingly “obvious” to all, with every possible “accident scenario” having been “done to death”, and come up “empty”, that MH370 “was not an accident”, but was indeed “a deliberate act”, thus, obviously, being a “crime”.

    Once the “transition” from “accident mindset” to “crime mindset” solidifies, would not the provisions of Annex 13 and the relevant enabling legislation in the various countrys involved become “no longer applicable” ?

  10. TBill says:

    @Ventus @Donald
    I agree with your sentiments. We should consider alternate possible causes of MH370 loss, but the fact that intentional diversion seems to fit the data well needs to be acknowledged, and then move forward.

    @David
    Re: Negative pressure landing
    It’s a long shot, but I am trying to suggest intent to land the aircraft under considerable negative pressure/vacuum to contain/implode the crash debris. The negative pressure relief valves would probably squash this theory, unless there is some way to disable or plug those up with. Sounds like the forward cargo bay is at least accessible in theory.

  11. Victor Iannello says:

    @ventus45: The ATSB has made it clear that the criminal investigation is not their responsibility. As such, the ATSB would not independently call the disappearance a crime. As far as I can tell, the only authority investigating possible criminal activity is Malaysia. We know that the FBI provided technical assistance to Malaysia to help recover the data on the captain’s computer. However, I don’t believe the FBI itself ever opened a case regarding the disappearance.

  12. Gysbreght says:

    @Victor Iannello: “As far as I can tell, the only authority investigating possible criminal activity is Malaysia. “

    France?

  13. Victor Iannello says:

    @Gysbreght: Yes, I failed to mention the French investigation, which has already concluded that there was nothing suspicious about any of the passengers or crew.

    As an aside, journalists in France seem to be gravitate toward the theory that the plane was downed in a US-led military operation, and there was a subsequent cover-up by the US and the UK with the help of Inmarsat. When the French investigation cleared the passengers and crew of any suspicion, this further fueled speculation about a military shoot down.

  14. Gysbreght says:

    @Victor Iannello:

    Thanks. I wasn’t aware that the French criminal investigation had reached a conclusion. Is it closed?

  15. Victor Iannello says:

    @Gysbreght: In January 2017, the AFP reported that they had cleared the passengers and crew of any criminal suspicion. As far as I know, the investigation is ongoing. If they aren’t pursuing passengers or crew on the plane, it is not clear what avenues they are pursuing.

  16. TBill says:

    @VictorI
    Re: France
    The France position is confusing to me. I know we relatively recently heard the “nothing suspicious about any of the passengers or crew” statement but I seem to recall at least one reference saying that may not be the comprehensive France position consensus.

  17. Oleksandr says:

    @Andrew,
    @ventus45,

    Either I am missing something, or the authors of the “Transport Safety Investigation Act 2003” smoked something when they compiled this document. What if a court orders ATSB staff to disclose some information? Or a court in Australia does not have such a right? Or ATSB reserves a right to deny court’s order?

    Well, I am only wondering where is the line separating non-disclosure of confidential information from legitimate requests of NOK to provide updates or additional information? This event has opened another can of worms. ATSB now looks very dirty.

  18. Victor Iannello says:

    @David: Since the Malaysian officials have nothing to hide, they can start by releasing all the radar data in their possession.

  19. Andrew says:

    @Oleksandr:

    The TSI Act seeks to protect the ATSB’s independence by making it unlawful for past or present ATSB employees or consultants to disclose information except in limited circumstances. An allowable defence is the disclosure to a court in civil proceedings where the court orders the disclosure, provided the ATSB has certified that such disclosure is not likely to interfere with an investigation. In making that order, the court must be ‘satisfied that any adverse domestic and international impact that the disclosure…might have on any current or future investigations is outweighed by the public interest in the administration of justice’.

    When all’s said and done, the Malaysian government has investigative responsibility for the disappearance of MH370, under Annex 13 to the Convention on International Civil Aviation. Australia is providing assistance under paragraphs 5.23 and 5.24 of Annex 13, but the Malaysian Ministry of Transport is responsible for the administration and release of any information about the accident investigation. I don’t think it’s fair to blame the Australian government for withholding information when it’s the Malaysians that control the release of that information.

  20. Gysbreght says:

    @David:

    I think it is fair to say that the ATSB has been in charge of the search effort, has selected the search area, is responsible for the choices made in that effort, and has an interest in hiding informations that put the wisdom of those choices into question.

  21. Gysbreght says:

    An example is information regarding Boeing simulations of “unresponsive crew” end-of-flight scenarios. The ATSB requested those simulations and specified the scenarios to be simulated. Why do they refuse to reveal those scenario’s and details of the results?

  22. Andrew says:

    The ATSB’s rebuttal to Ean Higgins’ story:

    Inaccurate media coverage on the search for MH370

  23. Oleksandr says:

    @Andrew,

    Re: “An allowable defence is the disclosure to a court in civil proceedings where the court orders the disclosure, provided the ATSB has certified that such disclosure is not likely to interfere with an investigation”

    That is exactly my point. Why does some bureaucratic dude appointed by the government has the right to stay above the law? What right does he have to withhold information on the basis of his own judgment? In my understanding in any democratic country it is up to a court to decide whether information can be released or not, after both sides present their arguments. It is understandable that ATSB can be bound by some international non-disclosure agreements, or disclosure of information can interfere with investigation. However, it must be up to a court, not up to a bureaucrat at ATSB to judge.

    In general, I have noticed that the ATSB is recently ignoring all the facts pointing outside of their search area:

    1. Drift studies: CSIRO, UWA, Deltares, IPRC, etc. show that ATSB’s selection is unlikely. Nevertheless ATSB concludes that the drift studies are consistent with their selection. Also, why does ATSB cite CSIRO’s study, which used considerably lower leeway factor than it was experimentally established?

    2. Debris. I don’t understand why neither ATSB nor Malaysians launch debris search campaign. All the fragments were found either by a chance, or by BG, who was then threatened for his activities.

    3. Forensic analysis of the found fragments. Where is it? Malaysians promised to included the flaperon analysis into the last year’s report. Australians were in possession of many fragments.

    4. “Curtin boom” and Haixun 37.5 kHz ping. Both are consistent with the original ATSB’s assessment. The former is “ruled out” based on a weird assumption and “inconsistent timing”. The later is ruled out because the hydrophone used by Chinese was certified up to 2 km depth, while the depth at the location was 4 km. Neither of these reasons is really convincing.

    5. The towelette found in Australia. What did forensic analysis reveal? Fingerprints? If some tourist left it on the beach, they should have been able to find fingerprints. Where is a report?

    6. Barnacle analysis. The Australian and French teams came to different conclusions. Where are reports?

    7. DSTG analysis. They have implicitly cut out investigation of magnetic hdg/trk trajectories by imposing the lower limit of 0.73M on the speed. Why? Neither DSTG nor ATSB did provide sounding justification. I have an impression that it was a way out to hide the second probability peak, inconsistent with ATSB’s area.

    In my opinion, it is not fair to blame Malaysians for withholding military-sensitive radar data, which, btw, could likely be supplied by other countries (e.g. Thailand and/or Indonesia), and not to blame ATSB for withholding other potentially useful information using a very ridiculous excuse in the legislation.

  24. David says:

    @Gysbreght. About Boeing simulations of “unresponsive crew” end-of-flight scenarios and the ATSB refusal to reveal those scenario’s and details of the results, I did pass on their reason, viz: “Given the commercial origins of the information, and the confidential nature, we do not have any additional details of those simulations for public release”.

    As to, “I think it is fair to say that the ATSB has been in charge of the search effort, has selected the search area, is responsible for the choices made in that effort, and has an interest in hiding informations that put the wisdom of those choices into question”, yes that would be a motive. That raises questions of integrity of the organisation and individuals and also as to whether “hiding” could be concealed with so many players, not just within the ATSB.

    However since this is a mature, monitored (by parliament) and carefully regulated organisation and there is no sign there was serious disagreement during the late 2016 Review I do not think you should be making accusations, unless you have more reason that disquiet they will not provide Boeing owned information.

    They have given their rationale explaining the basis for the shift to the new proposed search area I see no reason to doubt their integrity. They work has been recognised internationally.

    It would have been better had the Review had new faces and been more open but possibly confidentiality got in the way and also they though this would not be efficient. I do not know.

    So far as I am aware they have specified the search as would others. I think public expectations of being fully embraced in all this, including those of such forums as ours, are unrealistic, as frustrating as that is. Public involvement generally is seen worldwide to be contrary to the efficient conduct of investigations, with occasional exceptions. Did the French run full public disclosure during the AF447 investigation? They have not been very forthcoming about the flaperon have they? What of the Dutch with MH17? Did they make information public during the investigation other, like the French, with interim reports?

    The ATSB, as they point out, have provided numerous summaries of what they were doing and why. The principal investigator, as a search participant, has apparently no reservations about their decisions.

    I think we run the risk of making the ATSB the whipping boy for a worldwide system which is necessarily confidential. That said it would be nice to see the ICAO nations agree to provide some sort of review, assistance and monitoring of major accident investigations, that is either entailing major loss of life or likely to reveal an airworthiness related cause, materiel or procedural.

  25. David says:

    Third last para, last line, other THAN.

  26. Gysbreght says:

    @David:

    viz: “Given the commercial origins of the information, and the confidential nature, we do not have any additional details of those simulations for public release”.

    I took that as a convenient excuse. Did they ask for Boeing’s permission, and did Boeing refuse?

  27. Gysbreght says:

    @David:

    The ATSB states in their report “Final Investigation: MH370 – Search and debris examination update” of 2 November 2016:
    “Some of the simulated scenarios recorded descent rates that equalled or exceeded values derived from the final SATCOM transmission. Similarly, the increase in descent rates across an 8 second period (as per the two final BFO values) equalled or exceeded those derived from the SATCOM transmissions. Some simulated scenarios also recorded descent rates that were outside the aircraft’s certified flight envelope.”

    That is an incomplete statement, bordering on deliberate deception. They omit to say that this only occurred in an electrical configuration that, by their own statement in their email to you, is unlikely to have existed, and in the scenario for trajectory #3 for which they refuse to give details.

  28. Oleksandr says:

    In addition to my earlier post. My list above is indeed incomplete. More items:

    8. Boeing simulations.

    9. BTO and BFO data from other flights used by the ATSB and DSTG (a particular interest is the mysterious “geographic dependence”).

    10. Signal strength data, which was cut out from Inmarsat’s logs.

    11. Functionality of Jindalee on March 8, 2014 (perhaps sensitive information, but if one demands Malaysian radar data, why not Jindalee?)

    I am sure ATSB could disclose/clarify many more items…

  29. Gysbreght says:

    The statement also omits the delay between the time the 7th arc would have been passed, and the time those rates of descent were observed in the simulations.

  30. Oleksandr says:

    @Victor,

    You wrote: “Since the Malaysian officials have nothing to hide, they can start by releasing all the radar data in their possession.”

    How about this citation:

    ——

    “In May 2016, the JORN FAQ file / Fact Sheet was updated by the RAAF to address questions about Malaysia Airlines Flight 370. According to the update, “JORN was not operational at the time of the aircraft’s disappearance.” The update also stated that MH370 would have been unlikely to be detected by the system due to radar range, ionospheric conditions and a “lack of information on MH370’s possible flight path towards Australia.”

    However, in the immediate aftermath of the March 8, 2014 disappearance, information regarding JORN’s status was not released. This led to months of speculation. On March 18, 2014, sources cited by The Australian said that JORN was not tasked to look toward the Indian Ocean on the night of the disappearance of MH370 as there was no reason for it to be searching there at that time. On March 20, 2014, it was reported that Malaysian Defence Minister (also Acting Transport Minister) Hishammuddin Hussein requested the US to give any information from the Pine Gap base near Alice Springs, possibly alluding to JORN as well. On March 19, 2014, it was reported that an Australian Defence Department spokesman said it “won’t be providing comment” regarding specific information on tracking MH370 by JORN. However, several days prior to this, Australian Foreign Minister Julie Bishop told the Australian Parliament, “All our defence intelligence relating to Flight 370 has been, and will continue to be, passed on to Malaysian authorities…”

    In March 2015, before the July 2015 discovery of MH370 debris on Réunion Island, aviation technology expert Andre Milne requested information from JORN to prove or disprove that the aircraft ended up in the Indian Ocean, but he received no response from the Australian government in 2015. It was made public in May 2016 that JORN was not operational at the time of the disappearance.

    —–

    Source: https://en.wikipedia.org/wiki/Jindalee_Operational_Radar_Network

    Greg Hood was appointed as the chief commissioner of the ATSB starting from July 1, 2016. The tripatriate meeting took place on July 22, 2016. The information on Jindalee operation was updated in May 2016.

  31. Don Thompson says:

    @Oleksandr

    Any sources cited in that Wikipedia article might be valuable but citing Wikipedia itself cannot always be regarded as credible.

    Further, Milne is no aviation technology expert, his posts on matters aviation and MH370 are delusional.

  32. Don Thompson says:

    JORN: was it operational, was it not?

    OTH radar requires conducive atmospheric conditions, JORN relies on a network of ionosonde stations. If archived ionograms are available (Aus BOM, NOAA?) it may be possible to determine if conditions over the sIO on 7-8 March were conducive to JORN operating.

    Any takers for that research?

  33. Oleksandr says:

    @Don Thompson,

    It does not matter whether Milne is expert or not. Important points:

    (a) Australian Defence Department spokesman said it “won’t be providing comment” regarding specific information on tracking MH370 by JORN on March 19, 2014.
    (b) Australian Foreign Minister Julie Bishop told the Australian Parliament, “All our defence intelligence relating to Flight 370 has been, and will continue to be, passed on to Malaysian authorities…”
    (c) Hishammuddin Hussein requested the US to give any information from the Pine Gap base near Alice Springs, possibly alluding to JORN as well.
    (d) The timing when it was made public that JORN was not operational at the time of the disappearance coincides with the timing when Greg Hood was appointed as the chief commissioner, and when the tripatriate meeting took place.

    Do you think these citations from Wikipedia are inaccurate?

  34. Victor Iannello says:

    @Oleksandr: Authorities have said there was no JORN data related to MH370. I haven’t heard a credible reason to doubt this.

  35. Mick Gilbert says:

    @ventus45
    Can anybody seriously continue to support the contention that MH370 was “a bona-fide accident”, and not “a deliberate act”, thus being a “crime” ?

    Well, if you’re counting hands and taking names, then I most assuredly support the contention that you cannot yet rule out that MH370 was a bona-fide accident; an initial inflight emergency that seriously degraded flight system interfaces and that seriously injured/incapacitated the crew. Manifestly, there were a number of deliberate acts subsequent to the airplane reaching IGARI – an initial diversion at around 1720 UTC to Kota Bharu, a subsequent decision at around 1725 UTC to divert to Penang, the probable restoration of power to the Left AC Bus around 1825 UTC – but none of these necessarily constitute malicious leave alone criminal acts.

  36. Andrew says:

    @Oleksandr:

    Perhaps you should stop and consider the reason why information pertaining to accident investigations is protected in the first place. It is summarised in the introduction to Attachment E of Annex 13 to the Convention on International Civil Aviation, which states:

    “The protection of safety information from inappropriate use is essential to ensure its continued availability, since the use of safety information for other than safety-related purposes may inhibit the future availability of such information, with an adverse effect on safety.”

    Para 5.12 of Annex 13 states that such information:

    “…could be utilised inappropriately for subsequent disciplinary, civil, administrative and criminal proceedings. If such information is distributed, it may, in the future, no longer be openly disclosed to investigators. Lack of access to such information would impede the investigation process and seriously affect flight safety.”

    Para 5.12 of Annex 13 stipulates that the State conducting an investigation must not release certain information for purposes other than accident or incident investigation “unless the appropriate authority for the administration of justice in that State determines that [its] disclosure outweighs the adverse domestic and international impact such action may have on that or any future investigations”, known as the ‘balancing test’. The Australian legislation is more exacting in terms of protecting safety information, because it imposes an additional requirement for the Accident Investigation Authority (i.e. the ATSB) to certify that the disclosure is not likely to interfere with any investigation before the court can apply the balancing test.

    There is an ongoing debate about the protection of safety-related information and how it should best be accomplished. The following links provide some further reading, if you’re interested:

    ANNEX 13 TO THE CONVENTION ON INTERNATIONAL CIVIL AVIATION – ATTACHMENT E. LEGAL GUIDANCE FOR THE PROTECTION OF INFORMATION FROM SAFETY DATA COLLECTION AND PROCESSING SYSTEMS

    ICAO HLSC 2010-WP/7: ENHANCING SAFETY THROUGH THE PROTECTION OF CRITICAL SAFETY INFORMATION

    ICAO HLSC 2010-WP/30: INFORMATION PROTECTION FOR ACCIDENT INVESTIGATIONS

  37. Victor Iannello says:

    I have posted an update which includes a media release from the group called the MH370 China Families.

  38. Oleksandr says:

    @Andrew,

    The protection of safety information from inappropriate use is only a bureaucratic excuse in this case. It can be applied to the radar data or the ongoing criminal investigation, but I do not see why it should be applied, for example, to the BTO & BFO samples utilized by the DSTG, drift properties of the recovered fragments, barnacle and towelette forensics, Boeing simulations, and other information in possession of the ATSB, which has no implications on the aviation safety.

  39. TBill says:

    @David @all
    In my mind, part of the moral issue for USA/ATSB re: NoK is financial compensation for American and Australian passengers. If the USA/AUS administrations prefer to let Malaysia make the call on liability class (accident vs. intentional), but feel MY is making the wrong judgement, then the NoK are not getting support. Maybe they (USA/AUS) need to compensate for the amount of compensation lost.

    Also in my mind the ICAO compensation guidelines are probably too low, so that is one reason CEO’s are willing to accept a loss of plane incident without fretting too much about fixing the hardware/people systems. My prior company only wishes they had some compensation limit for an accident.

  40. Andrew says:

    @Oleksandr:

    You may see it as a ‘bureaucratic excuse’, but it’s also the law. I note the following:

    1. The Transport Safety Investigation Act 2003 (TSI Act) defines ‘restricted information’ as, inter alia, “records of the analysis of information or evidential material acquired in the course of an investigation (including opinions expressed by a person in that analysis)”.

    2. Section 12AD of the TSI Act states: “(2) The Chief Commissioner must ensure that the Chief Commissioner’s powers under this Act are exercised in a manner that is consistent with Australia’s obligations under international agreements (as in force from time to time) that are identified by the regulations for the purpose of this section.”

    3. One of the international agreements identified by the Transport Safety Investigation Regulations 2003 is Annex 13 to the Convention on International Civil Aviation.

    4. Australia is an accredited representative to Malaysia’s investigation of MH370 conducted under Annex 13.

    5. Para 5.26 of Annex 13 states: “Accredited representatives and their advisers…(b) shall not divulge information on the progress and the findings of the investigation without the express consent of the State conducting the investigation.”

    6. Claimants who are refused access to information under the Freedom of Information Act 1982 may seek a review of that decision by the Office of the Australian Information Commissioner and, if the decision is affirmed, may seek further review by the Administrative Appeals Tribunal.

