MH370 Search Update – Mar 18, 2018

Ocean Infinity’s search progress, from Richard Cole.

Recent Activity

Seabed Constructor, the vessel operated by Ocean Infinity to scan the seabed in search of MH370, is returning to port in Fremantle, Western Australia, to refuel, change crews, and resupply. Constructor is completing the second of three or four swings, each swing lasting about six weeks. So far, there have been no promising sonar “contacts” that might represent the debris field of the missing aircraft.

There remains about 3,000 sq km of seabed to search in the area that the ATSB and CSIRO designated as a priority. After that, the extended search area along the 7th arc would require scanning about 46,600 sq km to reach north to around 29S latitude if the width of the search was 25 NM on either side of the 7th arc. That will require more than one additional swing to complete. In fact, it could prove challenging to complete with even two additional swings, depending on the weather and how well the eight autonomous underwater vehicles (AUVs) perform. Although not publicly stated, there are indications are that at least one of the AUVs is having technical problems.

What We Know So Far

Any scenario that leads to a particular location (a “warm spot”) is based on a set of assumptions, and the failure to find the debris field in proximity to this location means that one or more of those assumptions are false. So we can be fairly certain that the large, blurry objects seen in the French satellite images were not from MH370, as the corresponding impact locations calculated by CSIRO were searched without success. Also searched was the warm spot that was calculated by assuming that the aircraft flew until fuel exhaustion on a path towards the South Pole. Unless an interesting contact was found but not yet disclosed, this scenario also can be dismissed.

In the coming weeks, other scenarios will be searched, including the impact site near 30S latitude that is based on two floating debris fields that were spotted during the aerial surveillance, and discussed at length in a previous post.

Reasons Why the Debris Field Has Not Yet Been Found

Although the area already scanned by Seabed Constructor was designated the highest probability by the ATSB and CSIRO, there are reasons why endpoints outside of this area are still possible.

  • A descent at 18:40 followed by a holding pattern, excursion, or other “loiter” before the turn to the south could mean the plane impacted along the 7th arc to the north of the priority area that has been searched. The last radar target was captured at 18:22, and after 18:28, the next ping arc derived from the BTO data is known at 19:41. There is simply no way to be sure of the path of the plane during this interval.
  • A shift in oscillator frequency of the satellite data unit (SDU) of the SATCOM, which would change the value of the fixed frequency bias (FFB) that is used to convert the location and speed data into a BFO that can be compared with the measured BFO values. In a nutshell, if the FFB shifted by +7 Hz after the power up at 18:25, endpoints as far north as 27S are allowed by the BTO and BFO data. It turns out there is an effect called “retrace” that causes oscillators that are powered down, cooled, and powered up to shift in frequency, and there are indications that a retrace shift of about -4 Hz occurred while 9M-MRO was on the ground at KLIA before the MH370 flight. A similar shift, but in the opposite direction (up) might have occurred due to the inflight power cycling.
  • Pilot inputs after 19:41 might have altered the path. The continuous, smooth progression of the BTO and BFO data suggests automated flight with few or no pilot inputs until fuel exhaustion. However, there is a remote possibility that the smooth progression of values was produced by a more complicated path that by chance replicated the simplest of paths.
  • There is also the possibility that the previous search was as the correct latitude along the 7th arc, but the width of +/- 25 NM from the 7th arc was not sufficient. The final two BFO values indicate a steep, increasing descent that if continued would mean the plane impacted close to the 7th arc. The debris is also consistent with a high-energy impact. However, it is possible, albeit unlikely, that a skilled pilot carefully recovered from the high-speed descent, regained altitude, and glided for some distance beyond 25 NM.
  • Although some of the area north of the priority search area was searched by aerial surveillance in the weeks following the disappearance, the search area was large and the coverage was spread thin. Also, some debris was seen from air, but never recovered due to the distance of ships supporting the search effort.

Simulation of Seabed Constructor’s Search Pattern

Finally, Richard Cole, who has carefully been tracking and analyzing the search patterns of Seabed Constructor, has produced a short video which shows the path of the vessel and how it relates to the launch and recovery of the AUVs. Richard is quite talented at extracting a lot of information from small amounts of data, and this video, like all his work, is commendable.

1,124 Responses to “MH370 Search Update – Mar 18, 2018”

  1. Mick Gilbert says:

    @Victor Iannello

    Victor, with regards to the possible impact of the retrace effect causing a change to the value of the fixed frequency bias subsequent to the power up at 18:25, what does that mean for the interpretation of the offset manoeuvre and more broadly for the flight path track interpretation between 18:25:27 and 18:28:15?

  2. airlandseaman says:

    Nice post Victor. And the Cole animation is an excellent visualization of how the search is being conducted.

    Regarding the OCXO retrace error (OCXO power cycle offset): Those of us with experience in the field have known about retrace error since the beginning. It has not been discussed much because it is very difficult to estimate separately from all the other possible small OCXO offset errors. Given what is known now (null areas south of S32), it is likely that the BFO bias did change slightly after the 18:25 reboot. That change was probably the result of some small retrace error (OTOO 1-5 Hz) and some medium term drift as observed in other MH370 flights (OTOO 1-10 Hz). Together, these errors in post 18:25 BFO bias could easily explain why paths to the more northern part of the 7th arc (S35-S25) produced increasingly larger BFO error, assuming a constant Bias of 150 Hz. If we assume a combination of retrace and drift error, the lowest BFO errors may be found for paths ending in this northern part of the arc.

    It must be emphasized that there is no evidence to support a specific post 18:25 BFO Bias change. Thus, we can’t simply calculate a new path ending further NE on the 7th arc. But the more area searched to the SW, without a find, the more likely it is that there was a bias change.

  3. Victor Iannello says:

    @Mick Gilbert: It depends on the amount of shift, but I haven’t studied it in depth Also, there are other explanations to explain the BTOs and BFOs besides a lateral offset. For instance, it’s also possible there was a turn and a change of speed rather than two turns to produce the offset.

  4. David says:

    @Pilatus. The validity of the final BFO.

    Your http://mh370.radiantphysics.com/2018/02/16/mh370-search-update-feb-16-2018/#comment-13304

    Also related to @Victor’s 4th dot point above.

    I assume what you referred to is at the ATSB’s page 101, footnote 29 (below). That follows a discussion about a hypothetical controlled ditching, in which the ATSB implies the final log-on was on the way to that, ie was pilot induced.

    The 0:19:29 BFO descent rate at its Holland minimum is compatible with a glide. It is the 0:19:37 which is inexplicable without an average acceleration between the two about equal to two thirds that of gravity. That seems possible and the ATSB believes it has been demonstrated in a Boeing simulation, though simulations are limited in what they can reproduce. Also, the simulation data released was incomplete.

    The footnote says, “It should be noted that these descent rates were derived assuming the SDU was still receiving valid track and speed labels from the ADIRU at 0019:37 UTC for use in its doppler pre-compensation algorithm.”

    If the ADIRU was working at 0:19:29 one is left grappling with why it (and its back-up) would not have been functioning correctly 8 secs later; and in a way that did not inhibit the transmission then. Still, the ATSB wording implies that a transmission of corrupt data would be possible, to me anyway.