  41. ALSM says:

    The Australian is full of Fake News, especially when it comes to ATSB and MH370. ATSB “correcting the record”….again.

    https://www.atsb.gov.au/newsroom/correcting-records/inaccurate-media-coverage-on-the-search-for-mh370/#.WPZiUAz1-FM.twitter

  42. Oleksandr says:

    @Andrew,

    A law is like a telegraphic pole: it is not possible to jump over it, but it is possible to find a way around. The ATSB had many such opportunities, but apparently Mr. Hood’s “mission” is to terminate the search. Not to hide something, but just get rid of this annoying problem. A task assigned from the top.

    Frankly I was shocked by the existence of such a law in Australia, which downplays the role of a court, and which opens a door to corruption. Confidentiality has nothing to do with this.

  43. Andrew says:

    @Oleksandr:

    I’m not going to buy into the conspiracy theories. If you bothered to do a modicum of research on the subject, you would find that FOI requests about MH370 made to the UK AAIB and the US NTSB have been refused for the same reasons as those made to the Australian ATSB; namely, the protection of sensitive safety information and the avoidance of damage to international relations. The FOI requests in those countries were refused by the bureaucrats, just as they were in Australia, because the law requires them to do so. As I already pointed out to you, there are avenues for the appeal of FOI decisions in Australia, as there are in other countries.

  44. Mick Gilbert says:

    @DennisW

    Dennis, that “ATSB secretly retracts its consensus claim on MH370 ‘death dive’” story, once again authored by Ean Higgins, in The Australian is yet another example of the appallingly low to non-existent level of professionalism displayed by The Australian’s aviation reporters. Once again, the ATSB issued a “Correcting the Record” statement with regard to that article as follows;

    “Ongoing false media reporting on search for MH370
    12 August 2016

    An article published in The Australian by Mr Ean Higgins on 12 August 2016 falsely accuses the ATSB of ‘secretly retracting’ information from a Joint Agency Coordination Centre operations update on 27 July.

    In recent weeks the ATSB has been very careful to accurately describe the Defence Science and Technology (DST) Group’s analysis of the accident aircraft’s rates of descent at the final satellite handshake (known as the 7th arc). This analysis concludes that the metadata associated with the final two satellite communications from the aircraft to the ground earth station indicates that the aircraft was in a high, and increasing, rate of descent.

    All the members of the Search Strategy Working Group have reviewed DST Group’s analysis and no objections to the analysis have been provided.”

    If the ATSB was guilty of anything on this particular matter it was poor internal processes for signing off on Operational Updates.

  45. Peter Norton says:

    Victor Iannello: « The group also continues to claim that the satellite data was altered to support the theory of a crash in the SIO, citing comments posted by Emil Enchev on a physics blog site. The comments have been removed, so we don’t know the basis for his claims. I have in the past asked the group to publicly release this evidence so that it could be properly reviewed. They refused my request, citing concerns about the safety of the crew and passengers of MH370, who they think could still be alive, and could harmed by their captors if Mr Enchev’s comments were released. »

    Emil Enchev deleted all his postings, but 6 of them are archived here:
    http://archive.is/yQmu3#selection-7099.0-7107.53 (and below)

    1 posting on 11 April 2014
    5 postings between 28 May – 30 May 2014

    Additionally, other people cite passages of his deleted comments.

    I looked at the content, but it seems inconclusive to me.
    And the majority of his postings are gone (unless they are saved somewhere …).

  46. DennisW says:

    @Mick

    Lame.

    The ATSB changed their wording for a very good reason – a lack of consensus. Why else would they change it. Their explanation completely ignores the issue of why a change was made.

  47. DennisW says:

    @Mick

    I am actually leaning towards Higgins despite my general disdain of the journalism surrounding this incident. Higgins no doubt has sources inside the ATSB that provided the information he published. Whether it was an internal memo or general employee meeting, I believe what he reported is accurate.

    The ATSB is very low in my regard for many many reasons.

  48. Andrew says:

    @DennisW:

    The Australian, 14 October 2016. Note the last paragraph:

    ATSB bulletin jumped gun on MH370 death dive ‘consensus’
    Ean Higgins

    The Australian Transport Safety Bureau issued a bulletin falsely claiming it had “consensus” from a team of international experts for its “death dive” theory that Malaysia Airlines Flight MH370 went down fast in a pilotless crash, before two overseas agencies had a chance to express a view.

    The ATSB made the claim as its chief commissioner Greg Hood joined the Malaysian government and Malaysia Airlines in a media campaign to hose down an alternative “rogue pilot” ­theory that Captain Zaharie Ahmad Shah hijacked his aircraft and flew it to the end outside the search area chosen by the ATSB.

    Internal ATSB documents obtained by The Australian show that while a senior investigator drew the incorrect “consensus” statement to the attention of colleagues only minutes after the bulletin was released, the organisation never issued a correction and instead secretly deleted the claim from its website the next day, after it had been widely ­reported internationally.

    The ATSB repeatedly refused to say why it had deleted the “consensus” claim, and falsely denied doing so in a subsequent post.

    Internal ATSB emails ­obtained under Freedom of Information statutes by The Australian reveal the truth behind the organisation’s media manoeuvres.

    The revelations come amid doubts expressed by independent experts about the reliability of the Inmarsat satellite data the ATSB uses for its rapid descent assumption, and claims the agency has, to avoid embarrassing Malaysia, steered away from the “rogue pilot” theory.

    MH370 vanished on a scheduled flight from Kuala Lumpur to Beijing on March 8, 2014, with 239 people on board. Its radar transponder was turned off and radio contact was broken.

    Radar and automatic satellite tracking data indicate the Boeing 777 reversed course early in the flight, and flew along the ­Malaysia-Thailand border and out over the Andaman Sea before making a sharp turn south to end up in the southern Indian Ocean.

    The ATSB designed the search zone based on its “ghost flight” theory that the pilots were incapacitated, and that after flying on autopilot, the aircraft came down quickly after running out of fuel.

    In July, two developments led to international debate about the ATSB’s strategy.

    Reuters reported the project director of the underwater search, Paul Kennedy of Dutch survey group Fugro, said the rogue pilot theory might be right after all.

    “You could glide it for further than our search area is, so I believe the logical conclusion will be, well, maybe, that is the other scenario,” Mr Kennedy told Reuters.

    The same weekend, New York magazine revealed a Malaysian police report indicated the FBI had determined Zaharie had charted a similar route on his home flight simulator.

    Days later, the ATSB issued a bulletin in the name of the federal government’s Joint Agency Co-ordination Centre for the MH370 search, downplaying the rogue pilot theory. The ATSB claimed in its July 27 bulletin the satellite data showed MH370 came down “most likely in a high rate of descent”. As originally released, the bulletin said: “This is indeed the consensus of the Search Strategy Working Group,” referring to experts including from the US and British air crash investigation.

    The documents obtained under FOI show that just a few minutes after the bulletin was ­issued, an ATSB senior investi­gator warned colleagues by email this was an “error” and that the sentence should be taken down.

    “It is certainly not yet the consensus position of the SSWG … 2 parties are yet to make a formal response on the subject,” the investigator said.

    The email chain shows another ATSB senior investigator agreed and gave instructions for the sentence with the “consensus” line to be removed from the ATSB’s and the JACC’s websites.

    But the ATSB did not retract the sentence until the next day, by which time it had been reported internationally, including in Mal­aysian and Chinese publications.

    As earlier revealed by The Australian, the deletion of the “consensus” line was discovered by British aerospace engineer Richard Godfrey, a member of the independent group of aviation experts who on their own initi­ative have been reviewing the MH370 scientific data.

    When, at the time, The Australian rang the ATSB spokesman who had issued the July 27 bulletin to ask why the deletion had been made, the spokesman hung up and JACC director Annette Clark declined to respond.

    Subsequently, ATSB MH370 spokesman Daniel O’Malley and JACC chief co-ordinator Judith Zielke would also not say why the “consensus” line had been secretly disappeared.

    When The Australian reported the deletion of the sentence, the ATSB issued a denial on its website, saying the report “falsely ­accuses the ATSB of ‘secretly retracting’ information”.

    In a statement after it had been made aware of the FOI material, the JACC said: “The information was retracted when it was learned not all working group members had, at that stage, provided formal responses. Subsequently a consensus view was reached.”

    ATSB bulletin jumped gun on MH370 death dive ‘consensus’

  49. DennisW says:

    @Andrew

    begin cut-paste//

    As earlier revealed by The Australian, the deletion of the “consensus” line was discovered by British aerospace engineer Richard Godfrey, a member of the independent group of aviation experts who on their own initi­ative have been reviewing the MH370 scientific data.

    When, at the time, The Australian rang the ATSB spokesman who had issued the July 27 bulletin to ask why the deletion had been made, the spokesman hung up and JACC director Annette Clark declined to respond.

    Subsequently, ATSB MH370 spokesman Daniel O’Malley and JACC chief co-ordinator Judith Zielke would also not say why the “consensus” line had been secretly disappeared.

    When The Australian reported the deletion of the sentence, the ATSB issued a denial on its website, saying the report “falsely ­accuses the ATSB of ‘secretly retracting’ information”.

    end cut-paste//

    Pretty much says it all. The ATSB is not to be respected or taken seriously. Godfrey is truly superb.

  50. Donald says:

    @All

    In it’s latest rebuttal, the ATSB states the following ” The ATSB is not aware of any officers who have concerns with the ATSB’s reported findings in this regard.”

    This is either an outright falsehood (which it is), or unfathomable stupidity. Take your pick. The subject at hand is EOF scenario and possible active pilot (ie…Zaharie).

    Basically, the ATSB is full of it. It’s simply not believable nor conceivable that some officers do not KNOWINGLY have different opinions. The evidence to support no alive pilot is not even flimsy. It’s non-existent. I’m a psychiatrist and find it almost impossible to believe that Zaharie would have taken his own life just after skirting INDO airspace. You simply don’t meticulously (which it was) plan an elaborate murder/suicide and not see it through until the bittersweet end…which would entail a specific destination point. Not some vague and non-specific SIO terminus.

  51. DennisW says:

    @Donald

    Yes, the ATSB is obviously lying, and they are not very good at it. Analytical incompetence is excusable. It is what it is. Lying is inexcusable.

  52. Brock McEwen says:

    Re: military radar:

    To “prime the pump”, I will get the ball rolling with two US assets: USS Kidd and USS Pinckney. The internets tell me…

    1) both are Arleigh Burke-class destroyers

    2) Arleigh Burkes have Lockheed Martin SPY-1D phased radar

    3) SPY-1D radars have a range of 250 nmi

    4) both “were conducting training and maritime security operations in the South China Sea when called upon to assist in the search for Malaysian Airlines flight MH 370 in the Indian Ocean.”

    Have I got any of this wrong? If so, please advise.

    Regardless: can someone please link me to data on where precisely in the SCS they were located on the night of March 7, 2014, and whether they, too, turn their radars off at night.

    Profuse thanks in advance for any help you can provide.

  53. David says:

    @DennisW.”As earlier revealed by The Australian, the deletion of the “consensus” line was discovered by British aerospace engineer Richard Godfrey”.

    No. I drew attention to it before that on the JW blog, to which you contributed at the time. I raised it in passing as a trivial issue.

    Subsequently it was used and dramatised in The Australian. It is a mystery to me why they employ Ean Higgins and allow his stuff, mostly tripe as this was, into print. (When I posted this most recent on this site, I did add ‘FWIW’). Odd because the paper does have fine journalists; but he lets them down.

  54. David says:

    @Gysbreght. You quote the ATSB, “Some of the simulated scenarios recorded descent rates that equalled or exceeded values derived from the final SATCOM transmission. Similarly, the increase in descent rates across an 8 second period (as per the two final BFO values) equalled or exceeded those derived from the SATCOM transmissions.”

    You say that this only occurred in an electrical configuration which they say is unlikely.

    Would you amplify why you think that there were no other configurations with such descent rates?

    Also I note that while they say that highly unstable descent rates resulted in that scenario there is no explicit statement that these met the BFO criteria: it might be that the instance you have in mind did not in fact qualify.

    I think you see these simulations as proving or disproving their BFO thesis, whereas I think their approach is from the other direction. They conducted the simulations having become satisfied that their BFO interpretation was sound, to determine which configurations might qualify. In the new simulations, “reasonable values were selected for the aircraft’s speed, fuel, electrical configuration and altitude, along with the turbulence level”*, so more than just the one, electrical. In other words, the descent rates were a simulation qualifier in the way the initial CSIRO drift analysis was, in asserting that the flaperon drift and its timing were consistent with the purported 7th arc crash area.
    The intent then was, from those simulations which met the BFO descent criteria, to find their final distances from the 7th arc, to set the width of the new search area.

    Before getting to your second point there are some I would like to raise within this general topic.

    • In the ATSB comments to me you have seen, the AC power loss at right-engine-failure hypothesis, “was put forward in an attempt to see if this scenario could lead to the aircraft travelling significantly further from the 7th arc than in the standard electrical configuration. According to the simulations, however, this configuration led to highly unstable flight paths that ended closer to where the 7th would have been for that scenario”.

    As to the last phrase, I do not see how the 7th arc could be shifted, essentially being set by BTO, nominally 2 mins after AC power failure. What I take it they mean is that the aircraft crash site for this scenario could be much more distant from the 7th arc. If it did meet descent rate criteria, being possible (even though unlikely) that scenario would reduce new search area success probability.

    • They say, “…flight simulators are unable to accurately model the dynamics of the aircraft’s fuel tanks”.* They make apparent that therefore the Boeing simulator did not simulate APU start-up which, “would re-energise the AC buses and some hydraulic systems. This could affect the trajectory of the aircraft. Similarly, the left and right engines may also briefly restart, affecting the trajectory.”*
    Surely this would affect confidence in the new search area’s width?

    • I note they say also that, “The engineering simulator uses the same aerodynamic model as a Level D simulator used by the airlines”*, yet in their comments to me (you have); “The Working Group’s simulations were performed in the Boeing Engineering Simulator which has a significant fidelity to the actual systems on the aircraft, and which was therefore considered more accurate than many other available simulators”. While supposing the two to be consistent shows why the Boeing results are different, yet they do not simulate an APU start up as I understand it while the Exner level D simulation did. An explanation would help.

    Finally, about your comment that our common reference, “omits the delay between the time the 7th arc would have been passed, and the time those rates of descent were observed in the simulations”, I can see your point. I suppose that since APU start-up was not simulated, they measured descent rates at 1:59 after engine failure and again at 2:07, then ascertained whether at those times the descent rates at least matched the BFO; and that there was a 10,800fpm descent rate gain between. For my part, adding to the importance of this timing assumption, I maintain still that they have not allowed 30-40 seconds for APU inlet opening. That could make a great difference in that dynamic situation.

    About some of these issues, it seems we will need to await the final report.

    *ATSB’s MH370 Search and debris examination update, 2nd Nov, 2016, p13/14

  55. ROB says:

    @Donald

    You are totally 100% right. A meticulously planned murder suicide which this obviously was, would require as much planning of the EOF situation as would have been required for the initial takeover, if it was going to succeed in its aim, and it obviously did go successfully – the ATSB couldn’t find the plane (even with the helpful and remarkable input from INMARSAT.

    Of course ATSB are lying (no credit to them, sadly) but lies have a habit of eventually being found out. They were handed a political poisoned chalice, when they were landed with this job. Both the Malaysians and the Chinese realized from the get-go that this was a political act, aimed at the Malaysian Establishment and the Chinese brutal suppression of muslim separatists in the far west of their counry – making a Malaysian airliner full of unwitting Chinese citizens disappear into thin air, would be taking a swipe at both the Malays and the Chinese.

  56. Oleksandr says:

    @Andrew,

    Re: “I’m not going to buy into the conspiracy theories”

    Neither I. If you meant the plane itself, my primary domain is the mechanical failure, followed by slow decompression. It seems there is a great vulnerability in the design of B777.

    If you mean the reluctance of the ATSB, Malaysians, French and Chinese and others to take actions, and also their unwillingness to release information, then, in my opinion, this is because of the blooming bureaucracy and desire of governments to stop spending funds for the investigation as the cost of the search operation became comparable with the price of a new aircraft. Release of new information could promote review of the assumptions and stimulate new studies, which could be not in favor of ATSB’s priority search zone.

  57. Mick Gilbert says:

    @DennisW

    Lame? Come on, now! It was a simple and very straightforward rebuttal.

    Re “Godfrey is truly superb.”; ask Richard for his opinion of the journalistic standards of The Australian in general and Ean in particular.

  58. Don Thompson says:

    The Australian and Ean Higgins destroyed any credibility, concerning MH370, when they engaged former airline pilot Byron Bailey to opine on the matter. His ‘knowledge’ of the end of flight scenario has tainted the Australian’s commentary, the paper can’t counter their pet pilot pundit and the ATSB at the same time.

    I do fail to understand why folks take so much press reporting as irrefutable fact.

  59. Gysbreght says:

    @David:

    “You say that this only occurred in an electrical configuration which they say is unlikely. “
    I said: “… in an electrical configuration …, and in the scenario for trajectory #3 …”

    “Would you amplify why you think that there were no other configurations with such descent rates? “
    When I asked in which trajectories this occurred, the ATSB replied 3, 7, 8, 9, 10. Trajectories 7, 8 , 10 represent the ‘unlikely’ electrical configuration. In simulations 3 and 7 the aircraft’s motion was outside the simulation database. The manufacturer advised that data beyond this time should be treated with caution. (I missed #7 in my post. Bad memory.) That caution is not exercised in the ATSB report.

    I don’t understand the sentence starting with “Also I note …”.

    “I think you see these simulations as proving or disproving their BFO thesis, …”
    They present these simulations as proof of their BFO thesis. That presentation is misleading, it ought to have been qualified. On re-reading the report I note that the ATSB also wrote:
    “Simulations that experienced a descent rate consistent with the ranges and timing from the BFO analysis generally impacted the water within 15 NM of the arc.”
    Based on the shape of the trajectories I suspect that statement is incorrect with respect to the timing relative to the 7th arc. I have asked the ATSB for those time delays but they still owe me a reply.

    “As to the last phrase, …” I don’t think the phrase suggests shifting the position of the 7th arc. The scenario did meet the BFO values but (I suspect) not within 8 seconds from the 7th arc (time of log-on request message, two minutes after loss of electrical power and autopilot).

    The report remark about dynamics of the aircraft’s fuel tanks leaves the reader in the dark as to whether or not the APU (was) started in the simulations. I don’t believe those dynamics could result in brief engine restarts, affecting the trajectory (see ‘lorry’).

    “About some of these issues, it seems we will need to await the final report.” Do you expect a final report from the ATSB, or do you mean the Malaysian final report targeted for the end of this year?

  60. Mick Gilbert says:

    @Don Thompson

    You wrote, “The Australian and Ean Higgins destroyed any credibility, concerning MH370, when they engaged former airline pilot Byron Bailey to opine on the matter.”

    Thank you, Don, on that matter we are most assuredly on the same page.

  61. Mick Gilbert says:

    @Victor
    @All

    Just to return briefly to recent discussions regarding fuel flows, would I be correct in inferring from the various discussions that there is an emerging collective view that in order for fuel exhaustion to occur as late as 0017 UTC the final leg south into the SIO would need to have been flown at a speed somewhat slower than LRC or ECON cruise, something closer to Best Holding Speed?

  62. Oleksandr says:

    @David,

    In response to Gysbreght you wrote:

    “I think you see these simulations as proving or disproving their BFO thesis, whereas I think their approach is from the other direction. They conducted the simulations having become satisfied that their BFO interpretation was sound, to determine which configurations might qualify.”

    Long time ago I presented a paper explaining the two abnormal BFOs of 273Hz (18:25:34) and -2 Hz (00:19:37) as a result of zero Doppler compensation term. This could also be a result of an erroneous compensation term due to erroneous navigation data fed into AES. There are two outstanding features:

    1. The two logon sequences are identical.
    2. Both the abnormal BFOs are paired with abnormally longs BTOs.

    Thus, it is almost certain that the underlying reason is the same in both the instances. Hence, if the BFO of -2 Hz is considered as representative of a rapid descent, then the BFO of 273 Hz should also be considered as representative of a rapid ascent. Yet the ATSB prefers using the BFO of -2 Hz, but ignoring the BFO of 273 Hz.

  63. Gysbreght says:

    From the same report:

    “Therefore the contributing elements consist of the:
    • tolerance or error of the BFO
    • direction of travel of the aircraft
    • oven-controlled oscillator warm-up drift
    • descent rate of the aircraft.”

    The first three bullets don’t explain the the BFO of -2 Hz at 00:19:37, therefore it must be the 4th.

    Simples, isn’t it?

  64. Gysbreght says:

    Simples. Except that there is no “unresponsive crew” end-of-flight scenario that results in rate of descent increasing by 10,000 fpm in 8 seconds, 2 minutes after loss of power.

  65. Victor Iannello says:

    @Mick Gilbert: Based on what we (think we) know about the temperature effect on fuel flow and the average PDAs for the engines, I’d say it’s doubtful that from 18:28 and later, the plane remained near LRC speed. However, there are many possibilities, including a short period of time at a slower-than-cruise speed followed by a longer period at cruise speed.

  66. TBill says:

    Question:
    ATSB: most consistent with “no human intervention during the final phase of the flight.”

    What is the final phase? After FMT (which I would not be so sure) or the very last part of the flight crash into the sea?

  67. DennisW says:

    @Mick

    “Lame? Come on, now! It was a simple and very straightforward rebuttal.”

    I read it a few times, and still can’t see how it addresses why they changed the consensus view.

    As far as your view of Bailey is concerned, I don’t see how it reflects poorly on the “Australian”. A newspaper does not create news. It reports news. Giving space to Bailey is what it is. A newspaper cannot act as a filter of opinion. When the New York Times reports what someone says it does not mean the NYT endorses it.

    One thing we know for certain relative to the ATSB is that they tossed 150M+ USD in the toilet by initiating an ill advised underwater search.

  68. Ge Rijn says:

    A telling article at last again.
    This makes finally clear what those NoK and all independent investigators are up against.
    Threatening your own ‘independent’ investigators with 2 years prison if they speak up with another opinion or release asked-for information without (Australian Government) permission.
    With the motivition; ‘the ­information could cause damage to the international relations of the commonwealth’.

    With this intervention the ATSB is officially reduced to a puppet-organisation proving it was totally powerless to conduct any reliable (aircrash) investigation on MH370 in the past, now and in the future.

    Proving also, what many suspected for a long time, there is a great deal to hide from the public. A really great deal.
    To find it necessary to proclaim this very intimidating signal to their ATSB investigators (and let it proclaim themselfs) and the very harsh message to the NoK that; ‘damage to the international relations of the commonwealth’ are regarded more important than their sorrow and struggle to find answers.

    IMO the ATSB has now really shown their cards. They know a lot more and they have been seriously shut down by the Australian Government.
    Maybe that’s even the implicit message they wanted to give to the world.

    We can only hope someone there will be brave enough to come forward and leak information.
    But I’m affraid it won’t happen. By their heritage and trade Aussies generally don’t belong to the most enlightend people on this earth. Pragmatic and opportunistic as they mostly are.

    I hope a brave exeption will make the difference.

  69. Victor Iannello says:

    @Peter Norton: Welcome, and thank you for the link. From what I can determine based on the surviving comments of Emil Enchev and the comments from others, Mr Enchev doubts the validity of the BTO data because processing delays are indeterminate. Well, that’s not true–the internal delay in the AES is limited to 50 μs by ICAO specification, and the standard deviation of the measured BTO values was determined to be around 29 μs. The group MH370 Chinese Families seems to be more interested in promoting conflicting theories than presenting facts.

    @Rob: Welcome to the discussion.

  70. Andrew says:

    “We can only hope someone there will be brave enough to come forward and leak information.
    But I’m affraid it won’t happen. By their heritage and trade Aussies generally don’t belong to the most enlightend people on this earth. Pragmatic and opportunistic as they mostly are.”

    This discussion started with ATSB bashing and has now moved on to bashing Australians in general. You are entitled to your opinions, but comments like that above are offensive and, in my opinion, have no place in this debate.

  71. DennisW says:

    @Andrew

    “This discussion started with ATSB bashing and has now moved on to bashing Australians in general. You are entitled to your opinions, but comments like that above are offensive and, in my opinion, have no place in this debate.”

    Totally agree.

    As far as “ATSB bashing” is concerned, I think you are over reacting. The reality is that none of us has any reason to hold our heads high. The ATSB, the IG, and the assorted individual contributors all acted in good faith. Unfortunately good faith and a dollar will get you a cup of coffee.

  72. Ge Rijn says:

    @Andrew

    I know what you mean. I deliberately throw some pepper to the (ATSB) Aussies. Please see my underlying objective.
    By the way being generally pragmatic and opportunistic is not meant as an offence to Australians. It’s just my experience being in the country and with the people there for a considerable time. I can understand why ‘in general’ those qualities are more important there than in Holland.

    I’m aware I was absolutely generalising.
    But with the goal of talking to ‘exceptions’ within the ATSB or the Australian Government, or anyone else to speak up and not settle for intimidation.

  73. Victor Iannello says:

    Whatever anybody’s feelings about the ATSB, and mistakes it might have made during the investigation, I believe the ATSB is the only member of the official investigative team that has been sincerely interested in finding the plane.

  74. Gysbreght says:

    @Victor Iannello:

    I wouldn’t say “the only member”, but otherwise: Well said!

  75. DennisW says:

    @VictorI

    “Whatever anybody’s feelings about the ATSB, and mistakes it might have made during the investigation, I believe the ATSB is the only member of the official investigative team that has been sincerely interested in finding the plane.”

    I tend to agree, but having spent many months in aggregate with Pac Rim customers, I was never able to consistently tune into their mindset, Aussies being the exception in that regard. I think decent and pragmatic US people and Australians are pretty much aligned.

  76. Ge Rijn says:

    @VictorI

    Yes, I’m surely willing to believe their intentions were objective and sincere in the beginning.
    But your article now shows they’ve been mouth-shut completely by intimidation.
    Something like this does not happen overnight.
    What many of us already sensed long time ago; they must have been struggeling with covering information for a long time.
    Crucial information.
    Probely restricted by their Government superiors.

    I therefore call upon the ATSB not to submit to threats of your Government on imprisonment while asked doing your job right.
    If you, any of you, have any honor left than speak up.

  77. DrB says:

    @Mick,
    @Victor,

    Mick said: “Just to return briefly to recent discussions regarding fuel flows, would I be correct in inferring from the various discussions that there is an emerging collective view that in order for fuel exhaustion to occur as late as 0017 UTC the final leg south into the SIO would need to have been flown at a speed somewhat slower than LRC or ECON cruise, something closer to Best Holding Speed?”

    I agree with Victor that it appears impossible to achieve 00:17 flame-out with any (combination of) ECON mode after 18:28, even assuming the rather good PDA used in the Flight Brief. Some of the time after then must have been flown with lower Fuel Flows. I slightly take issue with Victor’s statement about “many possibilities”. In my opinion there are probably only three or fewer. If all that time were flown at Best Holding, the endurance is certainly adequate but would be too long to be explained by a single speed change to Holding (even as late as 19:41). That leaves two other possibilities. One case is a reduction to Holding for a period up to about an hour, followed by ECON mode (perhaps even at LRC). That is consistent with Victor’s theory of a loiter. The last case that might fit all the satellite data is a constant CAS (maybe 250 KIAS?) at a reduced altitude. That case is a driver for me to refine my fuel model so it accurately represents those speeds between Holding and LRC. When this work is completed, I will present some endurance curves for all these flight scenarios. Then we will have a feel for what specific speed modes are consistent with the observed endurance.

  78. ROB says:

    @TBill

    I think the ATSB interpreted “no human intervention during the final phase of the flight” from the ISAT transmissions at 00:19, ie the plane appears to have dropped out of the sky after running out of fuel. A possibly contentious conclusion, but a convenient one. They have however, never specifically come out and said at which point in the flight they believed the flight changed from piloted to unpiloted, also convenient.

  79. Donald says:

    @Rob

    >They have however, never specifically come out and said at which point in the flight they believed the flight changed from piloted to unpiloted, also convenient.

    There are two scenarios in which this could have occurred. One planned, and one unplanned. It’s my opinion that both scenarios can be all but ruled out. Assuming a deliberate diversion (it couldn’t be more obvious and clearcut), the co-pilot almost certainly would be locked out of the cockpit. And with equal certainty, he would not have been able to regain access. This effectively eliminates a ‘thwarted’ plan (unless we go down the absurd on multiple levels rabbit hole of the ee bay hijinks). Then we have the equally unlikely scenario in which Z takes his own life sometime after the FMT but before the EOF.

    Lets remember that Z had planned this for a significant period of time (btw, I’d love to know what the conversation he had with the MAS engineer was about). Without boring everyone with some psycho babble, he would NEED to visualize his and the airplanes final resting place, for myriad reasons. He would NEED to feel the satisfaction and gratification accompanying his successful accomplishment. This could only be achieved through an absolute certainty of success, which he COULD NOT embrace until near fuel exhaustion and/or a target destination. I’m not saying he ditched the airplane (though I believe he controlled the airplane DURING EOF and saw it down in the manner he desired), and it’s possible he depressurized (or however else he may have taken his own life) just before EOF. But he most assuredly would have been alive and well for the vast majority of the ‘final leg’ south into the SIO.

    Maybe the ATSB has a few incompetent psychologists in their ranks…just maybe.

  80. Mick Gilbert says:

    @Victor
    @DrB

    Thank you for those summaries, gentlemen.

  81. Mick Gilbert says:

    @DennisW

    My views on the journalistic standards at The Australian are most assuredly not based on one story or one contributor but rather on a pattern of stilted, misinformed, incorrect, deceptive and inflammatory reporting on all range of matters.

    It is worth noting that in 2015 and early 2016, News Corp regularly ran Captain Bailey’s articles as news stories, not opinion pieces, in both The Australian and The Daily Telegraph. Despite those stories containing any number of factual errors, such as the B777 carrying three transponders that have automated fail-over capability (yes, he flew B777s but that’s what he wrote!), Captain Bailey’s work continued to be published as “news” up to late January 2016. Then came the good Captain’s “MH370: report’s ‘stupid’ flaws hinder search” stoty published in The Australian on 29 January 2016. That news article contained the following:

    “First, the ATSB [Australian Transport Safety Bureau] states “the right engine flamed out and in each test case the aircraft then turned left and remained in a banked turn”.
    That’s strange because, as any experienced multi-­engine pilot knows, if the right engine stops, the aircraft will want to turn right because of simple moment of forces. Strange also because when I flamed out an engine in the FFS, the thrust asymmetry compensation via the autopilot kept the aircraft flying straight.”

    Curieux, non?

    Now, the ATSB never said “the right engine flamed out and in each test case the aircraft then turned left and remained in a banked turn.” That quotation cannot be found in any ATSB report or update. When queried (repeatedly) as to the source of that quote, Captain Bailey replied as follows :
    “God this is boring
    page 11 00:05:00 Right engine flamed out – page 13 aircraft behaviour after engine flameout…each test case left turn. Learn to put 2 and 2 together. You obviously are not a pilot …”

    Ignoring the petty invective, the pages the Captain refers to are from the ATSB report “MH370 – Definition of Underwater Search Areas” dated 3 December 2015. On page 11, in a section of the report entitled “End-of-flight”, there is a colour-coded timeline diagram headed “Figure 6: End of flight sequence.” The first point in the timeline is named “Right engine flamed-out”. On page 13 is the “Search area width” section of the report. The second sub-section is entitled “Simulator data”; the second sentence reads “In each test case, the aircraft began turning to the left and remained in a banked turn.”

    Captain Bailey had fabricated the statement he attributed to the ATSB by:
    – adding the definitive article “The”;
    – inserting the phrase “Right engine flamed-out” (taken from a diagram on page 11);
    – adding the conjunction “and”;
    – inserting the sentence “In each test case, the aircraft began turning to the left and remained in a banked turn.” (taken from page 13 by skipping over some 500 intervening words); and
    – changing initial capitalisations to give the appearance of a coherent, complete and contiguous sentence.

    The matter was referred to both the Editor and the Australian Press Council, The Australian was forced to issue a correction and Captain Bailey’s work was thenceforth relegated to the Opinion pages.

    Other poor performances on Aviation matters include Mitchell Bingemann writing almost hysterically about Jetstar’s B787 Dreamliners flying with a known ­engine problem that could trigger a mid-flight engine shutdown. In two stories over four days Mitchell managed to get the model of B787 that Jetstar flies wrong (it’s a detail that is widely and publicly available). He also missed or ignored the statistics showing that the engines were in fact more than three times more reliable than the very stringent certification requirements called for but the pièce de résistance was his repeated use of the non-word “incidences” – “An investigation by The Australian found 22 incidences of in-flight engine failures …; The incidences have affected … ” As I expect you know, “incidence” is a mass noun (also known as an uncountable or non-count noun). Mass nouns, such as knowledge, furniture, advice and evidence, do not have a plural form. The word that he should have been using was “incidents”, the plural form of the word incident. Spell check should have picked up “incidences”.

    Last week in The Australian we were treated to Paul Cleary’s “Why Qantas 747 stalled en route to Hong Kong” which opened with “The Qantas 747 that suffered a “stick shaker” stall was most likely hit by turbulence that ruptured the delicate balance of air pressure that allow aeroplanes to fly.” Apart from the fact that it’s doubtful that QF29 actually stalled, there is no such thing as a “stick shaker” stall, there were no other reports of turbulence and given the traffic near QF29 at the time of the incident wake turbulence is unlikely, it is an imbalance of air pressure that allows airplanes to fly and, given the range of conditions airplanes safely fly in, the generation of lift is hardly a delicate affair.

    And today The Australian is carrying the story “Learjet flipped ‘in Airbus wake turbulence’” that somewhat belatedly reports the 7 January 2017 inflight upset of a Bombardier (Canadair) CL-600-2B16 Challenger 604 that encountered wake turbulence from an A380. Even though Bill Lear was involved with Bombardier’s early design work on the Challenger back in the early 70s nobody refers to a Challenger as a LearJet.

    To the extent that I haven’t already, I could bore you with further examples of poor general journalistic standards at The Australian but I’ll restrict myself to one involving sometimes Aviation reporter, Ean Higgins, on non-aviation duties. As part of the coverage of the Lindt Cafe siege inquest, Ean produced this gem:
    “Although she [Chief Superintendent Kerrin Smith, from Britain] was not asked whether she thought police should have stormed the cafe earlier, Smith’s body language suggested she thought they should have.”

    Now, consider that for just a moment. Reporting an answer that was never given to a question that was never asked would be pretty close to a fabrication at the best of times. However, in this case Chief Superintendent Smith gave her evidence by video-link from the UK. That did not stop Ean from managing to somehow interpret the body language visible in a tight head-and-shoulders shot on a court room TV to adduce the Chief Super’s answer to a question she was never asked! I mean, seriously?

  82. DennisW says:

    @Mick

    Thx. Nice summary.

    Our local papers (SF Bay Area) are not a whole lot better. 🙂

  83. David says:

    @Oleksandr. You might be familiar with the report Gysbreght referred to.

    https://www.atsb.gov.au/media/5771939/ae-2014-054_mh370-search-and-debris-update_2nov-2016_v2.pdf

    His extract if from its page 8 and an explanation follows.
    However that does not address your 18:25:34 273Hz, though that was explained as I remember it by Dr Bobby Ulich a couple of weeks ago. If neither of those answers what you raise I can help no further.

  84. Oleksandr says:

    @David,

    You completely missed my points. We have two sequences of (BTO_n, BFO_n). These sequences are associated with the identical exchange protocols: see details of the first 7 or 8 signals after reboots in Inmarsat’s logs. Everything looks normal, except for n=2 in both the sequences. And not only BFOs are corrupted, but also paired BTOs. And these corrupted values are the only corrupted values in the whole record. If I recall correctly, estimated probability of this to happen by a chance is less than 1 in 3000. Just put these two sequences side by side and you will see.

    I suggested zero Doppler compensation hypothesis due to a simple reason: upon power reboot on the ground, ADIRU usually enters alignment procedure, during which the aircraft must not move. And it may take up to 15 minutes. FCOM contains a note that SDU may not be reliable during this time, but no further explanation is provided. If an aircraft does not move, Doppler compensation is zero (assuming that AES assumes that the satellite is geostationary). I don’t know what ADIRU output is during alignment, but the most reasonable assumption is that either ADIRU’s outputs buffers to ARINC are zeros, or AES knows that ADIRU is being realigned and assumes that no compensation is needed because aircraft is supposed to be on the ground. Many questions to AES and ADIRU designers.

    This zero compensation assumption gives surprising result. Firstly there is a solution corresponding to a flight-level cruise-speed at approximately 340 deg heading (see diagrams in my note) at 18:25:34, which might be consistent with the lateral offset suggested by Victor, and a number of emergency maneuvers suggested by myself. Secondly, the descent rate could be of order 20 to 30 m/s at 00:19 (under the assumption that velocity does not change over the interval of 8 seconds), which is a way more reasonable figure, than the one considered by the ATSB.

    Finally, I am not sure to what Bobby’s explanation you refer. You need to remember that the BFO was 142 Hz just 7 seconds earlier, which is perfectly consistent with the radar data, and that the ascent would likely result in further tracking by the radars, unless the disappearance was caused by reaching the range of radar’s coverage. The later is unlikely explanation, if the plane was tracked by Thai Phuket, or Indonesian Lhokseumawe, or Sabang radar.

  85. Peter Norton says:

    > Victor Iannello:
    > “@Peter Norton: Welcome, and thank you for the link.”

    Thank you. You are very welcome.

    > From what I can determine based on the surviving comments of Emil
    > Enchev and the comments from others, Mr Enchev doubts the validity of
    > the BTO data because processing delays are indeterminate.

    Yes, he wrote:
    “Inmarsat dont able to determine Distance to the MH370 by its ping. ACARS-SATCOM SYSTEM have INDEFINITE INTERNAL DELAY that dont allow such calculations.”
    I read his remark the same way you do.

    > The group MH370 Chinese Families seems to be more interested in
    > promoting conflicting theories than presenting facts.