    This might have been gone into previously and perhaps the ATSB wording should not be interpreted that way. Others may be able to comment.

  5. TBill says:

    @Victor
    Nice article, nice video by Richard. What I have been wondering, does Richard have a feel for all of the AUV paths or just some of them? In particular the coverage of the northern region (difficult Diamantina area) would be interesting.

  6. DennisW says:

    @David

    cut-paste from a Reddit poster below:

    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.”

  7. Mick Gilbert says:

    @Victor Iannello
    @airlandseaman

    Thank you, Victor, and thank you Mike for the additional explanation.

  8. Barry Carlson says:

    @DennisW

    Just to get all the “ducks lined up”, I originally reported to Duncan Steel that the ATSB report published on 27 July 2016 was altered on 28 July 2016. Following a request from Richard Godfrey, I then sent him the following:-

    I don’t keep the reports. It’s more a case of just keeping a check on when the ATSB do any updates.

    In the case in question, the report published on 27 July 2016 included near the end the following,

    “The last satellite communication with the aircraft showed it was most likely in a high rate of descent in the area of what is known as the 7th arc. This is indeed the consensus of the Search Strategy Working Group.”

    They following day, 28 July 2016, I received a notification that text in the containing paragraph had changed, i.e.
    “This is indeed the consensus of the Search Strategy Working Group.” was removed.

    I also checked the Last updated content in the HTML mark-up and that confirmed the page was last updated on 28 July 2016.

    My take on what I saw, is that there is now a lack of consensus in the SSWG.

    … Richard Godfrey then pursued the matter further.

  9. David says:

    @DennisW. Yes thanks a much-ado-about-nothing, which I described just as “curious” on the JW site immediately after the ATSB deletion. Subsequently it was explained as per your quote’s final paragraph.

    I think that was Ean Higgins’s work. Some of us have reservations about most of that, though I recall that you find him interesting.

    Be that as it may, what I raise now supposes that the ATSB comment I refer to, which might be different to the Pilatus “notation”, is not the result of dissent at all. It well post-dates what Higgins stirred up.

  10. DennisW says:

    @Barry

    Thx.

    I am still reeling from the Mick post thanking Victor and ALSM. I never get thanked for anything. My analyst claims it is because I am an A-hole. Ami (my SO) agrees with that opinion. I think more often of taking one of my Glocks behind the barn with a blue tarp. You can only do so much.

  11. Pilatus says:

    @David,

    “Still, the ATSB wording implies that a transmission of corrupt data would be possible, to me anyway.”

    Precisely my point!

    The fact that the note has been provided opens the possibility that that the final extreme descent rate may never have happened.

  12. David says:

    @Barry Carlson. Had not seen yours. Maybe this is what started it? From JW:

    “David
    Posted July 28, 2016 at 2:56 AM
    @Aaron. A curiosity, about the SSWG. At one stage on the ATSB Operatonal Update yesterday 27th July it read, “The last satellite communication with the aircraft showed it was most likely in a high rate of descent in the area of what is known as the 7th arc. This is indeed the consensus of the Search Strategy Working Group.” Today that second sentence has been deleted.”

    @DennisW. In my above, “Yes, thanks…..”

    That deals with your problem.

    This exemplifies MY problem. My posts go unread. Should you reassure me that you have read this and I will faint.

  13. DennisW says:

    @David

    My take on it is that it is a simple matter of “geek” people with advanced degrees getting their panties wadded up. I have seen it over and over again. I would not attach any significance to it.

  14. Mick Gilbert says:

    @DennisW

    Well, thank you for noticing, Dennis. It may well have been a while between drinks but don’t forget this one.

  15. DennisW says:

    @Mick

    Thx. I am just being whiney.

  16. David says:

    @Pilatus. Thanks. I was unsure to what you were referring and it all got rather mixed up with SSWG dissent. Lets see if there is a ready explanation.

    It would need to be possible for crook data generated by the ADIRU/SAARU combo to get past cross checks and inhibitions and further for those circumstances to be feasible in MH370 at that time and not before.

  17. David says:

    @Mick Gilbert. “Errant nonsense”

    The ATSB has been quoted, “Mr McMahon contacted ATSB via Facebook and its general enquiries email in 2016 and 2017 respectively. At no stage did the ATSB suggest his evidence could be missing flight MH370,” a spokesperson told Newshub.

    “The images sent to ATSB by Mr McMahon, below, were captured on 6 Nov 2009, over four years before the flight disappeared.

    “Spurious claims such as these must be particularly upsetting for the family and friends of those lost on Malaysia Airlines flight MH370.”

    Beats me how he got the traction. Shame on them all. Thankfully not The Australian in this case. Progress.

    http://www.newshub.co.nz/home/world/2018/03/mh370-conspiracy-australian-officials-slam-investigator-over-cover-up-claims.html

  18. Richard Jones says:

    I am immensely grateful for all the careful analysis of the available evidence done by many very competent people whether being paid or not. You collectively have a much greater understanding of the details of satellite to aircraft to earth communications than I could ever hope to have.

    However, “high probability” of finding, based on the assumptions made and analysis done, never meant 100% probability. The other side of the coin is a “low probability” of not finding. I can accept that fact, even though I do have a very strong personal and professional interest in #MH370 being found.

    Keep up the good, solid analysis and also the current commentary on the OI search. Never give up till the main wreckage of MH370 is found.

  19. Richard Godfrey says:

    @DennisW, @Barry Carlson, @David,

    I confirm what Barry Carlson wrote above and that I had asked him about the deletion in the ATSB Operational Update. I was following up on a previous email from Barry Carlson to Duncan Steel.

    On 28 Jul 2016, at 12:14, Barry Carlson wrote:
    “I have a “change” utility keeping watch on the ATSB Operational Update page, and today part of your yesterdays quote, i.e. ” This is indeed the consensus of the Search Strategy Working Group.” has been removed. It’s not the first time I have found them quietly modifying their updates.”

    I had been contacted by The Australian in the past a number of times, but I now refuse any comment, as my statements have been inaccurately reported and taken out of context.

  20. Rob says:

    @Peter Norton

    “Rob says: ditch just forcibly enough to sink it as rapidly changing as possibly, in as few pieces as possible

    IIUC this is the “hard ditching” scenario (sort of in-between a soft ditching and a hard crash).
    The problem with this scenario is that the person in control of the plane doesn’t know whether he will survive the ditching or not.
    If you commit suicide, you want to know how your life ends beforehand and also choose the fastest, least painful way, in which you suffer and struggle the least. This would be the case in a hard crash, leading to instantaneous death.”