    Or maybe they don’t act in bad faith but understandably just cling to anything seemingly providing any hope ?

  86. David says:

    @Gysbreght. Thank you for reminding me about the information the ATSB provided you. As discussed, in these more recent simulations they varied speed, fuel, electrical configuration, altitude and turbulence. Musing, since the only scenario where fuel remaining could be varied was electrical (right-engine-first), 1-6 and 9 were subject to just speed, altitude and turbulence variations. It may be that the right-engine-firsts, 7, 8 & 10, were subject just to fuel changes.

    Of those descents that ‘qualified’ by descent rate characteristics, 100% of the right-engine-firsts did but just 2 of 7, or about 30%, of the left-firsts. A broad takeaway is the indication that loading does not affect BFO descent rates or their increase much; at least at altitude (I disregard data base exceedance, which I presume was lower down) though there is some distinguishing feature of 10 which turns it right. As to the low rate of qualification of the left-firsts, having speed, altitude and turbulence variation (maybe 2 of those have turbulence variation, and the remaining 5 have speed or altitude) there is sensitivity to those changes, obviously.

    Thus I assume these recent simulations might disqualify some speeds and altitudes albeit with very small sample sizes. Adding the earlier simulator outcomes to these 10 simulations might account for the ATSB confidence that they have it right. Here arises though the issue whether the simulation results are seen to prove that there would have been wide BFO compliance (a very high qualification rate) or are just a means of demonstrating that compliance is possible (a low rate would do), then from the latter deducing that the actual must have been one of those.

    Another feature is that with 100% of the right-engine-first trials (again, small sample!) having both ‘qualified’ and being “highly unstable” I think you get a clue that high instability at altitude could be expected for qualification more generally, though that seems pretty obvious. I would be interested to know though whether rapid pitch down instability was manifest, which is my opinion as to what happened as you are aware.

    To me their phrase indicating the 7th arc would move in a right engine first scenario is unambiguous though I do not understand it. I had it in mind that the aircraft could fly on past the 7th arc, depending on fuel left for the left engine APU combination but I withdraw that: I overlooked there would be no autopilot.

    You say, “The scenario did meet the BFO values but (I suspect) not within 8 seconds from the 7th arc (time of log-on request message, two minutes after loss of electrical power and autopilot)”. You said earlier that when you asked, “in which trajectories this occurred …”, the ATSB answered 3,7,8,9,10. I took it that ‘this’ entailed meeting both descent rate criteria and the 8 sec rate change. Do I take it that in fact ‘this’ includes just descent rates? In the above I have supposed it included both.

    As to the APU start, I do think they make clear the simulator makes no fuel provision for that or for that matter engine restarts and in both cases describe the effect of them conditionally; that is, “if” those had happened. Somehow though they are confident the APU start did happen, even though engine starts are left aside. As they say, an APU start can, “re-energise the AC buses and some hydraulic systems” and, “This could affect the trajectory of the aircraft”. Also affecting that would be APU inlet opening, which they have overlooked.
    Even though right engine first is seen as unlikely they should have addressed the effect of these non-hypotheticals on trajectory and new search probabilities.

    The specs for these 10 added trials should be releasable by the ATSB, though apparently not the outcomes; though they are of little use without.

    Out of the above points, the most important question which emerges is the BFO timing issue you have raised. Though probably related, the phrase under fig 6, the simulations being, “aligned at a point consistent with when the final BTO transmission may have occurred” is double-Dutch to me.

    Whether or not the ATSB has overlooked the APU inlet opening time, the 2 minutes is an estimate only (APU start time dependent on such as altitude, attitude?) I hope they do not have results which are sensitive to that, though I suspect they are, particularly as to when and for how long their 8 secs compliance window occurs/lasts. I doubt they will be free to answer, or describe this in any report, though if that is the case your question about whether they have actually asked Boeing about this would be apt.

    Overall, I regret this has decreased my confidence in the new search width, that being unhelpful.

    PS I do not know why your particular interest in simulation 3 (“and in the scenario for trajectory #3 …”) does not extend to 9; or whether 9 & 10, being opposite engines’ failures-first, tracking so closely together is a coincidence.

    @Oleksandr. You missed my last sentence.

  87. Oleksandr says:

    @David,

    No, I did not miss your last sentence. I only stressed once again why I think ATSB’s attempt to reproduce -2 Hz makes no sense. They have to reproduce the BFO of 273 Hz in the same way, and to do that they must abandon their constant altitude assumption at 18:25. Or, alternatively, they have to include one more item in their list of possible reasons, which is the erroneous Doppler correction term. Why did they omit this potential cause?

    It has been clearly demonstrated by the DSTG that there is some mysterious ‘geographic dependency’ in the BFO error (compared to models) up to 20 Hz, which cannot be explained by the items listed in the ATSB report (cited by Gysbreght above). This is a solid proof that ATSB’s list is incomplete. What are other possibilities? I can suggest only the Doppler compensation term.

  88. Gysbreght says:

    @David:

    The ATSB stated in its report dated 3 December 2015:
    “In the case of MH370, due to the individual engine efficiency, it is likely that the right engine
    flamed-out first followed by the left engine. Given the amount of fuel uplifted in KL and historic fuel
    burn data for each engine, it is estimated that the left engine could have continued to run for up to
    15 minutes after the right engine flamed-out.”

    Therefore, unless stated otherwise, I assume the right engine flamed out first in all simulations. In the electrical configuration where the loss of engine power from one engine resulted in the loss of autopilot (AP), there would be no thrust asymmetry compensation while the remaining engine continued to operate either at the thrust it had before the flame-out, or at maximum cruise thrust. Trajectory 9 is virtually identical with trajectory 10. Although it was not in the same electrical configuration as no.s 7, 8, 10 it must have been in a configuration that resulted in the same directional trim asymmetry.

    Trajectory no.3 is the one I don’t understand. It has the widest turn radii of all, and in the later stages distinct indications of phugoid motion. Therefore the phugoid must be the explanation for the high rate of descent observed, whereas in the other case where that occurred it was due to increasing bank angle, turning into the spiral dive that is evident at the end of those trajectories. I would like to know the scenario that produced a phugoid with such high amplitude.

    You said: “You said earlier that when you asked, “in which trajectories this occurred …”, the ATSB answered 3,7,8,9,10. I took it that ‘this’ entailed meeting both descent rate criteria and the 8 sec rate change. Do I take it that in fact ‘this’ includes just descent rates? In the above I have supposed it included both.”
    My question, and the ATSB’s reply referred to the text of the bullet, which included the phrase “increase in descent rate across an 8 second period (as per the two final BFO values)”.

    You said: ““aligned at a point consistent with when the final BTO transmission may have occurred” is double-Dutch to me.”
    If you follow the trajectories in Figure 6 from the top, they all pass through a single spot about two-thirds down. That point is where the 7th arc woud have been in the simulations, two minutes after the loss of the autopilot.

  89. Don Thompson says:

    @Mick G

    Well done on your call out of the Australian/News AU misrepresentations.

    :Don

  90. Gysbreght says:

    @David: You said “However that does not address your 18:25:34 273Hz, though that was explained as I remember it by Dr Bobby Ulich a couple of weeks ago.”
    I recall drB’s hypothesis, and found it very interesting but forgot the details. Could you direct me to that post, or perhaps drB could chime in?

  91. Oleksandr says:

    @Mick Gilbert,

    “given the range of conditions airplanes safely fly in, the generation of lift is hardly a delicate affair. ”

    Let me disagree with this. It is actually a very delicate affair. There are many effects, especially when it is about the turbulence. Let’s take, for example, so-called “vortex detachment” effect, which can cause an abrupt loss of the lift. Or approaching M=1, when some wing and other control surface areas became supersonic, while others remain subsonic.

  92. ROB says:

    @Donald

    Z’s plan went smoothly – a sure indication of careful, advanced preparation. The recovered debris points to a controlled impact with the water. The bulk of the debris consists of RH structural components, in particular, RH wing trailing edge components. This would have been the most difficult part of the plan to pull off – how do you control the impact in such a way that the aircraft sinks rapidly but leaves little tell-tale debris on the surface? Well, firstly you choose as remote an area as possible to comes down in, given the available fuel. Tick that one! Then you have to control the speed and descent rate in such a way that the airframe sinks rapidly without breaking up. Almost impossible to guarantee a totally successful outcome, no matter how much you plan in advance. The RH flaperon, RH outboard flap, RH engine cowling fragments got away, so not 100% success here. Interestingly, Z is reported to have researched previous accidents, on the net. AF447 would have been of particular interest.

    I cannot imagine the Aussies having any competent, psychiatrists. A few amateur ones, possibly.

  93. Gysbreght says:

    @David: With regard to trajectory no.3, perhaps I should have added:
    The large radius of turn indicates a low bank angle. I associate a low bank angle with a low average rate of descent, because I demonstrated in the analysis of ALSM’s simulator test that the rate of descent is driven by the bank angle.

  94. Victor Iannello says:

    @Peter Norton said, “Or maybe they [MH370 Chinese Families] don’t act in bad faith but understandably just cling to anything seemingly providing any hope ?”

    Perhaps, but repeatedly citing the doubtful claims of Emil Enchev hurts their case. In the same release, they question the satellite data and then ask for the search to continue along the 7th arc.

  95. David says:

    @Gysbreght. Thanks. Belated correction. By right engine first I meant AC loss on right engine failure before left, likewise left engine first meant AC loss at left engine failure. Perhaps that was apparent but if not that might now aid comprehension.

  96. Ge Rijn says:

    @ROB

    To make your list of confirmed debris more specific and to correct something:
    There is also the Left wing L1 flap track fairing, the Left wing outboard flap trailing edge piece, the Left wing Flaperon closing panel.
    This makes 3 to 4 pieces Left wing/ Right wing. Fairly even.
    The two engine cowling pieces could not be connected to the left or right engine.
    Then there is the nose gear door piece. And a number of other control surface related pieces not yet confirmed. All pieces you could expect seperating during a ~level, low AoA, relatively low speed entry on the water.
    It’s the sheer number of exactly this kind of pieces in relation to the few (3) cabin pieces that leave no other conclusion IMO.
    The cabin could have been breeched on a point or a door got open or seperated leaving some cabin debris to float out.

    The debris found from the tail (hor/vert stabilizer) seems more problematic but in a low AoA entry the tail would hit the water first and take the brunt of the impact also subjected to seperating debris later from the engines and wings.

    But also here is we have the problem of lack of data. Very little data on the debris investigation is available. No detailed reports. No independent cross-investigation. Only those few public overview reports the ATSB has published. There must be much more.

  97. ROB says:

    @Ge Rijn

    Nice to speak to you again. BTW, Brexit appears to be progressing nicely! Theresa May has just played a blinder and called a snap election. Things seem to have normalised a bit in Holland – it got quite ugly when the Turkish President threw a tantrum and made all sorts of unreasonable and unfounded accusations.

    Now re the debris: As I understand it the only piece confirmed as coming from the LH side, is the small piece of outboard flap trailing edge. The flap track fairing section you mention, actually belongs to the RH outboard flap, and the flaperon seal panel you mention probably belongs to the RHS. There is no evidence tying it to the LHS, in my opinion. The larger of the two engine cowling pieces was identified as coming from the RH engine, which suggests the RR piece also came from that side.

    Sorry mate, your information sources are out of date.

  98. Gysbreght says:

    @David: “Perhaps that was apparent …”

    No, it wasn’t, at least not to me.

    … the right-engine-firsts, 7, 8 & 10, were subject just to fuel changes ???

    Never mind.

  99. TBill says:

    @ROB
    “This would have been the most difficult part of the plan to pull off – how do you control the impact in such a way that the aircraft sinks rapidly but leaves little tell-tale debris on the surface?”

    My latest hypothesis, and this is a long shot so somebody has to invite me to look at a B777 to see if this is possible, is perhaps the aircraft was intentionally landed under negative pressure. This would probably necessitate tampering with the 4 negative pressure relief valves in the forward cargo bay. I suppose if the aircraft was ditched fast the relief valves may not open in time for complete repressurization, but the relief valves are probably quite effective.

  100. Ge Rijn says:

    @ROB

    Nice to see you again too. It’s getting a long way isn’t it..
    Just a short remark on May and Holland.
    May got chicken and more sensable and the Dutch proved again they don’t except intimidation also from the highest level (like the ATSB has done according @VictorI’s topic).

    On the debris. You should know I do some homework;

    http://www.mh370.gov.my/index.php/en/426-debris-examination-reports-28-feb-2018

  101. Gysbreght says:

    @TBill: What would be the effect of negative cabin pressure differential, for example in the impact conditions depicted in Victor Iannello’s excellent video “45S2 no pilot input.wmv”?

  102. Ge Rijn says:

    @ROB

    If you have more recent info I’m ofcourse glad to read.

  103. Ge Rijn says:

    @TBill

    Please can you explain to me what you mean by negative pressure in this regard.
    A (positive) pressurized cabin above certain altitude descending to higher pressure on lower altitide making the cabin pressure negative?

    Or a neutral (low pressure) decompressed cabin on high altitude descending to low altitude making this pressure negative towards the outside pressure?

    Do you suggest possible implosion or explosion effects?

  104. ROB says:

    @Ge Rijn

    Thank you. Yes, you have done your homework, and thanks for sharing it with me. Very interesting. But it does confirm the larger engine cowl piece as from the RH engine, as well as the flap track fairing as from the LH wing.

    Good stuff.

  105. TBill says:

    @Ge Rijn
    I am postulating an implosion to control debris. I am under the belief, under normal pressure, there would tend to be a explosive ejection of debris if a plane hit the water hard.

  106. TBill says:

    @Gysbreght
    I do think I recall that video flight sim of hitting water. I stress this is a long shot theory about hitting the water under vacuum. I’d be looking for this idea explaining some observations, if it has any hope. For example, messing up the BFO/BTO, plane flying longer due to lack of bleed air etc. The possible actions I see required would be manual close of outflow valves, cutting bleed air, setting landing altitude at max FL140, and probably tampering with the negative pressure inflow valves which will quickly bring in air I assume. Also a fast descent may be required as air will be leaking in and the hull may not stand up before impact.

  107. Ge Rijn says:

    @TBill

    I can believe also under normal pressure, but even under negative pressure (what’s the significant difference?) there could/would be an explosive ejection of debris if the plane hit the water under high speed or extreme pitch down.

    As I postulate the debris trail shows no signs of a pressure related explosion. Maybe a door blew out resulting in some cabin debris floating out but even that seems unlikely to me because of pressure, rather by impact (like Aisiana 124).

    Other there is no found debris at all to support a high speed explosive impact.
    ~90% trailing edge, wing related, surface control related, engine related, nose gear door related parts.
    Only three light cabin parts that show no sign of very high speed impact damage. All reqocnisably intact. Nothing else found still.

    Kind of implosion would have happened in case of a ditch-like impact. Also the Hudson ditch caused considerable implosion damage to its belly. But not by pressure difference but by impact force of the water.

    I cann’t think of a way a pilot can put the cabin pressure higher than the pressure on low altitudes. The cabin pressure must be either equal or lower. Resulting in implosion rather than explosion.

    Anyway I don’t believe this pressure-issue has been a factor.
    The debris tells its story IMO: ~level entry, ~low AoA, relatively low speed impact.

  108. Ge Rijn says:

    @TBill

    Forget to say I appreciate you bringing up an issue like this.
    Looking at every possible angle as always.

  109. TBill says:

    @Ge Rijn
    Part of the “explosive” nature of an impact is due to the compression of the air inside the aircraft which then tends to generate extreme pressure inside the aircraft. I have no idea if this could be prevented by landing under vacuum, but I am asking if anyone thought of this for MH370? Or any prior air crash ever? It does not have to be a severe impact, or severe vacuum, if the fuselage broke open, the vacuum could rapidly bring in water, depending on the geometry of the impact.

    As Will over on JW says, “go wide or go home” so I am going wide (out of the box) here.

  110. David says:

    @Gysbreght. “… the right-engine-firsts, 7, 8 & 10, were subject just to fuel changes ??? Never mind.”

    My apologies for my aberration.

    Translation from my correction,”…AC power loss at right engine failure, 7, 8 & 9″. Is that your query?

  111. David says:

    Gysbreght. …7,8 & 10

  112. Gysbreght says:

    @David: Let it rest. You’ve explained what you meant. I appreciate your comments.

  113. David says:

    @Gysbreght. “Let it rest”. Pleased to.

    “I recall drB’s hypothesis, and found it very interesting but forgot the details. Could you direct me to that post, or perhaps drB could chime in?

    He seems to be off-air. With any luck it was:

    https://drive.google.com/file/d/0BzOIIFNlx2aUUUVycGJLM21SX2c/view

  114. David says:

    @Rob. “….. it does confirm the larger engine cowl piece as from the RH engine” Right hand cowl but could be from either engine.

    First item:

    http://www.mh370.gov.my/phocadownload/3rd_IS/Debris%20Examination%20Reports%20-%2028022017.pdf

  115. Victor Iannello says:

    @David, @Gysbreght, & @Oleksandr:

    @DrB’s work is better understood after reading Ian Holland’s recent paper on the BFO.

  116. Gysbreght says:

    @David: Thanks for that link. But, I’m sorry, that’s not what I had in mind.

    In a post here some time ago, drB compared the BFO logged in the log-on request message at 18:25:27.4 with that immediately following it at 18:25:34.5, and the same for 00:19:29.4 and 00:19:37.4. He noted that in both cases the difference was similar in magnitude but opposite in sign, consistent with the sign and approximate magnitude of the north-south component of the airplane groundspeed in both cases. (If I remember that correctly.)

  117. ROB says:

    @David
    @Ge Rijn

    David, you are right. I misread the report. RH cowl, but could be either RH or LH engine. I discovered my error an hour or so ago, and was about to tell Ge Rijn.

    I didn’t see this report until now. I had to “switch off” after they abandoned the search, to preserve what was left of my sanity. The recent Australian article, particularly the suggstion that some ATSB might be disagreeing with the official position caught my attention.

  118. DennisW says:

    @VictorI

    FWIW, I did request ac copy of the reference [7] of the Holland paper which is cited several times in his published paper. This reference might be very relevant to the events before and after 18:25:27.

    [7] “Internal study regarding SATCOM ground-station logs,” MH370 Flight Path Reconstruction Group – SATCOM Subgroup.

    Nothing. Not even an acknowledgement.

    Same can be said for some additional information I requested relative to the data from the previous flights of 9M-MRO. Nothing.

    Hard to imagine why this information would fall into any restricted category. It is also hard to accept why an author would refuse to provide a reference cited in the body of his published work.

  119. TBill says:

    @David @Gysbreght
    Thank you for the link to DrB’s path of 17-March…not sure I saw that one. I agree with DrB’s path as far as it looks like BULVA or BEDAX to ISBIX or MUTMI then South goes thru the Arcs quite well. There was an earlier 9-Jan version with some rationale of the various turns.

  120. Gysbreght says:

    @DennisW: “Nothing. Not even an acknowledgement.”

    As for my last two emails to the ATSB. Just asking for information. No rant.

  121. David says:

    @Gysbreght. Consider calling Dan O’Malley. The media room international number is listed as +61 2 6257 4150.

  122. David says:

    @Rob. You switched off, “to preserve what was left of my sanity.”
    Would it help me find my use-by date?

  123. Oleksandr says:

    @Victor,

    To justify the proposed hypothesis of the “oscillator warm-up drift” Holland needs to get rid of 18:25:27 BFO. In his paper he states:

    “Regarding the log-on event at 18:25:27Z for MH370, it was noted that there was a non-zero bit error rate (BER) associated with the log-on request at that time. The associated received signal level and carrier to-noise density ratio (C=N0) were also unusually low. As such, the first BFO was deemed untrustworthy.”

    Firstly, it does look like Holland introduces another remarkable coincidence, as this particular BFO and associated BTO perfectly match position and velocity of MH370 derived from the radar data.

    Secondly, a very interesting peculiarity is the signal level. I was talking about this missing column of Inmarsat logs from the time of the public release. I don’t know why they cut it out and claimed that the logs were unaltered. Anyway, would not the low signal level be consistent with antenna steering problem?

    Thirdly, Holland forgot to mention that these abnormal BFOs are paired with abnormal BTOs. Also warm up effect? If yes, then for some mysterious reason it did not affect 18:25:27 BTO. I did not see such a long BTO on the first logon in KL. Was it observed in other flights?

    Fourthly, DSTG noted “geographic dependence” of BFO, which obviously cannot be explained by any item from ATSB’s list. What is the cause if not the compensation term?

  124. Oleksandr says:

    @Dennis,

    Re: “Nothing. Not even an acknowledgement.”

    As we recently learnt, ATSB staff can be imprisoned for 2 year, if he/she acknowledges you request without ATSB’s chief’s permission.

  125. DennisW says:

    @Oleksandr

    I thought we were discarding that as nonsense journalism produced by The Australian.

  126. Victor Iannello says:

    @All: Yes, I understand the frustration felt by those that have requested more information from the ATSB about BFO analyses and BFO data. Without going into details, Mike Exner and I have had similar experiences with related requests, including an interest that Mike expressed in learning more about BTO and BFO errors as a function of C/No and BER. At this point, I don’t think much more information will be forthcoming.

  127. DennisW says:

    @Oleksandr

    “Fourthly, DSTG noted “geographic dependence” of BFO, which obviously cannot be explained by any item from ATSB’s list. What is the cause if not the compensation term?”

    The cause is obviously oscillator drift. As soon as the DSTG computed a mean and variance for BFO they violated a fundamental statistical principal. The mean and variance are meaningless for processes that are not both ergodic and stationary. Oscillator behavior is neither.

    The second thing I asked Holland for, and received no response, are plots for other flights like Figure 4 in the DSTG book. You will find that all of them behave in a similar fashion. Figure 4 simply representing the worst case observed.

  128. DennisW says:

    @VictorI

    “At this point, I don’t think much more information will be forthcoming.”

    No doubt, but how can a PhD level person, Holland, not provide the reference he cites numerous times in an authored paper. That is an inexcusable breach of scholarly etiquette. Unbelievable, IMO. If you are not prepared to provide your references, you should not invoke them. It really is that simple.

  129. DennisW says:

    @VictorI

    Just for shits and grins I went back to the Holland paper. He invokes reference [7] seven times. Seven f’ing times, and he will not provide it.

    I am not really listening to your and Exner’s interactions with the ATSB. As near as I can tell you are both in love with those guys. Likewise Andrew and Mick. No point in even reading your comments in that regard. I am a Higgins fan simply because he is calling it like I see it. The ATSB are bunch of inept and dishonest wankers.

  130. Mick Gilbert says:

    @DennisW

    Whoa, whoa, whoa! Slow your roll there, Digger. Just because I’m Australian does not mean that I am a fan of, leave alone in love with, the ATSB. I have been roundly critical of the ATSB in any number of public forums for, amongst other things, their failure to release the information that they have. That said, I understand the practicalities of the matter; the ATSB is not the Annex 13 investigative authority, the Malaysians are.

  131. sk999 says:

    ALSM,

    Many thanks for keeping tabs on the ATSB website. CSIRO managed to acquire and chop down to size an actual B-777 flaperon.

    Amazingly, the only contribution of the French to this exercise was to provide photos. One would have thought that they would have expedited this study by providing the genuine article itself, or, in the alternative, have participated in the study, make sure that the chopped-down replica was accurate, and been coauthors on this report. Why is it that the term “disfunctional” comes to mind whenever the French are involved?

    It would also seem that there is much information to be teased out of the multitude of other pieces of wreckage that have been recovered.

  132. DennisW says:

    @Mick

    I am asking questions relative to information the ATSB and their helpers have already published. They put the info into the public domain. I am not generating questions relative to anything that is not already out there. If you cannot answer a question about what you published or provide a reference to which you refer how lame is that? It has nothing to do with Annex 13.

  133. Mick Gilbert says:

    @DennisW

    I’m not here to defend the ATSB and I can appreciate the frustration with unanswered queries; I’ve never received even so much as an acknowledgement to my requests for information from any of Boeing, PPG Aerospace or Malaysia Airlines.

    However, this recent round of discussion has got me thinking about this from a different perspective; I’m now wondering if there has been another air crash investigation where as much information has been put into the public domain prior to the publication of the final report as MH370?

  134. DennisW says:

    @sk999

    “It would also seem that there is much information to be teased out of the multitude of other pieces of wreckage that have been recovered.”

    Just consider the flaperon itself. I have no doubt competent forensics could determine whether the ragged edges were produced by flutter, the ALSM position, or by stresses incurred by a water impact. That info would be extremely helpful relative to reinforcing the rapid dive theory or the controlled low angle impact. How hard can that be, and why is that information not forthcoming? It truly is maddening.

  135. DennisW says:

    @Mick

    “However, this recent round of discussion has got me thinking about this from a different perspective; I’m now wondering if there has been another air crash investigation where as much information has been put into the public domain prior to the publication of the final report as MH370?”

    Good question. I need to noodle that one a bit since MH370 is so unique.

  136. sk999 says:

    DennisW,

    My comment r.e. “much information” referred to drift studies, not what stresses and subsequent damage may have happened during the final descent and impact. I have zero expertise in the latter. Further, even if I had any expertise, it is doubtful if it would contribute one iota to determining the location of the wreckage.

  137. DennisW says:

    @sk999

    Same here. I have avoided the whole final descent discussion because it is well outside my range of competence. I am more concerned with getting to +/- 100 nm at 00:11.

  138. ROB says:

    @David

    Will “switching off” help you find your use by date?

    I think it high unlikely. It gives you a temporary respite only, until something conspires to draw you back into the maelstrom.

  139. Gysbreght says:

    @Mick Gilbert: ” I’m now wondering if there has been another air crash investigation where as much information has been put into the public domain prior to the publication of the final report as MH370?”

    Have you browsed through the NTSB’s docket for high-profile accidents like the Hudson ditching or the Asiana crash at SFO?

  140. Mick Gilbert says:

    @Gysbreght

    Yes, of course I am familiar with the investigation report registers maintained by ICAO, the NTSB, the AAIB, the BEA, etc, etc.

    My question was,

    “I’m now wondering if there has been another air crash investigation where as much information has been put into the public domain PRIOR to the publication of the final report as MH370?”

  141. Gysbreght says:

    @Mick Gilbert: “I’m now wondering if there has been another air crash investigation where as much information has been put into the public domain PRIOR to the publication of the final report as MH370?”

    The French BEA published the final report on the AF447 accident in July 2012, three years after the accident on 1st June 2009. During those three years BEA’s interim reporting was no less extensive than that of the ATSB on the MH370 search activity. I never had the impression that information was being witheld, and do not recall concerns in that regard being expressed in the discussion forum (PPRUNE) where I participated at that time.

  142. ALSM says:

    Here is an MOT presentation given yesterday to Chinese NOK:
    https://goo.gl/nED3O8

    Nothing new, but interesting to see what MOT considered important to highlight.

  143. Gysbreght says:

    @Mick Gilbert: One area where BEA’s interim reporting was quite extensive was the analysis of the mode of failure of debris recovered, in particular the vertical stabilizer.

  144. Oleksandr says:

    @Dennis,

    “The cause is obviously oscillator drift. As soon as the DSTG computed a mean and variance for BFO they violated a fundamental statistical principal. The mean and variance are meaningless for processes that are not both ergodic and stationary. Oscillator behavior is neither.”

    How can the oscillator drift justify the mysterious “geographic dependence” noted by the DSTG? I would suggest that this effect is rather related to imprecision of navigation data fed into AES.

  145. Ge Rijn says:

    @ALSM

    New to me is the latest two finds from SA are also confirmed higly likely MH370.
    The big no.7 flap track fairing piece (which is the second no.7 fairing piece/Liam Lotter) and the rather big chunk of Right wing aileron.

    That makes 16 confirmed, almost certain, higly likely or likely trailing edge/wing/control surface related pieces and 1 nose gear door.
    Are you still convinced most of those pieces seperated due to flutter?

  146. Oleksandr says:

    @ALSM,

    I do not trust CSIRO’s studies anymore because their conclusions are obviously dictated by the ATSB, who commissioned their study. Specifically, CSIRO plays with the leeway factor to justify what they were ordered to justify. And they have succeeded in this difficult task.

  147. Oleksandr says:

    @Victor,

    Re: “Yes, I understand the frustration felt by those that have requested more information from the ATSB about BFO analyses and BFO data. Without going into details, Mike Exner and I have had similar experiences with related requests, including an interest that Mike expressed in learning more about BTO and BFO errors as a function of C/No and BER. At this point, I don’t think much more information will be forthcoming.”

    Then why are you so frustrated by the unwillingness of Malaysians to release military radar data, but not frustrated by the unwillingness of the ATSB and DSTG to release BTO & BFO related data, which may potentially be more useful than the radar data?

  148. Victor Iannello says:

    @Oleksandr: To be clear, I’d like all the relevant data released by whichever party controls it. The reality is that much of the data we seek is owned by Malaysia, Boeing, Thales, and Inmarsat, and confidentially shared with the ATSB. Considering the failure to find the plane, I think more should be released so that others can independently review and analyze the data.

  149. Oleksandr says:

    @Victor,

    This time I totally agree with your formulation.

  150. Oleksandr says:

    @Sk999,

    Re: “Amazingly, the only contribution of the French to this exercise was to provide photos.”

    No. They have conducted hydraulic lab tests of the flaperon. They have determined the properties of the two positions the flaperon could float (the leeway and drift angle). They have run a reverse drift model, which is consistent with the majority of the forward drift models. They did it long time ago. See JW blog a year ago:

    http://jeffwise.net/2016/05/02/french-judiciary-report-raises-fresh-doubts-about-mh370-debris/

    I am actually very interested in the properties of the second large fragment, which ended up in Tanzania. Why were French allowed to publish some data despite of confidentiality, while the ASTB uses some ridiculous bureaucratic excuse to justify the silence?

  151. Ge Rijn says:

    @Oleksandr

    Just to add something I noticed from the CSIRO-report (not on your comment to @ALSM):

    ‘Figures 3.1 and 3.2 also show that
    a crash location near 35°S is one that results in the flaperon
    being unlikely to have reached Australia. This cannot be said of other potential crash sites south of 32°S, but can be said of the 32°-30.5°S segment.’

    Firts I think a mistake went in the above text; ‘crash sites south of 32S’ I think should be south of 35S.
    Anyway I read they make a note the 32-30.5S segment is also a good possibility.

    Later they state:

    ‘Debris would very likely have been seen if the crash had been
    between 32°S and 27°S. The same cannot be said of the zone between 38°
    S and 33°S, which was not searched, or only lightly searched (34°S-33°S), thus leaving it as a possible location of the aircraft.’

    It seems obvious the study rejects the 32-30.5S segment (or even till 27S) almost exclusively based on the (lack of) aerial search results between 38 and 33S.
    They seem to forget this area was thoroughly search by side scan sonar without result.

    This area has been rejected by now even by the ATSB.
    So I really don’t understand why CSIRO keeps holding to ~35S with this kind of strange arguments.

    Aerial search has not spotted/found anything anywhere.
    To conclude then it must have been in an area you did not aerial search (but which has been searched extensively by sonar) sounds naïve to me or maybe even dictated as you suggest.

    I suppose they had bad luck with the aerial search by coming too late on the scene dealing with a very small, already widely disperged debris field.

    .

  152. Ge Rijn says:

    To mention something else from the CSIRO report I found quite interesting.
    And maybe of interest to Jeff Wise too (I know I should probably comment on his blog but his current topic is Germanwing, besides I’m sure he will read this blog too).

    In the video provided you can see how easily the flaperon is flipped over noted every minute by wind and waves.
    IMO this could well explain why barnacles on both sides could grow and survive for a considerable time.

  153. Victor Iannello says:

    The ATSB finds itself in a logical dilemma. Based on the results of the new drift studies, CSIRO has high confidence that the plane will be found near 35S latitude. Additionally, the BFO results say the plane should have crashed close to the 7th arc. However, this portion of the 7th arc was declared unlikely by the DSTG Bayesian study. It was also already searched close to the 7th arc without success.

    Further north along the arc has been determined to be unlikely because a debris field was not detected by the aerial search. Yet, a search zone centered at around 30S was the ATSB’s recommendation in June 2014, just on the heels of the aerial search.

    And despite the ATSB’s and CSIRO’s high level of confidence that wreckage will be found near 35S latitude, Malaysia and Australia are not willing to resume the search there.

    If the ATSB truly believes the plane will be found near 35S, but further out from the 7th arc, it needs to explain why the DSTG results are wrong, and why the BFO analysis is wrong.

  154. DennisW says:

    @Oleksandr

    “How can the oscillator drift justify the mysterious “geographic dependence” noted by the DSTG? I would suggest that this effect is rather related to imprecision of navigation data fed into AES.”

    Earlier I made an incorrect reference – it is figure 5.4 not figure 4 of the “Bayesian Methods…”book. The associated comment in the book is cut and pasted below.

    begin cut-paste//

    The residual error is clearly not zero-mean, and the mean varies with time. Substantial effort was made to characterise this structured bias. It was found to have a geographic dependency but it has not been possible to determine a quantitative function to compensate for this change in bias.

    end cut-paste//

    I regarded the notion of a geographic dependence raised by the DSTG to be in error. It is really why I asked Holland for BFO error plots from other flights. My position is they would look very similar i.e. resembling a random walk. The DSTG notes above “the mean varies with time”. A cut paste from Wikipedia below relative to stationarity.

    begin cut-paste//

    In mathematics and statistics, a stationary process (a.k.a. a strict(ly) stationary process or strong(ly) stationary process) is a stochastic process whose joint probability distribution does not change when shifted in time. Consequently, parameters such as mean and variance, if they are present, also do not change over time.

    end cut-paste//

    It is well known that oscillator statistics do not exhibit stationarity. The oscillator frequency (despite oven control) will change with external temperature, vibration, and a slower aging change with time.

    I just went to the garage and took a photo of a GPS receiver mounted on one of my motorcycles to illustrate the vibration mitigation routinely used for off road applications. Without the shock mount GPS performance is noticeably impaired. Temperature and aging are non-issues while actually tracking satellites.

    http://tmex1.blogspot.com/2017/04/gps-shock-mount.html

  155. Oleksandr says:

    Wouldn’t it be easier to name some reason why the aerial search at 30S has failed? And also remove 0.73M threshold from the DSTG study to resolve the dilemma. Perhaps not, because they would face another dilemma.

  156. Don Thompson says:

    The documentary record provided by Richard Cole shows that the deep ocean search covered Arc7-S35º, primarily by Go Phoenix/Phoenix International and the ProSAS60 towfish, but with good overlap from Equator and Discovery.

    The deep ocean search around Arc7-S35º has been deemed to deliver a null result. The deep opean search at Arc7-S35º didn’t go out to +/-40nm, I would estimate from REC’s graphics +/-20nm.

    :Don

  157. Victor Iannello says:

    @Don and @Oleksandr: Yes, considering the collective results of the Bayesian study, the BFO analysis, and previous scanning, I would not be nearly as confident as CSIRO and the ATSB that the plane will be found near 35S latitude. Quite a dilemma. Of course, if there is no further searching, the ATSB is safe from being proven wrong once again.

  158. Andrew says:

    @DennisW:

    “I am not really listening to your and Exner’s interactions with the ATSB. As near as I can tell you are both in love with those guys. Likewise Andrew and Mick. No point in even reading your comments in that regard.”

    No Dennis, not in love with the ATSB at all, but I think I have some understanding of the restrictions the ATSB faces when it comes to information sharing, which is more than I can say for some commentators here. You can bury your head in the sand if that’s your preference, but the simple fact is the ATSB can’t share all the information you desire, for a variety of reasons that have already been explained but you refuse to accept.

    To be honest, the tone of the discussion over the last few days has been extremely disappointing. I understand the frustrations at not being able to access the information you desire, but to be blunt, it’s about time a few people here stopped acting like spoilt children. Calling the ATSB a “bunch of inept and dishonest wankers” might make you feel good, but I doubt it will help your cause.

  159. DennisW says:

    @Andrew

    “the ATSB can’t share all the information you desire, for a variety of reasons that have already been explained but you refuse to accept.”

    I am asking the ATSB about information they have already published. How can a sincere scientist publish a paper that includes seven references to a document that is not publicly available? If the document was not referenced, I would not be asking about it.

  160. TBill says:

    @all
    I was just about to ask ATSB about what we know about jet fuel quality, now it seems my timing is bad. Perhaps there is some way ATSB could handle questions in some kind of triage: (1) things they can answer , (2) things that could be forwarded to MY to be considered for final report, etc.

    @Mick Gilbert @Gysbreght
    As an NTSB example of a crash, TWA Flight 800. That was during the the early of the internet, not sure Google yet existed, at that time I used a search engine called Dog Patch. Within a month of the crash I was able to see fuel tank explosion was suspected cause (missile was the leading favorite press theory). From that, I knew immediately that fuel tank issue was the likely explanation and put some of my efforts there. Missile conspiracy theories still very active for TWA800 unfort…one can only imagine MH370 legacy. TWA800 was very open from my perspective, lots of different industry groups working together to arrive at a solution. No big secrets.

  161. Ge Rijn says:

    @Andrew

    I feel talked to too so respond. Dispite the ATSB ‘rectification’ on the ‘bad news’ the Australian published a signal has been given to all officially connected to the investigation. The Australian is no tabloid paper dealt for free in train stations. We even cann’t read it unless to pay for it online.

    I cann’t image those journalist did not do their homework.
    To me it’s a clear signal the Australian Governement has used its power to mouth-shut the ATSB about internally deviding issues and some urge to go public by some.

    The need to do this intervention only raises the superstition there must be crucial information to hide to outsiders like us and the NoK.
    With no need actualy. An ongoing investiagation is by definition restricted in providing details. Almost everyone understands.
    MH17 is still under investigation but silent.
    There should be no need for venting threats and mentioning law-articles.

    Anyway IMO this does not help the ATSB to be taken 100% seriously as an indepedent organistation.

  162. TBill says:

    Re: CSIRO
    Of course 35-S is pretty much right on DrB’s proposed path end point, I am thinking.