    Yes Peter, this is the hard ditching scenario. Now, what you and others who focus on the problems of pilot suicide in relation to MH370 seem unable to fully understand is that the pilot didn’t hijack his plane and fly it to S38, E89 simply in order to kill himself. The situation is slightly more complex than that. The pilot hijacked his plane, switched off the transponder, disabled ACARS, then flew in such a way as to attempt to avoid subsequent path detection/reconstruction in order to make the plane completely disappear from the face of the earth (not to mention the ocean). This was primarily a grotesque act of revenge and anger, probably against the Government, possibly the airline and just possibly against humanity itself. That he was going to die was to him just a necessary and obviously acceptable consequence of the operation. He had probably given little thought to exactly how he would die. He primary aim was to complete his mission as planned. The plan was to make the plane disappear, pure and simple. He did not want, and did not intend for any wreckage to turn up at a later date, which is the reason why he ended up running out of fuel at S38. It was as far into the remotest, quietest part of the SIO that the fuel would take him, and it would ensure enough daylight when the time came to make a hard ditching and disappear for good.

    Mulling over statistics on pilot suicide only muddies the water, and distracts focus away from what really happened that night. The Germanwings horror is an example of pilot suicide. Do you see any similarities between MH370 and Germanwings, apart from the locked cockpit door?

  21. Rob says:

    @VictorI

    “A shift in oscillator frequency of the satellite data unit (SDU) of the SATCOM, which would change the value of the fixed frequency bias (FFB) that is used to convert the location and speed data into a BFO that can be compared with the measured BFO values. In a nutshell, if the FFB shifted by +7 Hz after the power up at 18:25, endpoints as far north as 27S are allowed by the BTO and BFO data. It turns out there is an effect called “retrace” that causes oscillators that are powered down, cooled, and powered up to shift in frequency, and there are indications that a retrace shift of about -4 Hz occurred while 9M-MRO was on the ground at KLIA before the MH370 flight. A similar shift, but in the opposite direction (up) might have occurred due to the inflight power cycling.”

    However, couldn’t the change in FFB, the retrace, have caused a shift in the negative direction following the in-flight power up? Surely there is no knowing which direction the bias shifted? If the shift was negative, it would favour southerly flight paths? Is it possibly because you had crossed the Rubicon and already ruled out southerly paths, that you assume the retrace effect will necessarily favour a more northerly path?

    I point this out not as you might think, to be merely disruptive, but to draw attention to possibly another example of theory reinforcement or confirmation bias.

    Unless a fundamental rethink is done, a critical review of the assumptions underpinning the favouring of northerly paths, then the result might very well be that OI fail to find the wreckage, through no fault of their own. There is a risk that a future analysis of this enterprise will conclude that misplaced preconceptions, errors of judgement, suppressed psychological biases and pure lack of centralised decision making combined to snatch failure from the jaws of success?

    Just thinking.

  22. Jack.H says:

    I wish the Inmarsat handshaking data might be wrong. The investigator (in realities not lead by Malaysian, but AU, ATSB British & the so-called American’experts’) was totally discarding all those eye witnesses at southern area. The ‘transparency’ that was being pointed out by this blog by looking at Malaysian authority update release were not good enough to counter back the main core or fundamentals of searching. Wild accusing the searching issues to the internal Malaysian political matter are the dumbest thing I ever heard in a modern civilization of ‘smart-gadgets age’. We are still hoping the on going search will return positively anyway.. praying for that to the God.

  23. airlandseaman says:

    Regardless of what some ATSB working group members may have believed about the 0019 BFO data accuracy and descent rate at various points in the past, there is no question now.

  24. David says:

    @Richard Godfrey. “I now refuse any comment, as my statements have been inaccurately reported and taken out of context.”

    Yes I imagined that was what had happened.

  25. Richard Godfrey says:

    @Victor

    Many thanks for the excellent summary of the current situation regarding the search for MH370.

    1. I agree we do not have definitive evidence of what happened between 18:28 UTC and 19:41 UTC, but we do have circumstantial evidence, from both the home simulator of ZS and the “Yellow Pin” shown in the initial Malaysian Report, of a flight path out over the Andaman Islands.

    2. I agree the BFO FFB may have shifted after the 18:25 UTC power up and that MH370 End Points further northwards toward 27°S are allowed by the satellite data, but the transoceanic drift analysis, although much less precise than the BTO and BFO, does show a tendency toward 30°S and a slight tendency away from 29°S northwards and an increasing tendency away from 27°S northwards.

    3. I have not yet seen any definitive evidence of pilot input after 19:41 UTC.

    4. I believe that +/- 25 NM is a reasonable search width and I expect the aircraft to be found well inside this width from the 7th Arc.

    5. I agree the aerial search was very thin (11% coverage) and floating debris could well have been missed. The RNZAF aerial photos however, still need to be accounted for.

    I conclude that Ocean Infinity should continue searching further north than 29°S +/- 25 NM from the 7th Arc, if they have not found the aircraft before, weather and resources permitting. This would require an extension to the 90 days search timeframe.

    @Richard Cole

    Many thanks for the great video. I presume it takes a lot of time to put together, otherwise I would politely request the same for other areas. I imagine fitting all the SC stopping points to AUV launches and pick ups, which in turn fit an underwater search pattern, is a non trivial and time consuming exercise.

  26. Victor Iannello says:

    @Rob: Sure, a small downward shift of the FFB (around -1 Hz) improves the BFO fit for the more southern path that interests you. However, drift models would have to be very wrong in that case AND a long glide would be required. It’s not confirmation bias. It’s simply not the case I was discussing.

  27. David says:

    @ALSM

    Do you know what the ATSB had in mind as to a process or circumstances which would allow such an outcome?

  28. Rob says:

    @DennisW

    Parking me in the “lame” category will not win the argument. In fact it only diminishes your own position. If you have no effective counter argument to put forward, then it would better serve the cause if you just admitted it.

    You have never once proposed one single realistic, believable explanation for the events, in all this time.

    But the real insult was to lump me in with Ge Rijn. You are a whacko period.

  29. Don Thompson says:

    @David quoted the ATBS’s footnote that included, “assuming the SDU was still receiving valid track and speed labels from the ADIRU” while then translating the statement into “the ATSB wording implies that a transmission of corrupt data would be possible”.

    That’s a rather disingenuous representation of the possibilities.

    The ADIRU is the source of position, attitude, ground track and ground speed data. The ADIRU communicates its data over the Flight Controls ARINC 629 buses. Concerning the AES, the ‘IRS’ data words necessary for the SDU are relayed through AIMS Data Conversion Gateway Function (DCGF) and onwards over an ARINC 429 bus from the AIMS cabinet to the SDU.

    ARINC 629 transfers comprise sequences of data words, ARINC 429 transfers for the ‘IRS’ data are single data words.

    Both the ARINC 629 and 429 bus systems employ single bit parity for data words as an integrity check. The data transfers from the ADIRU, on to the Flight Controls bus, and from the AIMS/DCGF to the SDU are periodic, at a rate of at least 20Hz.

    If a transmission error was to be detected in a single periodic transfer the message/data word would be discarded by the receiving bus terminal. If a data source has no valid data to distribute, messages and data words continue to be distributed but with an indication for the absence of valid data. Without valid ‘IRS’ data the SDU would eventually, implicitly, revert to the Log Off state and the AES would not attempt any further transmission until valid ‘IRS’ data became available.

    The ADIRU and AIMS are both supplied with 28VDC power from multiple sources including the hot battery bus, therefore the end of flight assumption of all generator failure should not have interrupted the operation of either platform.

    The possibility of erroneous data propagating through the avionics network is very low.

    Now, consider that the over-the-air SATCOM data bursts are not instantaneous. At 00:19, the R-ch burst at 600bps for the Log On Request SU has a duration of 960ms, while the following Log On Acknowledge SU on the R/1200 channel has a duration of 460ms. The duration of these bursts implies that the SDU applies multiple doppler pre-compensation updates to that function’s DAC in its IF-RF up-convert signal chain while each burst was ‘clocked’ out. It’s also worth considering other GES metadata for the final received burst: e.g. estimated burst error rate = 0; C/No = 43.38, neither give an indication that the GES demodulator had processed an exceptional burst.

    So not only would ‘corruption’ require erroneous data to propagate to the SDU, but to propagate over multiple periodic transfers to the SDU. My view is that such an event is a long way out on the far end of a slope of diminishing possibilities.

  30. Richard Godfrey says:

    Are Ocean Infinity planning a quick turnaround in Fremantle within 14 hours?

    SEABED CONSTRUCTOR Arrival Planned 22/03/2018 04:00 from Sea to Australian Marine Complex 4.

    SEABED CONSTRUCTOR Departure Planned 22/03/2018 18:00 from Australian Marine Complex 4 to Sea.

    The implication is:

    Repairs and maintenance issues solved in 14 hours.

    Each crew task handover including any open complex technical issues within 14 hours.

    Resupplying for 65 people and 6 weeks within 14 hours.

    Refuelling within 14 hours.

  31. Don Thompson says:

    @Richard G,

    Constructor accomplished its port call at Port of Spain over 30th Nov to 2nd Dec, involving two berthings, within 23hrs time ‘tied up’.

    If offload and replenished supplies are containerised, only refuelling might be the critical path.

  32. TBill says:

    @Richard
    Richard G. and I asked similar questions above, that is, we are wondering what else Richard Cole knows about the AUV paths? Especially the AUV paths in the northern sector of volcanos and trenches would be of interest.

  33. TBill says:

    @Victor
    I am holding in my critique at least one-half day, but holy mackerel after all of this lack of finding aircraft with passive flight assumptions, can’t we bump a notch over “remote” for active flight possibility?

  34. Ge Rijn says:

    To me it’s a bit odd they did not complete the rest of the area but stopped in the middle of it and return to Fremantle. Completing the area would have taken them only a few days extra.

    Is the hurry to leave and go back an indication they found something interesting or just indicating they just go back and have found nothing so far?

  35. Victor Iannello says:

    @Ge Rijn: I suspect the timing of the return trip to Fremantle is related to the contractual agreements with the crew.

  36. Victor Iannello says:

    @TBill: Regardless of how I characterize the possibility of pilot input after 19:41, how would you suggest changing the search strategy of progressively searching to the north along the 7th arc?

  37. Ge Rijn says:

    @TBill

    To shime in. I see the tone has changed lately.
    @VictorI in this topic included specifically (for the first time in a topic this way as I remember well) the possibility of a recovery and glide outside the +/-25Nm by a skilled pilot.
    Also he uses the frase ‘high energy impact’ again. Which leaves more room to what kind of impact occured and not only focusses on a high speed nose down impact. I also preferre this frasing.
    For the debris shows it must have been a high energy impact beyond a soft ditching a la Hudson-ditch.

  38. Ge Rijn says:

    @VictorI

    I guess you’re right. It’s just tense times.
    They stopped around your SPOLE track point. Just at a ‘warm spot’ so to say.
    That made me wonder.

  39. DennisW says:

    @Richard

    3. I have not yet seen any definitive evidence of pilot input after 19:41 UTC.

    I think your point along with running out of fuel are not being given the consideration they deserve. If the PIC wanted to ditch the aircraft in any sort of controlled manner dead engines would not be the method of choice. It is my strong opinion that the plane was pointed South after the FMT in some autopilot mode and that was that.

  40. Rob says:

    @VictorI

    to me: “Sure, a small downward shift of the FFB (around -1 Hz) improves the BFO fit for the more southern path that interests you. However, drift models would have to be very wrong in that case AND a long glide would be required. It’s not confirmation bias. It’s simply not the case I was discussing.”

    Fair enough, but I expect the drift studies to be shown in due course for what they are, which is merely an unhelpful distraction.

    CSIRO admitted to me (via ATSB) in late 2016 that S40 was possible. S40 only fell out of favour after the ATSB failed to find the aircraft. They failed because the assumptions they used on which to base the search, were incorrect. That they were incorrect was partly due to an inadequate appreciation of the problem they faced, ie a rogue pilot commandeering his plane and deliberately, calculatingly flying it to obscurity in the SIO (possibly understandable if you are not experienced in rogue pilot events and to be honest, who is?) The other reason was a restricting political pressure from Malaysia.

    I’m pleased to see that you are now entertaining the need to possibly reconsider the +/-25nm constraint. With any luck we just might get you on board with the piloted glide scenario before time runs out.

    But most importantly, the job we’re all in now is not to score points off each other, but to give OI the best chance of finding the plane. That would be one supreme achievement for which will all be equally proud, and it’s still possible to pull it off.

  41. Ge Rijn says:

    @DennisW @Richard Godfrey

    It makes more sence to reason the other way:
    ‘I have not yet seen any definitive evidence of NO pilot input after 19:41 UTC’.

    Pilot control in an obviously controlled flight till leaving radar detection and a controlled turn into the SIO is much more proof of an all pilot controlled flight than the very, very, very remote possibility the flight suddenly changed in an inactive pilot flight after ~18:40 or any FMT.

    I think it’s a big mistake people holding on to this misconception on facts and statistics.
    The ATSB-search failed on this misconception.
    OI has a better chance if they listen well and have learned from the past. I trust they have.

  42. TBill says:

    @Victor
    “…how would you suggest changing the search strategy of progressively searching to the north along the 7th arc?”

    I am currently in agreement with the search approach. I hate to prematurely mention birthday waypoint, but it exemplifies, pending further study, a possible “warm spot” for spot check outside Arc7 +-25.

    The “active” pilot proponents are not a cohesive group. Let me attempt to categorize the “active” pilot camps:

    Camp-1: Active pilot had target in mind such as Broken Ridge (my camp)
    Camp-2: Active pilot had a style of crash in mind (long glide to minimize debris)
    Camp-3: All of the above (Ge Rijn’s camp – may I say?)

    The Camp-1 sees Broken Ridge as priority search area. Camp-2 needs wider search area. Seems to me, Camp-2 tends to focus on 38S as their area.

    What could help in my mind is developing some example flight paths.

    I see as a “Slanty Dog-Leg” as one BFO-straight-looking, active path. One example is slanting over to Broken Ridge from initial 180S path, and another example I have not yet worked on much, would possibly be slanting over to Path L894 from 180S trajectory (eg; somewhat similar Deplorable Edward’s path).

    Although I am not (yet) a long glide proponent, maybe working out a talking case, showing what had to be done to silence IFE etc, would be instructive. I’d aim that test case at the birthday waypoint just to see if we can make search case for that. Really bizarre possibility to me, if the pilot was sending us hints like that and we had no interest because we want to hold so strictly to passive hypothesis.