  163. Ge Rijn says:

    @TBill

    It could be but then the high speed dive near the 7th arc would not fit for nothing was found there.
    It would actualy only serve as additianol prove there was no high speed dive.

  164. Oleksandr says:

    @Andrew,

    ” but the simple fact is the ATSB can’t share all the information you desire, for a variety of reasons that have already been explained but you refuse to accept. ”

    It is understandable that the ATSB cannot share all the information, but it is ridiculous that they refuse to provide information, which has nothing to do with security or safety, and which was gathered by themselves. The legislation you cited was apparently written either by drunk lawyers, or unqualified staff, who had poor understanding of what they wrote.

  165. Oleksandr says:

    @Dennis,

    Re: ” am asking the ATSB about information they have already published”

    They have not published this information. They have cited it. I have no doubts that it exists somewhere. The ATSB can show us only that information, which fits their narrative, and hide everything else. What to show, and what to hide is decided by one person. Unfortunately ATBS’s chief has more rights in Australia than a Federal Court. Above the law.

    Re: “I regarded the notion of a geographic dependence raised by the DSTG to be in error. It is really why I asked Holland for BFO error plots from other flights.”

    Take a look at Fig. 5.4. The “geographic dependence” is obvious. I know nothing about oscillator drift, but I can hardly imagine how it can explain such an average trend. It rather resembles Schuler oscillations (due to the Schuler period, which is 84.4 minutes). Also, it is well known that ADIRU’s heading and position slowly deviate from the actual ones due to the measurement and integration errors over the time.

  166. DennisW says:

    @Oleksandr

    I recall you bringing up the Schuler period previously. IMO, that was a very insightful thing to do. My belief, however, is that the ADIRU is updated by GPS, and that both the Schuler effects and drift are removed. I would guess the position, speed, and track of the aircraft are known with great precision.

    Again, if we had data from other flights we could home in on the answer.

  167. Oleksandr says:

    @Dennis,

    I think it was Gysbreght, who first associated Schuler’s period with DSTG’s Fig. 5.4.

    Re: “My belief, however, is that the ADIRU is updated by GPS, and that both the Schuler effects and drift are removed”

    If I recall correctly, ADIRU’s position is periodically updated by GPS, but not heading.

    Re: “track of the aircraft are known with great precision”
    Yes and no. There is a complex interplay between ADIRU, SAARU, GPS and FMC. There is always some discrepancy in the data provided by these units, and it is up to a “user system” to decide how to use this information. In my understanding the AES primarily takes data from the ADIRU via ARINC bus, but not from GPS.

  168. ALSM says:

    I am not sure how Honeywell did it at the time 9M-MRO was manufactured, but virtually all modern air nav systems use a Kalman filter to combine air data, 3 axis rate sensor data, 3 axis magnetometer data, and 3 axis GPS data…each with appropriate errors, weighting functions and time constants, etc. The end result is similar to “GPS updating the INS” except that the Kalman Filter produces a continuously updated estimate with no discrete “updates”. If the GPS or INS data is lost (usually momentarily when it does happen), the Filter coasts through the outage with no discrete changes.

  169. DennisW says:

    Thx, Mike.

    An interesting sidebar is the GPS Car Navigation systems we sold to BMW, Hyundai, Porsche,…

    With a Garmin in your car you are still searching for satellites when you exit a multi-story garage. With the factory installed systems you can spiral down the garage access ramp and your car is perfectly positioned at the street exit ready to take you to your selected destination. Similar to the modern air nav systems. The Kalman filter in the car case takes GPS, a single axis rate gyro, and odometry to produce a position. Of course, the system does the same thing on the way in to your parking space, and remembers your position and orientation when you shut the car off. Huge difference especially in urban canyons and the like. Of course, the price difference is not trivial.

    In good GPS tracking situations the rate gyro and odometer are calibrated by GPS, and are much more accurate than they would be standing alone.

    Disclaimer: I am the registered owner of a significant stock position in a company which provides OEM car navigation systems.

  170. DrB says:

    @VictorI,

    You said: “If the ATSB truly believes the plane will be found near 35S, but further out from the 7th arc, it needs to explain why the DSTG results are wrong, and why the BFO analysis is wrong.”

    On the first point, Oleksandr has already mentioned the 0.73M lower speed limit, and I would add to that the lack of any consideration of available fuel. The problem is not that the DSTG did not predict any possible crash locations N of 36S, it’s just that those locations were overwhelmed in the probability assessment by the large number of paths at 36-38S. These southern end points are mostly not reachable with the available fuel. In a way, the DSTG results are not “wrong” per se, but they failed in their objective because of faulty assumptions.

    On the second point (why “the BFO analysis is wrong”), I’m not sure that statement can be proved at this point in time. Is it impossible for a B777 to experience 4500 fpm ROD and then 15,000 fpm ROD 8 seconds later and still hit the ocean more than 20 NM away? I don’t know whether it is possible or not, but I have not so far seen a clear-cut case made for it being impossible. It’s too bad we don’t have more information about the end-of-flight simulator runs. Perhaps Gysbreght has an opinion on this point he would share. Personally, I don’t know of any possible explanation for the BFOs at 00:19 other than a rapid descent. Could there have been some bad NAV data from the ADIRU to the SDU due to the power interruption at second-engine flame-out, and could this cause the observed BFOs?

  171. DrB says:

    @Gysbreght,
    @Oleksandr,

    Way back in November 2015 I published a table of the MH370 Log-on pairs of BFOs. It is Table 1 on page 11 in my Addendum 5. I’ll repeat the basic information here. The change in the second (ACKNOWLEDGE) BFO from the first (REQUEST) BFO is 0 Hz at 16:00, +136 Hz at 18:25, and -184 Hz at 00:19. Remember that the power had been on the SDU for several hours prior to the 16:00 re-boot at the gate. The power interruption then was very brief, and the thermal warm-up transient in the OXCO was essentially nil. This is also true for the brief interruption (~1 minute) at the 00:19 re-boot. As I showed in my OXCO model paper, the effect on the frequency at 00:19 is very small (a few Hz). Thus the 16:00 and the 19:00 log-ons are not contaminated significantly by the OXCO warm-up transient. However, the 18:25 log-on BFOs are very much affected. After compensation by the OXCO warm-up model, they do not indicate any maneuver or climb or descent beyond the 15 NM lateral offset apparent in the BTO data alone. We have also learned that this maneuver is called for by ICAO regulations in this situation.

    I will also comment on the BTOs observed during these three log-on events. At 00:16 the BTOs were 14,780 and 14,820. At 18:25 they were 12,520 and 12,600. At 00:19 the BTOs were 18,400 & 18,380. The second and third log-ons were in flight, and these required applying corrections of multiples of 7820 microseconds to the ACKNOWLEDGE BTOs only. After this correction, the agreement of the REQUEST and the ACKNOWLEDGE BTOs for all three log-ons is satisfactory. I don’t think there is anything to be inferred from the BTO timing cycle corrections that would reduce confidence in the BFOs. Two of the three MH370 log-ons are unique, as far as I can tell, among the data from prior flights. I doubt there was a single case of in-flight re-boot and log-on, as this is not supposed to happen in normal operations. That is apparent in Holland’s paper, where all the prior log-ons he analyzed were on the ground.

    To summarize, all the BTO and BFO data at 18:25 are consistent with an SDU OXCO transient due to an extended period in an unpowered state plus a lateral offset maneuver occurring between 18:22 and 18:26. At 00:19 the OXCO transient is insignificant, and another explanation is required for the BFOs then – either a rapid descent or possibly some corruption of navigation data fed by the ADIRU to the SDU (or both).

  172. ALSM says:

    @DrB
    @Gysbreght
    @Oleksandr:

    It should also be noted that the BFO and BTO values are generated by independent means and measured at the GES by independent means. IOW…There is nothing related to the generation and measurement of a BTO value that is affected by the carrier frequency, and there is nothing related to the generation and measurement of a BFO value that is affected by the round trip signal propagation time. These are independent measurements, so there is no reason to believe that if either value appears to be questionable for any reason then the other should also be questionable.

  173. Victor Iannello says:

    @DrB: I wasn’t really asking for reasons that the DSTG’s Bayesian and BFO analyses were wrong. I’ve been questioning their assumptions about speed range and the number and type of manoeuvers since the report was released in December 2015. Rather, I am curious whether the ATSB is now questioning those same assumptions.

  174. Victor Iannello says:

    For those interested in the concerns raised by some of us after the DSTG report was released, look at this paper by Richard Godfrey and me that was published on Duncan Steel’s website in Dec 2015 that discusses the bias towards straight, high-speed flights.

  175. Peter Norton says:


    > Victor Iannello: “Repeatedly citing the doubtful claims of Emil Enchev
    > hurts their case. In the same release, they [MH370 Chinese Families]
    > question the satellite data and then ask for the search to continue
    > along the 7th arc.”

    I agree with you on the content.
    And you certainly know more about them than I do, but rather than acting in bad faith, I think they are just poor lost souls desperately clinging to anything seemingly promising any glimmer of hope.

    As to the apparent contradiction you are pointing out:
    Hope and despair are not always bound by pure logic.

    However, if they question the satellite data, I can indeed see some logic in them asking for the search to continue along the 7th arc in order to disprove the theories based on the satellite data.

    (… or else find the plane there, which would be a – albeit lesser – desired outcome, too.)

  176. Perfect Storm says:

    Victor Iannello: “look at this paper by Richard Godfrey and me that was published on Duncan Steel’s website in Dec 2015”

    Has this been deleted from Duncan’s website ?
    And if so, why ?

  177. Victor Iannello says:

    @Perfect Storm: I don’t know, but it’s certainly his right to run his site as he pleases.

  178. Brock McEwen says:

    I only had time to skim the latest official drift work tonight, so just quick, early observations:

    1) The conclusions of the latest drift paper seem to clash violently with two key aspects of their own results:

    A) Presence of debris on Oz shores, and
    B) Absence of “low-windage” debris on SA shores after 22 months

    Both are predicted – by their own model, and by many others – if impact occurred at 35S.

    Neither were observed.

    (I hope I’m wrong, and that these clashes are due to my eyes reading faster than my brain could process. Corrections/confirmations welcomed.)

    2) If empirical testing has merely validated CSIRO’s original drift model – and the size of recovered debris has merely corroborated the ATSB’s original theory (hypoxia–>flameout at altitude–>high energy impact), then we need to find out the precise date the ATSB/SSWG/JIT/JACC first set eyes on Figure 4.2 (or its predecessor).

    As discussed before in this forum, the mid-most third of the wide search area – which Fig. 4.2 claims is the worst part to search, because [aerial search] – was from June to Oct of 2014 the section the ATSB claimed was the best part to search. Did search leadership have a version of Fig. 4.2 in hand when writing their 2014 reports? If so, why did it not even rate a mention? And if not, why not?

  179. Gysbreght says:

    @drB:

    ” Is it impossible for a B777 to experience 4500 fpm ROD and then 15,000 fpm ROD 8 seconds later and still hit the ocean more than 20 NM away?”

    It is entirely possible for MH 370 to experience 4500 fpm ROD at the time the log-on request was transmitted which defines the 7th arc. I claim that it is impossible without pilot imputs for the airplane to reach 15,000 fpm ROD 8 seconds after the 7th arc. In all simulations that occurred much later, seconds before crashing into the sea, if it occurred at all.

  180. David says:

    @DrB. “Is it impossible for a B777 to experience 4500 fpm ROD and then 15,000 fpm ROD 8 seconds later and still hit the ocean more than 20 NM away?”

    Besides Gysbreght’s, you have also the ATSB opinion, from Boeing simulations,”Results from recent simulations showed high rates of descent broadly consistent with the BFO analysis. These simulations indicated that the aircraft was likely to be within 15 NM of the 7th arc.”

    Page 17, https://www.atsb.gov.au/media/5772107/ae2014054_final-first-principles-report.pdf

    I find it difficult to believe that their results are unsound though I can understand that Gysbreght would information all the same.

    @Rob. Thanks for your advice

  181. Richard Cole says:

    @Brock

    There are a number of new links at David Griffin’s web site.

    http://www.marine.csiro.au/~griffin/MH370/

    This includes a presentation at CSIRO in the last few days which contained a number of drift model outputs, as a function of position on the arc. The sequence illustrates Griffin’s point there is a ‘sweet-spot’ at 35S where little debris moves east to Australia. I have posted those pages here:

    https://www.dropbox.com/s/39alkjiny5nd53c/Griffin_MH370_CSIRO_2017-extract.pdf?dl=0

    On figure 4.2, this uses the current Griffin drift model to relocate the areas aerially searched in 2014 back to their positions on April 7th. Search authorities could not have had that particular information in 2014. In any case we have long known that the early search was split between the south, based on a simple flight path model, and an area further north. It is not surprising that there were gaps.

  182. Gysbreght says:

    In the simulations the airplane stayed within +/- 15 NM of the 7th arc because it turned. “In some instances, the aircraft remained airborne approximately 20 minutes after the second engine flameout”. If the autopilot was lost when the first engine flamed out it could even have flownn up to 15 minute longer. Victor Iannello’s simulations showed that if the airplane maintains wings level it it can fly 100 NM beyond the 7th arc. An ape can be trained to keep the wings level.

    The ATSB is witholding the conditions that were simulated and quotes the results selectively in a manner designed to mislead the superficial reader. Their failure to respond to my query tells me that my suspicion is correct.

  183. David says:

    @Gysbreght. I think in those instances the descents were inconsistent with the BFOs. I raised earlier the notion that meeting the BFOs is a qualifier.

    @DrB. More specifically on whether it is possible that they could “hit the ocean more than 20 NM away”, I believe they nominated an average 25 miles width (with a weighting for arc altitude) because, “The simulated scenarios do not represent all possible scenarios, nor do they represent the exact response of the accident aircraft”.(Search and Debris update p12).

    If that is so the answer is yes if you accept what they say, they having access to the simulator results and being informed by Boeing opinion.

  184. Gysbreght says:

    @David: In all simulations the radius of turn where the tracks cross the 7th arc is quite large. Can you think of a reason that would make the airplane pitch down violently while still at low bank angle?

  185. David says:

    @Gysbreght. Yes. Low speed, pitch down, no AC no autopilot.

  186. Gysbreght says:

    @David: You’re not giving a reason for the pitch down that increases ROD by 10,000 fpm in 8 seconds. You’re not giving a reason either why the ATSB would decline to provide the time between the 7th arc and 15,000 fpm ROC.

  187. David says:

    @Gysbreght. To amplify, we went into this a couple of weeks ago, when I described what might be the pitch down reaction when the left engine finally failed at low speed, the stabiliser and elevator holding the nose up at that point, high AoA.

    Whether or not, there were simulator runs which were consistent with the BFOs. While you do not have the information you would like I find that hard to argue with.

  188. David says:

    @ Gysbreght. By ‘finally failed’ I mean autopilot disengagement on loss of AC power was lost as the engine ran down below idle

  189. Victor Iannello says:

    @Richard Cole: Thank you for the links, and welcome to the discussion.

  190. Don Thompson says:

    @Oleksandr wrote:

    There is a complex interplay between ADIRU, SAARU, GPS and FMC. There is always some discrepancy in the data provided by these units, and it is up to a “user system” to decide how to use this information. In my understanding the AES primarily takes data from the ADIRU via ARINC bus, but not from GPS.

    The AES is delivered data that is sourced from the ADIRU via the AIMS/Data Conversion Gateway Function.

    The FMC, the navigation function, may be commanded to derive its positional reference from a range of sources (ADIRU, GPS, radios) but the avionics ‘truth’ is the ADIRU.

    When the ADIRU is initialised, on the ground, GPS position sourced via the FMC is used for validation of the entered position. Subsequent to initialisation, GPS position and velocity is used in the ADIRU sensor calibration logic, which is used to calibration the gyros and
    accelerometers. The calibration does not correct the ADIRU errors, it only reduces ADIRU drift errors during flight (source: B777 AMM).

  191. Gysbreght says:

    @David: “the stabiliser and elevator holding the nose up at that point, high AoA.”

    All the simulator runs, Boeing, ALSM, Victor Iannello and others show that it doesn’t happen the way you suppose.

  192. Gysbreght says:

    @David: Since you have a better rapport with the ATSB than I have, why don’t you ask them to confirm your understanding that the BFO criteria were met within 8 seconds from the 7th arc in those five recent Boeing simulations?

  193. Gysbreght says:

    @David: I forgot to mention Andrew in the list of simulations.

  194. Gysbreght says:

    @David: “we went into this a couple of weeks ago, when I described what might be the pitch down reaction when the left engine finally failed at low speed, the stabiliser and elevator holding the nose up at that point, high AoA.”

    While partly correct, let’s look at it quantitatively:

    The stabilizer does not trim below the minimum maneuver speed of 205 kt IAS at the ZFW (approximately the top of the amber band on Andrew’s PFD). The autopilot can command the elevator to hold the nose up down to a speed just above stickshaker speed, which was 170 kt IAS in Andrew’s simulation. When the autopilot quits, the elevator is released, and the airplane pitches down to the angle of attack that corresponds to the trim speed of 205 kt IAS. At that AoA and a speed of 170 kt IAS, the lift is reduced to about (170/205)^2 of that required for level flight, i.e. the normal load factor reduces to 0.688 g, and the airplane accelerates downwards at the rate of 1-0.688 = 0.312 g. As the airspeed increases, the lift increases and the vertical acceleration reduces (in absolute value).

    Increasing the rate of descent by 10,000 fpm in 8 seconds represents a vertical acceleration of 0.648 g.

  195. Ge Rijn says:

    On the sequence of CSIRO-drift graphics link @Richard Cole posted I wonder why only the 35S segment is so out of line.
    The graphics bordering south and north of this segment show a very different pattern and lots of landings on WA (which is also contradicting some previous studies concerning the segments north of 36S).

    What makes this 35S segment so special within only ~1 degree latitude?
    What is happening only there that makes a dramatic change in the currents and results within only ~1 degree latitude?
    It seems a freak of nature to me and I would therefore highly value a specific explanation from CSIRO/Griffin on this.

    Further the 31S till 29.5S segments actualy show a better picture than 35S. All three have 2 possible landings on WA. With the 35S and the 30.25S/29.5S segments are the same with 2 landings in the Shark Bay area.

    But the distribution and coverage of all found debris from Tanzania till SA is a lot better within the 31S till 29.5S than with the 35S segment in this graphic sequence.

    I have the opinion now this CSIRO-35S segment is way to much out of line within their own sequence and needs specific explanation.

  196. Victor Iannello says:

    @Gysbreght & @David:

    I think that @Gysbreght makes a good point. A mismatch between the airspeed and the trim speed doesn’t seem sufficient to cause the vertical acceleration (down) that is suggested by the BFO values.

    Here is a list of possible causes of the accelerated descent:
    1. A steep banked turn.
    2. Pilot input.
    3. System malfunction.

    Are there others? In particular, is there a system configuration that could cause this without the need for a bank or for pilot input? I can’t think of any.

  197. Gysbreght says:

    @Victor Iannello: I agree with your 1, 2 and 3.

    “A mismatch between the airspeed and the trim speed doesn’t seem sufficient “

    Also that mismatch only exists at second engine flameout, and disappears when the airspeed is back to the trim speed. The 7th arc is two minutes after the loss of engine power.