  43. Ge Rijn says:

    @TBill

    I have to set things clear right away.
    I’ve stated for a long time my priority area is between ~32.2S/33S and ~95E/97E. 32.26’/97 is still my favorite but it could be well inside the named area.
    I’m not in the camp of a long glide anymore for a long time.
    After recognizing the facts of the final BFO’s I acceptted there could not have been a very long glide after the steep descent.

    I am in your Camp 1 and 2. Camp 2 (like me) possibly needs a wider search at certain areas but definitely not around 38S!

  44. Peter Norton says:

    > Rob says:
    > He had probably given little thought to exactly how he would die.

    I don’t know a single such (real-world) case. Do you?

  45. Niu Yunu says:

    @Richard Cole
    @Victor Iannello

    Interesting video. How to explain the ship’s seemingly erratic movements in-between AUV launches? The ship seems to go to random places, where it neither launches nor collects AUVs. So what is the purpose of that? How can the ship’s track be explained?

  46. Richard Cole says:

    @Richard G

    I may have a go at animating the work inside the arc earlier in Swing #2. It is a deal of work to set up the charts and capture the images but I did this first one as an explainer for the principles of the AUV movements, which I could see playing with the model but were not clear in the still images I was posting.

    General Point on AUV paths:

    Where the AUV releases and recoveries are on a regular pattern then it is easier to work out what the paths are between those pairs of points. In the south of the region in the animation (where the water was deeper and the seabed less regular) there were additional launches but no obvious pattern that the AUVs were following, so I have not shown those launches, as discussed in some brief notes in this thread.

    https://goo.gl/qykvXC

    I could add the southern launches to the animation (and some guesses on recoveries) that would at least indicate that Constructor was doing something during some of its excursions south.

    In general, the less regular the sea floor (like the far north), the less obvious the AUV search pattern and hence the harder to indicate in an animation.

  47. Don Thompson says:

    @Niu Yunu

    Richard C and I have shared some thoughts on the AUV work so I hope Richard doesn’t mind me contributing to the answer.

    The AUV launch phase can be discerned by a ship stop, the launch, followed by a 3 to 6 knot path where the ship is guiding, tending-to-depth, an AUV to enhance navigation accuracy. Once at working depth, the AUV is left to continue its mission autonomously. However, the ship does ‘check-in’ with the AUV during its mission via telemetry over acoustic comms and it’s these ‘check-in’ tasks that give the impression of randomness while zig-zagging between locations. While working a long linear track or an advancing line pattern, each AUV’s search box can be readily deduced with some basic constraints: the track spacing and the endurance of the AUV.

    When working over complex bathymetry features the AUV missions aren’t so readily deduced but correlating detailed bathymetry imagery and the ship’s track does help identify an AUV track.

  48. Richard Godfrey says:

    @Ge Rijn

    I posted earlier today “I have not yet seen any definitive evidence of pilot input after 19:41 UTC.”

    You stated: “It makes more sence to reason the other way: ‘I have not yet seen any definitive evidence of NO pilot input after 19:41 UTC’.”

    My evidence for no pilot input after 19:41 UTC is in slide 6 and 7 of the presentation on “Reconstructed Flight Paths”, please see link below.

    http://mh370.radiantphysics.com/2017/11/25/possible-mh370-debris-seen-in-aerial-search-in-spring-2014/#comment-10052

    My apologies that I missed your evidence for pilot input after 19:41 UTC. Please give me the link.

  49. Niu Yunu says:

    Don Thompson: “it’s these ‘check-in’ tasks that give the impression of randomness while zig-zagging between locations”

    I see. Thank you, Don. Am I correct in assuming that the zig-zagging is due to:
    what you just explained + “there were additional launches but […] I have not shown those launches”
    If some ship movements shown in the video correspond to AUV launches that are NOT shown in the video, this would also explain the non-linear movements between the AUV launches seen in the video.

  50. Richard Godfrey says:

    @DennisW

    You stated: “It is my strong opinion that the plane was pointed South after the FMT in some autopilot mode and that was that.”

    I agree!

  51. Don Thompson says:

    @Niu Yunu,

    I think you’ve got it! Prior to 2018-02-20 the AUVs were running long, linear tracks. Subsequent to 2018-02-20 the AUVs have predominantly been deployed to run advancing line ‘boxes’.

    It’s best to hold with interpretations of mission tracks until Constructor is clear of an area.

  52. Niu Yunu says:

    @Don Thompson: Thanks to your explanations, the video makes much more sense now. Thank you.

  53. Rob says:

    @Ge Rijn

    “I am in your Camp 1 and 2. Camp 2 (like me) possibly needs a wider search at certain areas but definitely not around 38S!”

    I sympathise with you dilemma. But luckily there is a mitigating factor on cue to rescue the situation. Previously, the putative S38 extended glidepath was not sufficiently constrained/defined to permit an economical search. This is no longer the case. My Great Circle path hitting the 2nd (1941) arc at S1.0333, E93.65 and flaming out at S37.60 E88.66, while en route to waypoint S41 E88, and exactly satisfying the range/endurance constraints is specific enough to permit the search area to be limited to a 4,300 sq km circular area of radius 20nm centred on S39.20 E88.36, with confidence. How long would it take OI to search this area, assuming no preliminary bathymetry required?

  54. TBill says:

    @Ge Rijn
    OK got it
    P.S.- for what it’s worth, an alternate birthday variation (31.71S 96.1E) appears to be within 22nm of Arc7 and also is within DrB’s recommended search zone in Site2 Area3.

  55. Ge Rijn says:

    @Richard Godfrey

    You know I have no evidence of pilot input (control) after FMT like no-one has.
    But the (your) evidence you referre to is no evidence either as you realise well too I hope.

    And ofcourse the plane was flown in some kind of A/P mode after FMT. It was not manualy flow for ~5 hours after FMT. No pilot would do that.
    It’s all about what happened at the end-of -flight between ~0:11 and ~0:19.
    Was it pilot controlled or not during that time.
    I don’t know, you don’t know, nobody knows for sure.
    But lookking at the whole flight and statistically, chances are quite realistic it was controlled till the end.

  56. J Magnus says:

    @Rob

    A first time poster here. I’ve really enjoyed reading these threads and appreciate the thoughtfulness of the commentary.

    At the risk of adding another unpopular theory, but hopefully contributing a novel thought with respect to near 40S long-shot theories, I wanted to propose that the pilot’s simulated route may have been actually carried out, and that it was a trial run of sorts.

    This might sound nuts, but I think the pilot could have been headed to a spot that is 7,900 miles beneath either the Pentagon or World Trade Center (i.e., the spot you would reach if you drilled directly through the earth to the opposite side), for possibly symbolic reasons, and because these antipodal coordinates were potentially reachable from the starting point, at least in a fuel exhausted glide, which was what he simulated. If you imagine the earth turned upside down and follow the deleted simulation track, it is a path that crosses very near the “antipodal” coordinates of the D.C. area.  It is essentially a run up the east coast. I would argue that the simulation may have been a test to see if he could glide over and ditch on one of these locations.