  198. TBill says:

    @Victor
    I was just checking in to post a list too

    Outline for Possible End of Flight BFO Explanations:

    A. “Natural” trajectory for unpiloted B777 aircraft
    B. Hardware or Systems data errors related unusual conditions
    1. On my list is depressurization or repressurization
    C. Intentional piloted trajectory
    1. Pilot arbitrary choice to dive
    2. Attempt to hit a specific geographic/underwater target
    3. Attempt to engineer a particular crash mode (eg; to hide debris)
    4. Attempt to generate power (RAT/engine spools)

  199. Richard Cole says:

    @Ge Rijn

    The difference in the model output at 35S is discussed in section 3.3 of Griffin, Oke and Jones “The search for MH370 and ocean surface drift“ (Dec 2016) http://www.marine.csiro.au/~griffin/MH370/MH370_Ocean_Driftv2.9.pdf

    They say there:
    “The satellite altimeter data tells us, via the model, that there were alternating bands of eastward and westward flow, including two prominent ~150km-wide bands of westward flowing surface current crossing the 7th arc in March 2014 – one near 35°S and one near 30°S (Fig. 3.3.1). The current near 35°S carried floating items about 500km west before turning north (Fig. 3.3.2).“

    So debris at 35S is carried away from Australia. This measured feature of the ocean flow and the lack of debris on the Australian coast is the core of his argument. Clearly, for small number of pieces there is a randomness in how many reachthe shore and then in their chances of being found, but there were not a large number of pieces on the shore.

  200. TBill says:

    Maybe
    D. Physical damage to aircraft (stress damage/hull etc)

  201. Ge Rijn says:

    A descent of 15000ft in 8 seconds to ~4000ft (if this is what happened according to the final BFO’s) would not have resulted in a high speed dive impact. In one way or another the plane was level and relatively slow on impact.
    Pilot controlled or not.
    The overwhelming kind of trailing edge, wing related debris found, their locations, their damage.
    It’s just clear as a whistle to me this must have been a low AoA, relatively slow speed ~level impact (IMO controlled).

    IMO the denial of this obvious information is hampering progress.
    Leading to endless discussions about end of flight scenarios we cann’t settle with the Inmarsat data or simulation data at hand.

    IMO the debris tells a very clear story.

  202. Ge Rijn says:

    @Richard Cole

    Thanks. I did not read that explanation before.
    It seems it was a temporaly shift of current in those areas in a quite narrow band. What caused this? A river of 150km wide going to the west between two rivers going to the east?

  203. Oleksandr says:

    @Bobby,

    Re: “I will also comment on the BTOs observed during these three log-on events. At 00:16 the BTOs were 14,780 and 14,820. At 18:25 they were 12,520 and 12,600. At 00:19 the BTOs were 18,400 & 18,380.”

    I meant BTO of 51700 microseconds paired with the BFO of 273 Hz, and BTO of 49660 microseconds paired with the BFO of -2 Hz.

    How does you theory explain these BTOs?

    ————–

    @ALSM,

    Re: “It should also be noted that the BFO and BTO values are generated by independent means and measured at the GES by independent means. IOW…There is nothing related to the generation and measurement of a BTO value that is affected by the carrier frequency, and there is nothing related to the generation and measurement of a BFO value that is affected by the round trip signal propagation time. These are independent measurements, so there is no reason to believe that if either value appears to be questionable for any reason then the other should also be questionable.”

    Which one is questionable:
    1. The BFO of 273 Hz or BTO of 51700 microseconds?
    2. The BFO of -2 Hz or BTO of 49660 microseconds?

    How do you explain that the erroneous BFO occurred simultaneously with the erroneous BTO at least in one of the instances, if these are completely independent, as you stated? And how do you explain that no other BFO or BTO among the whole Inmarsat log is abnormal?

  204. ROB says:

    @Ge Rijn

    The Malaysian Debris Examination Report is quite revealing, in my opinion because the inference is the aircraft must have hit the water in a very flat attitude (possibly right wing down) with a relatively low forward velocity, ie as if it were in a stall. Items 9 and 15 are of particular interest: Item 9 from the LHS, item 15 from the right. They are in a basically similar condition and must have been separated from the top surface of the wing by a similar mechanism, probably when their corresponding flaperons were forced upward on impact with the water. The RH flaperon ended up on Reunion. So where is the other flaperon? Was it so badly damaged that it sank, is it still afloat somewhere in the ocean or was it only partially dislodged, and remained attached to the wing. The fact that both the section of RH outboard flap and the RH flaperon made landfall on the other side of the SIO would suggest that if the LH flaperon had ended up floating in the ocean, it too would have been recovered before now. This prompts another question: if the aircraft hit the water in a stalled condition, was it pilot induced? I think it’s more likely that it was.

  205. DrB says:

    @Till,

    You said: “Re: CSIRO
    Of course 35-S is pretty much right on DrB’s proposed path end point, I am thinking.”

    If one assumes that the average 9M-MRO PDA is the 1.5% used in the Flight Brief rather than the much higher number derived from ATSB Fuel Flow sensor readings, then the Constant True Heading (CTH) path to 35S has two issues:

    1. There is too much fuel for exhaustion to occur at 00:17 unless there is a source of extra drag or fuel consumption beyond normal flight. I don’t know what that could be. It seems it cannot be the APU running constantly after 18:25, or the re-boot at 00:19 would not have occurred.

    2. The means to achieve CTH cannot be an aborted hold (based on Andrew’s simulator test) unless one believes the pilot set the NORM/TRUE switch to TRUE before 18:40. This seems very unlikely to me. Alternatively, the pilot could manually set a CTH using the MCP, but again, the pilot would need to change the NORM/TRUE switch.

    There is an alternative scenario that ends up in the same neighborhood. It uses Magnetic Track. I have noticed for a long time now that CMT and CTH routes have nearly equal curvature. Once I finish my Fuel Model refinements, I will go back and investigate CMT routes. Fuel is still an issue, but this can perhaps be resolved by flying at a reasonably high speed but at a low altitude. One should ask, why would a pilot set a CMT? The only possibility that appeals somewhat to me is if it were exactly 180.0 degrees CMT. Perhaps a pilot might choose to do that just to go South if the circumstances were dire.

  206. Oleksandr says:

    One interesting aspect of CSIRO’s drift model is that it also predicts considerable concentration of debris at the eastern shore of Madagascar, similarly to my models. So, what is the ATSB waiting for in terms of the debris search and recovery if they are confident in CSIRO’s results?

  207. Oleksandr says:

    @Don Thompson,

    You forgot to mention that ADIRU receives SAARU data via one of the three ARINC channels. I don’t know if this data is used for comparative purposes only, or it affects ADIRU output. It must affect somehow (for example, by flagging the discrepancy), because if it does not affect output at all, there is no point to compare.

    Re: “When the ADIRU is initialised, on the ground, GPS position sourced via the FMC is used for validation of the entered position.”

    What is the ADIRU output when the ADIRU is being initialized / aligned (i.e. in the process of initialization)? It may take up to 15 minutes, so it is hard to believe that SATCOM is not functional for 15 minutes on the ground only because of the ADIRU.

  208. DrB says:

    @Oleksandr,

    You said: “I meant BTO of 51700 microseconds paired with the BFO of 273 Hz, and BTO of 49660 microseconds paired with the BFO of -2 Hz.

    How does you theory explain these BTOs?”

    It’s not my theory. It’s ATSB’s published explanation of multiples of 7820 microseconds being included in the ACKNOWLEDGE message timings for the two in-flight log-ons.

    A good question to be asked is why this same type of correction was not needed for the ground-based log-on at the gate at 16:00. I suspect there is some difference in the SDU software depending on whether the aircraft is stationary on the ground or is in flight. To my knowledge this topic has not been discussed relative to MH370.

    The reason why unusual BTO and BFO values both occurred at 18:25:34 is that this is for a log-on Acknowledge message in flight after an extended period of power outage to the SDU. This circumstance only occurred once in all the 9M-MRO data (including the MH370 flight and the prior flights). The reason for an unusual BFO at 00:19:37 is different; it can’t be due to a thermal OXCO transient because the power-off time was quite short.

  209. Oleksandr says:

    @ALSM,

    This is in addition to my previous post. You wrote:

    “There is nothing related to the generation and measurement of a BTO value that is affected by the carrier frequency, and there is nothing related to the generation and measurement of a BFO value that is affected by the round trip signal propagation time”

    I think you interchanged BTO and BFO in this statement, but apart from this, what is the probability that the BTO value becomes corrupted exactly at that time, when the process of oven warm up affects BFO? And this occurred not once, but twice. And this did not occur when the plane was on the ground. Why? And no other BTOs were as long as these two samples. Why?

    Sorry, I do not believe in such coincidences.

    Re: “The end result is similar to “GPS updating the INS” except that the Kalman Filter produces a continuously updated estimate with no discrete “updates”.”

    What is your explanation of the “BFO drift” effect in Fig.5.4 of DSTG’s report if you believe that navigational data fed into AES are always accurate?

  210. Oleksandr says:

    @Bobby,

    To my knowledge the ATSB has never explained a basis for the delay. It is understandable that these delays can be discrete, but what did cause the delay?

    Also, why does the ATSB suggested a different offset for 18:25:27, which was 4600 but not 7820, also R-Channel? In general, all the first BTOs in logon sequences were subjected to the additional delay of 4600 microseconds, including 18:25:27. Why would the second BTO in 18:25 sequence be affected by an extended period of the power outage to the SDU, but not the first one?

  211. Oleksandr says:

    @Bobby,

    In addition to the previous:
    You stated “reason for an unusual BFO at 00:19:37 is different”

    How do we know this? Sorry, I think it is just a circular logic. To justify a rapid descent, the ATSB needs everyone to believe that -2 Hz is valid, so that the ATSB invented the hypothesis that the reasons for the anomalous BFO and BFO in the logon sequences are different. They faces a number of problems:
    (1) they need to discard original Inmarsat’s paper, which suggested these numbers unreliable;
    (2) 18:25:27 BFO and BFO do not fit this theory, so they (e.g. Holland) need to discard this sample;
    (3) there is no single multiple factor for BTO, so they need to invent several factors (i.e. 4600 and 7820 microseconds);
    (4) they say that over the period of 8 seconds the plane experienced 0.68g acceleration, but before that the plane was descending at less than 3,000 fpm (ref to Holland). To derive these numbers they need to make a hidden assumption with regard to the location and heading before the engine flameout. Not talking about such a high acceleration.

    Now it is suggested that -2 Hz is valid, because the there is no other reason for an unusual BFO at 00:19:37 except a rapid descent. Nice bazaar.

  212. DrB says:

    @Oleksandr,

    You said: “Why would the second BTO in 18:25 sequence be affected by an extended period of the power outage to the SDU, but not the first one?”

    The BTOs are not affected by the length of power outage, only the BFOs are affected due to the OXCO thermal transient.

    You also said: “To my knowledge the ATSB has never explained a basis for the delay. It is understandable that these delays can be discrete, but what did cause the delay?”

    Here is the quote from the ATSB’s report on 3 December 2015 on page 20:

    “The communications corresponding to the burst timing offsets (BTO) used to determine the location arcs were transmitted on 2 communications channels; the R1200 and the R600 channels.

    The R1200 log-on acknowledgement BTO measurements that occurred at 18:25:34 and 00:19:37 were initially ignored due to an anomaly in the values. Instead, the R600 BTOs at 18:25:27 and at 00:19:29 were used to create the 1st and 7th arcs.

    Because the R1200 BTO measurements were assessed as having a lower standard deviation, identifying the cause of the anomaly, in order to possibly correct it, would improve the accuracy of the model.

    An analysis of similar transmissions from the previous flight was conducted which identified a variable offset coefficient with a factor of approximately 7820 μs (i.e. N*7820).

    The manufacturers of the SDU and the ground station equipment were unable to determine a specific reason for the offset, however they did note that the reference clock used for the time slots was 128 Hz, which equates to 7812.5 μs.”

    This is the means I used to derive the corrected BTO values I listed in my previous post (subtracting multiples of 7820 microseconds). Note that this occurs on the R1200 channel only, not on the R600 channel. For the two in-flight log-on REQUEST messages on the R600 channel, subtracting 4600 microseconds was recommended by ATSB. Thus no corrections were applied to the 16:00 log-on on the ground to either the R600 (REQUEST) or R1200 (ACKNOWLEDGE) channels. In some way the SDU in this regard behaves differently on the ground than in flight. Perhaps Don T knows why.

  213. DennisW says:

    @Oleksandr

    “What is your explanation of the “BFO drift” effect in Fig.5.4 of DSTG’s report if you believe that navigational data fed into AES are always accurate?”

    There is no reason for the GPS in the aircraft to provide a frequency reference. It could be done quite easily, but there is no motivation to do so. The aircraft Doppler correction on the ISAT data is good enough with the oven oscillator accuracy in the AES. The nav data provided to the AES need to be accurate in order to compute the AES downlink frequency needed to compensate for aircraft Doppler relative to the satellite.

  214. ALSM says:

    Interview with David Griffin and comments on the recent CSIRO Update

    https://blogs.csiro.au/ecos/mh370/#comment-14886

  215. TBill says:

    @DrB
    The other possibility is to change the scenario a bit, right? Let’s say add’l time and distance at full speed after 1840 cutting back to slower speed at Arc2 or maybe even after Arc2. Right now I am personally holding Arc3 to Arc5 as the steadiest period.

  216. ALSM says:

    Oleksandr:

    I did not “… interchanged BTO and BFO in this statement…”. The statement was as I intended. The bottom line is that BFO and BTO processes and values are independent and there were no erroneous values of BFO or BTO circa 00:19 as you assume. They all make perfect sense to me (and many others), and they are consistent with other independent observations and data as discussed ad nauseum.

  217. Brock McEwen says:

    @Richard C. Thanks much for the response, and the link.

    An analysis culminating in Fig. 4.2 had to have been performed by June, 2014; it is an attempt to answer a critical and intuitive question: “how can the empty aerial search help inform seabed search strategy?” That’s why I said “or its predecessor” – I’m not talking about the latest iteration, but the first. What did that first analysis say, and who saw what when?

    I find the graphs underwhelming in support of s35. Not only is this “sweet spot” something we haven’t seen from any other drift expert, it appears to pertain only to debris with flaperon-esque leeway profiles. For zero debris on Oz shores to be plausible, a large array of debris – with a rich mix of plausible windage vectors – must be modeled, and this “hole in the doughnut” at 35s must be evident across all vectors.

  218. Oleksandr says:

    @Dennis,

    Re: “The nav data provided to the AES need to be accurate in order to compute the AES downlink frequency needed to compensate for aircraft Doppler relative to the satellite.”

    Would accuracy of navigation data resulting in 20 Hz BFO error due to the compensation term be acceptable?

  219. Oleksandr says:

    @ALSM,

    “there were no erroneous values of BFO or BTO circa 00:19 as you assume.”

    Where does it come from? You are repeating the same mistake again, which previously led you to 38S. You again use circular logic to make conclusions based on assumptions, and then justify assumptions by conclusions.

  220. Don Thompson says:

    @Oleksandr

    My bad, you gave me an incorrect impression for your comprehension of the ADIRS functioning.

    SAARU monitors the ADIRU, it tracks the ADIRU’s data as transmitted on the L & R A629 Flt Ctrl busses. When the ADIRU data is good, the SAARU simply passes that data on, along the C A629 Flt Ctrl bus. Should the SAARU detect ‘bad’ data from the ADIRU it transmits its own processed data on the C A629 bus.

    If the AES receives no valid ‘IRS’ data, it simply operates via the LGA signal path for which it needs no ‘IRS’ data. No beam steering required and as the aircraft is on the ground, doppler correction is moot.

  221. Oleksandr says:

    From DSTG book:

    “For some communication messages, typically during initial log-on, there was a very large difference between the measured BTO and the nominal delay. Analysis showed that rather than simple outliers, these
    anomalous BTO measurements could be corrected by a factor of N × 7,820 μs where N is a positive integer. The origin of these anomalous BTO measurements has not been fully determined…”

    What does ‘typically’ mean?

  222. Oleksandr says:

    @Don,

    “If the AES receives no valid ‘IRS’ data, it simply operates via the LGA signal path for which it needs no ‘IRS’ data. No beam steering required and as the aircraft is on the ground, doppler correction is moot.”

    So the Doppler correction is essentially assumed to be zero, correct?
    And what if the aircraft is in the air?

  223. Don Thompson says:

    Oleksandr asked, “Where does it come from”

    Powerpoint, that’s where it comes from. Communicating for too long in Powerpoint-esque soundbites has made a lot of people incapable of describing a concept concisely and coherently.

    Terms used, such as ‘anomalous’, were ill-considered. They weren’t anomalous in a functional sense, I suggest they were anomalous in the sense of recorded experience.

    The GES channel unit demodulators measure BFO: simple, they measure what they receive. Each demod is fixed to a specific channel, it’s fixed to demodulate a specific carrier’s data and extract certain management metadata as it performs that function.

    The burst structure for each return channel type is different, the point at which the timing/sync ‘mark’ occurs is unique to each channel type. Hence, the differing ‘bias’ for BTO.

    :Don

  224. Don Thompson says:

    Oleksandr asked “And what if the aircraft is in the air?”

    If the aircraft is in the air then the ADIRU has long been initialised & is operational, the AES operates using the HGA signal path.

    If the ADIRU fails in-flight, things aren’t good, satcom is a low priority issue (but the crew has HFVOICE as an alternative to SATVOICE).

    After circa 15:55UTC, on the ground at KLIA, all satcom operations used the HGA. There would be no R/1200 (or T/1200 or C/21000) channel utilisation via the LGA signal path.

    :Don

  225. DrB says:

    @all,

    I am trying to understand the Checkpoint fuel tables in the December 2007 FCOM (pages PI.21.4-6) for the Trent engines. The tables give the fuel required (at LRC) and the time for various distances to be traveled at each specified altitude. I think the purpose of these tables is to verify if adequate fuel is onboard to reach a destination at that distance, altitude, and weight. My question is, do the fuel and time tables include allowances for descent and landing?

    Based on the variation in the average speeds I have calculated for each adjacent pair of distances, it seems like the descent and landing fuel and time are included. I also see that there is a fuel offset at zero distance of -1.5 MT (at FL370). Perhaps this is the fuel needed for descent and landing.

    I was attempting to derive the LRC fuel flows at various altitudes, especially those between FL250 and FL100, which are included in the Checkpoint tables but which are not included in the LRC table on page PI.21.3. Does anyone know how to do this in such a way as to get the level-flight fuel flow not including the descent and landing?

  226. DrB says:

    @Don Thompson,

    The R1200 log-on acknowledge message at the gate at 16:00:13.4 is peculiar in several ways compared to the same (log-on acknowledge)messages at 18:25:34 and at 00:19:37. Do you understand why:

    1. The BTO then does not contain multiple increments of 7820 microseconds.
    2. It occurs 17 seconds after the log-on request message (not 7-8 seconds).
    3. There is a second message with the same function in the revised data log at 16:00:13.9 in the P-channel.

  227. DennisW says:

    @oleksandr

    “Would accuracy of navigation data resulting in 20 Hz BFO error due to the compensation term be acceptable?”

    The navigation data does not depend in any way on the BFO error. The ISAT system is a communication system not a navigation system. The AES oscillator is free running and not corrected by anything.

  228. Brock McEwen says:

    @Don: in 2014, a Metabunk contributor named “guardeddon” replied to S Gunson’s post re: Malaysian radar, linking readers to a detailed compilation of published open-source data on equipment, deployments, and nominal ranges. Were you the “guardeddon” in that convo? If so:

    1) FYI: I think I found a minor typo in your location of sqdrn 322. Though your paper gave its longitude as E[x]°51′, I infer from GE and context that you would have meant E[x]°0’51”. I infer this primarily from the fact GE shows me nothing at the former, and what appears to be a radar station within 4″ of the latter. I seriously doubt the difference is material to anyone’s analysis of anything; I mention it only because I know we share a fondness for precision, and because the error did seem to make it into the table Duncan published.

    2) Did you ever resolve the question with S. Gunson re: specific hardware at each of Butterworth, 321 (hill behind Kota Bharu) and 322 (hill behind Kuantan)? He thought they were all GM400’s; you thought they were all something less powerful.

    3) Part of your argument was that the (lone, in your view) GM400’s deployment was reportedly at another location – but when I followed your supporting links, one was defunct, and the other didn’t seem to mention GM400 specifically. Can you help me with the dead link in particular?

    3) In that exchange, you said you’d collected similar data for Thailand and Indonesia, as well, and were going to add it to the report. Did you ever publish an expanded version of this paper? (Other than the tabular summary Duncan posted, I mean.)

    4) Did you collect similar info for any other nation?

    5) I offer belated but profuse thanks for collecting and collating all this data – it is much appreciated. I usually begin with my thanks, then launch into my questions; I fear this risks minimizing the appreciation I feel for fellow fact-collectors.

    @All: apologies for the lack of equipment/location names in the above. I’m not trying to be cryptic – just careful about repeating anything sensitive that I don’t yet know for a fact is already in the public domain.

  229. Perfect Storm says:

    from https://blogs.crikey.com.au/planetalking/2017/04/21/revised-mh370-wreckage-drift-analysis-adds-pressure-renewed-search/

    « Our earlier field testing of replicas of the flaperon was unable to confirm numerical predictions by the Direction Generale de L’Armement (DGA) that the flaperon drifted left of the wind. Field testing of a genuine Boeing 777 flaperon cut down to match photographs of 9M-MRO’s flaperon has now largely confirmed the DGA predictions, at least with respect to drift angle. The impact of this information on simulated trajectories across the Indian Ocean is that the July 2015 arrival time at La Reunion is now very easy to explain. »

    Why did they use replicas in the first place (instead of a genuine flaperon) ?

    And why wasn’t the MH370 flaperon used, that was found on Réunion ? Yes, it’s a piece of evidence, but they had 1 year to examine it, so I guess they have already done all the work on it. And putting it into the water, let it drift for a while and recovering it later won’t damage it anyway.

  230. Don Thompson says:

    Brock,

    The TUDM/RMAF equipment radar equipment is as described in my briefing paper on the subject. As is so often the case, Gunson’s information was simply incorrect. A UK journalist, knowledgeable in the technology and familiar with the region, attended Malaysia’s LIMA 2015 defence industry exhibition and confirmed my information with industry representatives.

    I have compiled similar information for Thailand’s southern air defence region, RTADS-III, Indonesia’s northern region, Kosekhanudnas III, India’s Andaman & Nicobar command, and Singapore. One site in the Kosekhanudnas III sector remains unidentified, the radar head location for the Sibolga SATRAD 234 unit. I have the location for its ‘barracks’ but not the ‘high site’ of the radar unit.

    I haven’t yet updated the paper.

  231. David says:

    @Gysbreght and others still following the end of flight simulations vs final BFO descent rates.

    https://www.dropbox.com/s/yx1jrn3c1eiom4a/Final%20BFOs%20and%20pitch%20down.docx?dl=0

  232. Andrew says:

    @DrB:

    “I am trying to understand the Checkpoint fuel tables in the December 2007 FCOM (pages PI.21.4-6) for the Trent engines. The tables give the fuel required (at LRC) and the time for various distances to be traveled at each specified altitude. I think the purpose of these tables is to verify if adequate fuel is onboard to reach a destination at that distance, altitude, and weight. My question is, do the fuel and time tables include allowances for descent and landing?”

    You are correct; the tables can be used to determine the fuel required to reach the destination, given the distance, altitude and weight. The tables include allowances for the descent and landing.

    “Based on the variation in the average speeds I have calculated for each adjacent pair of distances, it seems like the descent and landing fuel and time are included. I also see that there is a fuel offset at zero distance of -1.5 MT (at FL370). Perhaps this is the fuel needed for descent and landing.”

    Perhaps we’re looking at different charts, but in my copy of the FCOM the minimum air distance shown in the Low Altitude table is 200nm and in the High Altitude table it’s 400nm. I can’t see any reference to a ‘fuel offset’ of -1.5MT. The typical fuel required for descent and landing in the B772 is about 1,000kg from FL410. About 300kg is used during the approach from 3,000ft to landing, and the rest during the descent. Descent times from various altitudes are shown in the the ‘Descent at .84/310/250 KIAS’ table. You could use those figures together with the descent fuel allowance (1,000kg, adjusted for altitude) to estimate the level flight components.

  233. Richard Cole says:

    @Brock

    If I understand it, the right hand panel of figure 4.1 and the online appendix address items of different windage and their probability of arrival on Australian shores. The ‘specialness’ of 35S is most apparent for the flaperon modelling.

    I don’t disagree that the analysis is unconvincing. It certainly won’t cause the search to be restarted and therefore is probably at a dead-end.

  234. Gysbreght says:

    @David:

    I’m not accusing the ATSB of lying. They have carefully chosen their language to avoid being accused of that. When they say “Similarly, the increase in descent rates across an 8 second period (as per the two final BFO values) equalled or exceeded those derived from the SATCOM transmissions.” I have no reason to doubt the accuracy of that statement. What they avoid saying is that those 8 seconds did not start at the 7th arc, but much later.

    In your paper you simply assume that the airplane descends at an angle of 21 degrees. It doesn’t do that two minutes after the loss of the autopilot.

  235. David says:

    @Gysbreght. OK best left there.

  236. Ge Rijn says:

    @Richard Cole

    Thanks you mention; ‘The ‘specialness’ of 35S is most apparent for the flaperon modelling.’
    Ofcourse this is the case with this study which I’m now more aware of.

    In a way the flaperon only represents one end of the debris-spectrum with it’s high windage, leeway and speed. And with it’s abondance of barnacles it’s one of only two pieces that can make a fairly save estimate of the arrival time.
    Which brings me to the piece on the other side of the debris-spectrum; the Mosselbay RR-piece.

    This piece would have drifted flat on the surface without the extra speed and leeway from the wind the flaperon recieved. It also had a lot of barnacles attached when first found in december 2015 which also makes a save estimate on its arrival time. Only ~5 months after the flaperon but the most distant piece from the 7th arc so far.
    While the flaperon is one of the closests to the 7th arc.

    So I think in important ways those two pieces show opposit characteristics that could serve as representive of all other found debris between this two opposits also representing a time window of ~5 months in which most other debris-items reached their African destinations.

    I suggest repeating the flaperon-experiment with a same kind of copy of the RR-piece.
    Then you’ll have results that could cover the whole spectrum a lot better IMO.

  237. Oleksandr says:

    @Dennis,

    Re: “The navigation data does not depend in any way on the BFO error. The ISAT system is a communication system not a navigation system. The AES oscillator is free running and not corrected by anything.”

    Indeed. What I meant is that the velocity and location, which is used for the estimation of the Doppler compensation term, is not exactly the actual velocity and location of the plane. Say, for example, 1% discrepancy in the velocity estimated by the ADIRU compared to the real one, may result in 10 Hz BFO error.

  238. Oleksandr says:

    @Don Thompson,

    ALSM stated: “there were no erroneous values of BFO or BTO circa 00:19 as you assume”. I asked where does it come from. Your response: “Powerpoint, that’s where it comes from. Communicating for too long in Powerpoint-esque soundbites has made a lot of people incapable of describing a concept concisely and coherently.”

    Is that powerpoint the one, where the IG insists on 38S? Your comments like this are useless.

    Re: “If the ADIRU fails in-flight, things aren’t good, satcom is a low priority issue (but the crew has HFVOICE as an alternative to SATVOICE).”

    Thank you for the comprehensive comment, but unfortunately it has nothing to do with my question.

  239. DennisW says:

    @Oleksandr

    OK. I understand what you are saying. The signature of the figure 5.4 error is so familiar to me. I tried to go back and bring up oscillator data I had logged in a previous life, but the drive I used to save it crashed. Thank god for the cloud. 🙂

  240. Don Thompson says:

    @DrB:

    Your questions:

    1. The BTO then [16:00:13.4 ] does not contain multiple increments of 7820 microseconds.

    It is not the ‘peculiar’ Log On/Off Ack burst. The later ones are, as you put it, the peculiar ones. Only peculiar in terms of BTO measurement. That AES burst occurs in the first slot, of four, available for an AES to transmit. At this time the AES remains sync’ed with a P/smc 600bps channel with a frame length of 2sec. The return R/smc 1200bps burst requires only a 500msec slot – the AES therefore has four slots in which to make the burst.

    2. It occurs 17 seconds after the log-on request message (not 7-8 seconds).

    While the AES must initiate a return channel burst within 300µsec of the frame boundary derived from the P-ch there is no latency constraint on the the datalink management exchanges (a higher level element of the protocol). The extended delay may even have resulted from an R-channel collision with a transmit from another AES. Regardless, I don’t see there being anything of concern as a result of that latency for the Log On process.

    3. There is a second message with the same function in the revised data log at 16:00:13.9 in the P-channel.

    Yes, the AES first acknowledges via the R/smc channel that it has received the channel allocation information, then the GES acknowledges the completion of the (datalink) Log On with its LLA SU transmitted on the allocated P-channel, IOR-P10500-0-3859.

    In case anyone’s missed it, the P-ch notation translates as:
    IOR: Indian Ocean Region
    P: P channel
    10500: channel bit rate
    0: beam ID, 0=global
    3859: coded frequency of the P channel.

  241. Oleksandr says:

    Bobby,

    Re: “I was attempting to derive the LRC fuel flows at various altitudes, especially those between FL250 and FL100, which are included in the Checkpoint tables but which are not included in the LRC table on page PI.21.3. Does anyone know how to do this in such a way as to get the level-flight fuel flow not including the descent and landing?”

    Just use differential, and ignore the first values in the columns. Earlier I described an example how I did it; see my post April 6, 2017 at 12:55pm.

  242. Paul Smithson says:

    With to end of flight. Comments on the hypothetical scenario set out below would be welcome:-

    1. Allow, for a moment, that the BFOs at 0019 are anomalous and the aircraft was not in steep descent.
    2. Consider the possibility that the Left a/c bus was powered up from the Right engine at SDU re-boot (the left IDG having been deliberately isolated earlier due to electrical problems on left side).
    3. Power-down of SDU should then arise when right engine goes down; left engine still fuelled and running for xx minutes.
    4. Autopilot remains engaged, if I’m not mistaken, using the hot battery?
    5. Does hydraulic power remain if left engine running, even if IDG is isolated?
    6. APU start-up follows loss of power from Right engine, creating the unexpected logon.
    7. Logon of IFE not completed because IFE is not powered (being one of the low priority items that is shed by the ELMS).
    8. Plane flies beyond 7th arc for some minutes under power from left engine.
    9. post-script If Left IDG was shut down/unloaded for nearly the entire flight, how much less fuel would have been used by left engine?

  243. Ge Rijn says:

    @ROB

    To come back on your comment about the debris and the No.9 and 15 pieces (left and right wing flaperon sealing panels).
    You know I can only agree on those panels seperated by water impact in a same manner most probably when both flaperons seperated too and forced those panels to break at the same time.
    Anyway it seems quite impossible to me those panels seperated due to flutter or any other aerodynamic force.

    Could well be that left wing flaperon is still drifting or lying somewhere waiting to be found. Time will tell.

    To me it’s disappointing still the official narrative is a high speed dive impact and quite some unofficial serious investigators keep defending this scenario too while this kind of facts on the debris (and many others) just clearly tell there could not have been a high speed dive impact.

    All the discussing on the final BFO’s and ‘dive-descents’ would be rather unnecessary or more clear if this was reqocnised IMO.

  244. ROB says:

    @Ge Rijn

    You are absolutely right. Flutter is dead in the water (if you will excuse the awful pun)

    The reason the ATSB are sticking doggedly to the high speed, unpiloted dive scenario it simply to allow them to discard the piloted glide scenario, gliding beyond the primary search area. The Malaysians won’t countenance a piloted EOF, for pretty obvious political reasons, and the ATSB have to go along with them. That is why the are drawing attention away from S38 extended glide, by focusing on S35 ( with the cooperation of CSIRO). The ATSB are banking on no one having the stomach to restart the search. It will be game over, with the ATSB coming out with haloes on their heads, and squeaky clean. The name of the game at present, is Damage Limitation. All very sad, but inevitable.

  245. Ge Rijn says:

    @ROB

    IMO whatever happened, piloted or not, the plane did not enter the water in a high speed dive but in a ~level, low AoA, relatively low speed attitude. I won’t repeat my arguments (for heavens sake..).

    I keep disagreeing with your 38S extended glide assumption.
    38S is excluded indefinetely by now by any means. I won’t name them all.
    Anyway this latest CSIRO/Griffin study proves once again any crash area south of 36S must be considered out of order.

    I assume also, from the beginning, this dissapearance must have been a planned and all controlled flight till the end. It’s what human logic/common sence dictates and the debris clearly tells us.

  246. Richard Cole says:

    @Ge Rijn

    >I suggest repeating the flaperon-experiment with a same kind of copy of the RR-piece.

    ATSB did develop a dummy RR piece and tested at the same time as the original dummy flaperon. The work is described in the December 2016 paper. It moved at a similar speed to the undrogued drifters.

    There is a set of kmz files for google earth derived from the Griffin model at:

    http://www.marine.csiro.au/~griffin/MH370/kmz/

    Playing with those indicates sharply different predicted behaviour with regards to arrival of debris in southern Africa. With origins at 37S (and further south) no test particles reach southern Africa at all. Origins much north of 35S and the test particles arrive there too early, as mentioned in the text.

  247. TBill says:

    @All
    Elementary I know, but….
    Is there an open source of the seven Arcs lat/long or kmz for Google Earth? I am thinking the Google links point to deleted material at Duncan’s site. I can calculate by hand from Yap but looking for source of calculated ring data at say FL350. I have some ring kmzs by Hussein but these seem to be on the ground (not in agreement with DrB/Victor/et al). What I do is put the rings on Google Earth, and then I can put the Flight Sim airplane icon on Google Earth and fly through the rings on Google Earth to develop path ideas.

  248. Ge Rijn says:

    @Richard Cole

    Thanks for the link.
    I know similar test were done. It’s no suprise the RR-piece would behave like an undrogged drifter though but the speed it arrived at Mosselbay only ~5 months after the flaperon suggests to me it must have started (far) more north of 35S. ~31S would fit this time frame much better IMO.

    Considering most found pieces have no leeway like the flaperon I think a more northern location of origin is more likely (north of 35S).

    Therefore I think an exact similar study by CSIRO with an exact RR-piece copy (like the flaperon) could be usefull.
    At least the RR-piece won’t be that hard to duplicate.
    Independently finding results next to the ATSB findings won’t do any harm.

  249. Brock McEwen says:

    @Don: thanks for the reply, and for listing the selected locations you researched. If the full report ever comes to fruition, I will be among its keenest readers.

    Re: TUDM GM400 location(s): is the dead link impossible to resuscitate, then?

    If – I just say “if”, because you may simply have missed my request for help on the dead link – you are replacing that link with “a UK reporter told me”, then I’ll need to follow up with that reporter, to drill through to actual evidence. Can I have his/her name, please? Feel free to DM or email it to me. Thanks.

    I don’t discount your assertion that S Gunson’s reporter was less informed/precise than your reporter. It’s just that they reported wildly different things. I have his reporter’s name, but not yet yours.

  250. Victor Iannello says:

    @Richard Cole: When I examine those KMZ files for low windage (1.2%), non-flaperon debris starting from 35S latitude, there is no debris that reaches Mozambique and South Africa by the end of December 2015 and some debris that reaches Australia. That directly conflicts with the discovery of the engine cowling (which would be expected to have low windage) in December 2015 in South Africa and the flap track fairing in December 2015 in Mozambique. Examining the high windage (3%) non-flaperon debris starting from 35S latitude also at the end of December 2015, there is sparse debris that reaches Mozambique and almost no debris that reaches South Africa, while much debris has reached Australia.

    Using the KMZ files generated by Griffin’s model, I don’t understand how the debris discoveries in December 2015 in Mozambique and South Africa are possible.

  251. DennisW says:

    @VictorI/Brock/Richard

    Why do we believe that ocean drift modeling (and observations) done in 2016 would accurately reflect what occurred a year earlier? Are ocean currents and surface winds static? Certainly there are trends, but just like the atmosphere there is great variability depending on weather and solar radiation (which in turn influences water temperature and movement). I have not been able to find a source that addresses this subject at all.

  252. Victor Iannello says:

    @DennisW: The ability of the CSIRO’s model to predict the drift of MH370 debris is another question. I was trying to simply compare their predictions with the debris data for a crash at 35S. It appears the debris record does not support their conclusions, but perhaps I am missing something.

  253. DennisW says:

    @VictorI

    Yes, I understood your question. Mine is different. There was even a significant tropical storm that passed into the SIO shortly after 9M-MRO went missing. Storms of that type are fairly rare in that region as I understand it.

  254. Oleksandr says:

    @Victor,
    @Richard Cole,

    I have run my drift model with random leeway, making certain assumptions with regard to the distribution. The most intriguing location in terms of the matching to found debris is 28.8S. Here is a plot:

    https://www.dropbox.com/s/uxqs9qf5g5pss7b/shoreline_dist_99.09E_28.84S.jpg?dl=0

  255. Victor Iannello says:

    @DennisW: There were about eight storms in the region that could have impacted the trajectories of debris from MH370. I’ll be posting a paper from a contributor here that discuss the impact of storms and critiques other aspects of the CSIRO report. I’ll jump to the contributor’s conclusion–the debris record does not support a crash at 35S.

  256. Don Thompson says:

    @Brock

    Concerning the bad links: one bit.ly references a TUDM website page, the site has been reorganised since. However, it is archived at archive.org, the pertinent text translates as, “This year, the government approved the construction of infrastructure projects Skn 331 new units in Kota Samarahan and new radar site known as the GM 403 is located on the French-made Mt. Kili, in Simunjan, which is expected to be completed and fully operational by 2016.” Progress is typically slow is Malaysia.

    An additional relevant reference cites “one new long-range air defence radar […] the Ground Master 400”

    Take the information I have collated entirely from open sources, and then independently corroborated, or leave it. I don’t intend to play games over it.

    :Don

Leave a Reply