    I believe that evidence that the SDU was in fact rebooted and could also be disabled, coupled with the deleted simulator track, combine to make a credible case for the plane having been directed southeast of the existing search areas. Jeff W recently recently argued the point on the SDU reboot calling into question the satellite data in his blog. I agree. In addition, I understand the strength of the science behind the drift modeling, but also know that 10-day weather forecasts also produce data and numbers to point to, but in my experience the output is relatively worthless. Not saying that is the case here, but perhaps this modeling is being given too much weight.

    Given the incredible, incredible expanse of earth and sea the plane could have been flown to, it is very odd to me that the simulation route (and potentially, actual route) happens to head right towards symbolically significant coordinates. There is nothing much in that area of the world to head to, leaving me with the feeling that if the end of this flight was intentional —I very much hope this not to be the case—that the end point may have been a symbolic target (i.e., to leave a pile of wreckage deep below, and exactly below). It may be a long shot, but I believe that these two points may be worth a quick search someday.

  57. Niu Yunu says:

    Captain Byron Bailey says:
    “If the flight crew was somehow incapacitated, the plane simply would have continued on auto-pilot to Beijing.”

    What would have happened in that case?
    Would the plane have …
    – autolanded at Beijing airport ?
    – crashed at Beijing airport ?
    – even been in danger of crashing into populated areas in Beijing ?

  58. Don Thompson says:

    @Niu Yunu asked “What would have happened in that case [that the plane simply would have continued on auto-pilot to Beijing]?

    Upon reaching the last waypoint defined in the FMS route, the FMS would revert to maintain the heading at that point.

    The final waypoint in the filed flight plan was the Dawangzhuang VOR-DME (VYK) reached by flying the A461 airway. An onward heading of 31º would be expected. However, China’s reaction to an uncommunicative intrusion to its airspace after the VVNB-ZJSA FIR boundary, expected at 1810z, might have been more significant.

  59. Andrew says:

    @Niu Yunu

    RE: ”Would the plane have …
    – autolanded at Beijing airport ?
    – crashed at Beijing airport ?
    – even been in danger of crashing into populated areas in Beijing ?”

    Byron Bailey’s comment assumes the pilots became incapacitated before the aircraft turned back towards the Malaysian peninsular, something that was never suggested by the ‘experts’, despite Bailey’s assertions.

    The aircraft can’t autoland without human intervention to set up the autopilot (ie select the initial approach altitude on the MCP and arm LOC/GS modes) and to configure the aircraft for landing (ie extend the flaps and lower the landing gear).

    If the pilots had become incapacitated before the turn back, the aircraft would have continued to fly along the track that was programmed in the FMC, presumably all the way to Beijing. The aircraft would have overflown the last waypoint programmed in the FMC at its previously maintained cruise altitude and then continued flying on its last heading until it eventually ran out of fuel. Given the aircraft was carrying additional reserve fuel to allow for a diversion to an alternate, etc, it would have continued flying a considerable distance past Beijing.

  60. Niu Yunu says:

    @Andrew, @Don Thompson: many thanks to both of you!

  61. Victor Iannello says:

    Here is a new piece authored by @Byron Bailey that appeared in The Australian. Contributor @Rob is mentioned.

    It’s too bad that Byron won’t engage in a productive dialogue here.

  62. Peter Norton says:

    @Andrew:
    since you are here, may I relay a theory from the other place to you for comment (in case you want to comment) ?

    > A few days ago I suggested that an inexpert pilot might have pushed the nose down
    > to recover from a stall.
    >
    > Another explanation for that action might be the DUAL ENG FAIL/STALL checklist
    > calling for “Airspeed …. Above 270 kts” for best windmill start capability.
    > https://www.dropbox.com/s/7w1n3jbsae026pv/DualEngFail%20Checklist.pdf?dl=0

    (link seems dead)

  63. Don Thompson says:

    Per Bailey’s latest airing in the Australian, “Hardy is a former British Airways Boeing 777 pilot who calculated accurately the final position of MH370

    Accurately? Calculated with a printout from Google Earth and a primary (elementary) school 12″ ruler. Hardy’s present tenure is with Turkish Airlines.

  64. Don Thompson says:

    Further

    …and “The ATSB’s bosses and media flacks have dug themselves even more deeply into their cave of silence, refusing to discuss Vance’s findings

    Blatant and mendacious misrepresentation: ATSB published their findings from first hand analysis of the Pemba flap. As Hardy states, Vance has had only access to photographic records, and presumably only those publicly accessible from the ATSB website.

    In reality, it’s Vance who has refused to engage in any discussion concerning his findings, prefer to hold his alleged ‘analysis’ for the publication of his book.

  65. TBill says:

    Byron Bailey article is reflecting the main conflict which I have been saying is obvious.

    Many people, many B777 pilots, hypothesize pilot hijacking. But the theoretical approach has avoided all mention of the active-pilot hypothesis, at least publicly. This is extremely frustrating for some.

    Not for me, because I am willing go beyond Simon Hardy’s old YouTube path finding tutorial, which I loved. I feel the current search area is very good based on BTO/BFO – for an active pilot case.

    One contradiction in Byron Bailey comments, is he embraces the Z simulator studies which had a curved path inwards towards OZ. But he endorses as a near 100% certainty, a out-dated flight path heading out to deep sea, away from OZ. Do Hardy and Keane agree with avoiding all scientific folks who may agree with them in principle but who might suggest some updates to the Hardy-path?

  66. Victor Iannello says:

    @Don Thompson: Full disclosure on Larry Vance. He did contact me and offered to meet me and others (at his expense) to present his findings on why he believes the damage to the flaperon and flap were consistent with a ditching with flaps extended. Unfortunately, he would not release material to us to review prior to the meeting because he wanted to tightly control the material in advance of the publication of his book. I questioned whether the meeting would be productive without a release of materials to us, and the meeting never materialized. I also felt that since he was for some time already very publicly claiming the evidence proved there was a ditching, he should be prepared to publicly back up that claim by releasing his analysis.

    So yes, there was an offer to engage in discussion with him, but he would not release his materials in a way that would allow us to do a full review.

  67. Victor Iannello says:

    @Peter Norton said: A few days ago I [not Peter Norton] suggested that an inexpert pilot might have pushed the nose down to recover from a stall. Another explanation for that action might be the DUAL ENG FAIL/STALL checklist calling for “Airspeed …. Above 270 kts” for best windmill start capability.

    That’s absolutely ridiculous. An experienced pilot does not need to achieve a descent of 15,000 fpm to reach 270 knots to restart the engines, which would be futile anyway since there was no fuel. Meanwhile, an inexperienced pilot would not be following failure checklists, and even an inexperienced pilot would know that after flying for 6 hours, the engine failure was probably due to fuel exhaustion, as the fuel indications would show.

  68. Victor Iannello says:

    @TBill said: But the theoretical approach has avoided all mention of the active-pilot hypothesis, at least publicly. This is extremely frustrating for some…I feel the current search area is very good based on BTO/BFO – for an active pilot case.

    So what’s your frustration? Your proposed endpoint with pilot input will be tested. The ongoing progressive search pattern will search endpoints consistent with many theories, including yours. Whether or not there was a small input that caused the path to deviate some from a great circle path is inconsequential.

  69. DennisW says:

    @Victor

    The Bailey article was horrible. I actually feel bad that the ATSB has to endure that.

  70. Peter Norton says:

    @Victor Iannello: thank you for your comment. I agree.
    (Just for the record: the way you quoted me makes it look like it was my own theory, but it isn’t. I only posted it here for comment.)

  71. Andrew says:

    @Peter Norton

    RE: Gysbreght’s comment “> A few days ago I suggested that an inexpert pilot might have pushed the nose down
    > to recover from a stall.
    >
    > Another explanation for that action might be the DUAL ENG FAIL/STALL checklist
    > calling for “Airspeed …. Above 270 kts” for best windmill start capability.”

    As you mentioned, the link is dead so I don’t know how Gysbreght obtained that information. The DUAL ENG FAIL/STALL checklist actually says “Set airspeed above 250 knots”. The pilot would obviously need to lower the nose to achieve that requirement, but he/she certainly wouldn’t need to ‘stuff the nose down’ (as suggested by Byron Bailey) and generate a rate of descent approaching 15,000 ft/min. I’d suggest that most pilots (apart from Byron Bailey, it seems) would gently lower the nose and smoothly accelerate to the required speed.

    Gysbreght suggested that an ‘inexpert pilot’ might mishandle the aircraft, which is certainly possible. However, I doubt that such a pilot would be familiar with the DUAL ENG FAIL/STALL checklist or how to access it via the ECL.

  72. Victor Iannello says:

    @Don Thompson: Has @Byron Bailey, Larry Vance, or anybody else ever explained how the flaps were extended after fuel exhaustion? As we all know, the RAT supplies hydraulic pressure to “center” flight control surfaces, but not to the flaps, as there is a checkvalve that prevents flow in that direction.

  73. Andrew says:

    Looks like Victor beat me to the punch while I was typing!

  74. sk999 says:

    Bailey – curious. In spite of his latest assault on the ATSB (which I have come to think is a national pastime in Australia), his proposed route is entirely compatible with that of the ATSB, aside from the end-of-flight scenario.

    In fact, his preferred route South matches that of the DSTG:

    “[Simon] Hardy is a former British Airways Boeing 777 pilot who CALCULATED ACCURATELY the final position of MH370 based on Zaharie being in control, and that was farther southwest than the ATSB searched.”

    Hardy’s final position was squarely in the middle of the high probability region of the DSTG’s heat map and was thoroughly searched.

    Much ado about nothing.

  75. Victor Iannello says:

    @Andrew: Yes, we had exactly the same reaction to this silly proposal. You were probably preparing your response when I submitted mine.

  76. Victor Iannello says:

    @Peter Norton: I’m sorry. I went back and inserted a bracketed comment in your contribution to avoid confusion.

  77. Victor Iannello says:

    @sk999: I hope @Byron Bailey chooses to engage in a discussion here. His theories will either stand up to scrutiny by informed readers here, or they won’t. I don’t know of a better forum to have this discussion.

    To be honest, if here were to engage, I think he’d find that there are many here that agree with a lot of his scenario. Of course, I’d politely ask him to stop using inexact terms like “ghost flight” and “death dive”, which drive me crazy because of the confusion they create.

  78. Peter Norton says:

    @Andrew:
    thank you for your comment.

    > Andrew says:
    >> RE: Gysbreght’s comment
    >> Another explanation for that action might be the DUAL ENG FAIL/STALL checklist
    >> calling for “Airspeed …. Above 270 kts” for best windmill start capability.”
    >
    > As you mentioned, the link is dead so I don’t know how Gysbreght obtained that information.

    Andrew, the information is from the B777 manual, page 64 (Sec. 2.0 Page 15).

    Sorry I didn’t mention that. I assumed that as a B777 Captain you would know.

    The manual says:

    DUAL ENG FAIL/STALL
    SECONDARY ACTION
    Airspeed………………………………………………………………….ABOVE 270 KTS
    Ensures best windmill start capability. Engines may accelerate to idle
    slowly. The time from fuel control switch to RUN to stabilized idle may
    be as long as two and a half minutes. If N2 is steadily increasing, and
    EGT remains within limits, the start is progressing normally.
    Be aware of possible asymmetrical thrust conditions depending upon
    individual engine start and acceleration times.
    APU Selector (If APU Available)……………… START, RELEASE TO ON
    Backs up automatic APU start

    > Andrew says:
    > The DUAL ENG FAIL/STALL checklist actually says “Set airspeed above 250 knots”.

    If you have the checklist, could you post a link ?

    > The pilot would obviously need to lower the nose to achieve that requirement, but he/she certainly
    > wouldn’t need to ‘stuff the nose down’ (as suggested by Byron Bailey) and generate a rate of descent
    > approaching 15,000 ft/min. I’d suggest that most pilots (apart from Byron Bailey, it seems) would
    > gently lower the nose and smoothly accelerate to the required speed.

    Agreed. It’s just curious that a veteran pilot with 45 years of experience (15 of which on type) would say that.

  79. DennisW says:

    @sk999 re: Hardy’s postion

    Thanks for that. I did not even bother to check it. I never assumed Bailey was that dumb.

    @Rob

    English scientist Robin Stevens, an independent MH370 investigator, calculated the time of MH370 fuel exhaustion coincides with the local mean time of sunrise.

    “local mean time of sunrise”??

  80. sk999 says:

    R.E. Larry Vance – “Unfortunately, he would not release material to us to review prior to the meeting because he wanted to tightly control the material in advance of the publication of his book.”

    Not sure why, but somehow I am reminded of Pons and Fleischmann.

  81. Rob says:

    @J Magnus

    I’m pleased to hear you enjoy reading our threads. Yes I agree, it does sound nuts.
    There was nothing symbolic about S41 E88, just a convenient and practical way to arrive at fuel exhaustion with the Sun a few degrees above the horizon. Simple and straightforward. Nothing mysterious, no connections with alien civilisations or a desire to take over the world. Just a demented, evil desire to hide his plane as effectively as possible. To some, reality may seem boring but take it from me, it’s really a lot more interesting and frightening than fantasy.

  82. Rob says:

    @DennisW

    coincides with the local mean time of sunrise.

    “local mean time of sunrise”?? Yep, local mean time of sunrise. Haven’t you ever heard of The Equation of Time? I know, but he’s a pilot for heaven’s sakes, not a scientist like you and me 🤓

    Now it’s time for some shuteye, it’s been a very long day.

  83. TBill says:

    @Victor
    “So what’s your frustration?”

    Who me?
    …maybe a little pin proximity paranoia or some other MH370 side-effect?

  84. Victor Iannello says:

    @Andrew said: If the pilots had become incapacitated before the turn back, the aircraft would have continued to fly along the track that was programmed in the FMC, presumably all the way to Beijing. The aircraft would have overflown the last waypoint programmed in the FMC at its previously maintained cruise altitude and then continued flying on its last heading until it eventually ran out of fuel.

    If the plane was in VNAV/LNAV, and if by chance the MCP altitude was dialed down to the airport elevation or less before incapacitation, (and there was no hold in the flight plan), the plane might have crashed on or near the runway.

  85. David says:

    @ALSM. “Regardless of what some ATSB working group members may have believed about the 0019 BFO data accuracy and descent rate at various points in the past, there is no question now.”

    You imply that that ATSB footnote comes from ‘various points in the past’. The footnote is the final documented view of the ATSB (and presumably the SSWG therefore).
    You purport to speak for the ATSB now. Supposing you do, you offer no explanation for that footnote or alternative interpretation of it.

    @Don Thompson. Disingenuous means insincere.
    The fact is that there was a 0:19:37 transmission. The ATSB has stated that it assumes that there was still valid ADIRU data, which led to the descent rates derived from that. If invalid data would halt any transmission why the footnote?

    I think it reasonable just on logical grounds to conclude that the footnote indicated there was a possibility invalid data could be transmitted. (My use of the word ‘corrupt’ was intended to be a synonym for invalid, as in a message being rendered not fully comprehensible in transmission.)

    I believe your admonition to be unwarranted therefore.

    Thank you otherwise for your fulsome explanation as to how unlikely invalid data transmission would be.
    You say, “Without valid ‘IRS’ data the SDU would eventually, implicitly, revert to the Log Off state and the AES would not attempt any further transmission until valid ‘IRS’ data became available.” In effect here you say the footnote was unnecessary/wrong.
    You add, “The possibility of erroneous data propagating through the avionics network is very low.” That leaves it open, if very unlikely.
    Finally, “My view is that such an event is a long way out on the far end of a slope of diminishing possibilities.” That leaves it extremely unlikely

    The 0:19:37 BFO is the basis for search width so its validity is dead important, moreso IMO than piloted-or-not and IFE non-connection. However it is entirely reasonable to suppose currently that search width is sound just as you and ALSM indicate and I have implied.

    It is entirely reasonable also to suppose there to be a basis for the ATSB reservation until that is withdrawn or explained, it being their extant
    position.

    I carry this contradiction forward.

  86. Mick Gilbert says:

    Regarding Seabed Constructor’s departure from the search area, it is worth noting that severe tropical cyclone Marcus is barrelling out of the Timor Sea and down into the Southern Indian Ocean. Based on these synoptics I wouldn’t bd surprised if SC has an extended stay in Fremantle this time around.

  87. Victor Iannello says:

    @David: Is it possible that for extreme attitudes or high accelerations, the output values of the ADIRU become out of range, and not valid? I seem to remember reading somewhere that this has occurred in the past in other accident scenarios. However, even if this occurred, it’s still likely a crash occurred soon after.

  88. DennisW says:

    @David

    I remain confused by your reservations concerning the ATSB actions relative to the 00:19:XX events. My understanding was there was never a lack of consensus, but rather not all of the parties had a chance to respond before the consensus statement was published.

  89. J Magnus says:

    @Rob

    Thanks for your response.

    To be clear, I am not agreeing with S41 E88, but rather the concept of reaching a southern latitude that is near S40.

    If it helps, I charted great circle routes between the below coordinates:

    10.1831 N 90.2245 E (simulator)
    45.1277 S 104.1408 E (simulator)
    10.1831 N 90.2245 E (simulator)
    38.90719 S 102.96312 E (antipodal DC)

    I totally agree that it sounds like tin foil hat stuff. I stumbled across this antipodal coordinate theory doing an unrelated project, and was struck by these unusual locations in the SIO. I actually wish I did not see them, and sincerely hesitated to post anything, given the reception I expected; however, I figure in a year or two’s time, if nothing turns up it might be worth exploring new ideas like this. Thanks for taking the time to read, and being open minded. I was just hoping this might be thought provoking.

  90. Andrew says:

    @Peter Norton

    RE: “Andrew, the information is from the B777 manual, page 64 (Sec. 2.0 Page 15).
    Sorry I didn’t mention that. I assumed that as a B777 Captain you would know.”

    I fly Airbus these days and haven’t flown the B777 for some years. I usually need to refer back to the manuals to check the specifics. The document in the link you posted is an old Continental manual that has been heavily customised for the airline. The Continental -200ERs are powered by GE90 engines. The DUAL ENG/FAIL STALL checklist for those aircraft stipulates a speed above 270 knots to facilitate a windmill start, the same as the -300ER. The same checklist for the RR Trent-powered B777s stipulates a speed above 250 knots. I can’t post a link, but the checklist simply says: ‘Set airspeed above 250 knots’.

  91. David says:

    @Victor. “….there was never a lack of consensus, but rather not all of the parties had a chance to respond before the consensus statement was published.” That happened in mid 2016. The footnote is October 2017.

  92. David says:

    In my 9:22 PM, for @Victor please read @DennisW.

  93. Andrew says:

    @DennisW

    RE: “My understanding was there was never a lack of consensus, but rather not all of the parties had a chance to respond before the consensus statement was published.”

    That’s my understanding of the events too and I really don’t understand what all the fuss is about. The ATSB published a comment in haste. They were subsequently advised of their error, whereupon the comment was amended. They later declared that a consensus had been reached, so what’s the big problem?

  94. Peter Norton says:

    @Andrew:
    sorry, I didn’t know there were substantial differences between the manual versions.
    The reason I chose this version is that it included the exact text Gysbreght quoted.
    Anyway, you all seem to agree that his theory has no merit.
    (PS: seeing my text you quoted, I realize it might be seen as impolite. I hope you didn’t take it that way.)

  95. David says:

    @Victor. “Is it possible that for extreme attitudes or high accelerations, the output values of the ADIRU become out of range, and not valid?…. However, even if this occurred, it’s still likely a crash occurred soon after.”

    I had not heard of that. If those were the only circumstances as you say there would be little difference in outcome. As a curiosity, do you recall what the effect was?

    I suppose there could be extreme acceleration and attitude if 0:19:37 was coincident in the aircraft hitting the water but that is a longshot.

    Supposing that attitudes and acccelerations not to be the basis for the ATSB reservations what is? (not ‘was’!) (also, rhetorical!). In the extreme if the data were seriously compromised somehow the aircraft could have been a continuing glide at 0:19:37.

    However accepting @Don Thompson’s view of the extreme unlikelihood of such invalidity, the ATSB reservations would not extend to that, but then I do not know why they are there at all, at least in my and the Pilatus interpretation of them.

  96. sk999 says:

    DennisW,

    It should be noted that Simon Hardy’s “final position” has shifted over time; unfortunately much of the detail is in articles “published” at flightglobal.com, which requires registration. The original “pin on a map” was at 87.4E, 38.082S. That is the position I declared had been thoroughly searched. His youtube video, posted in Nov 2016, defines a triangular zone that appears to be well SW of the original “final position”, and to the extent that I can interpret the coordinates, only a portion of it is in the Fugro scanned region. However, if Simon wants his work to be taken seriously by anyone other than Byron, he really needs to put out a proper written report.

  97. Pilatus says:

    @David,

    “I suppose there could be extreme acceleration and attitude if 0:19:37 was coincident in the aircraft hitting the water but that is a longshot.”

    That is one possibility I considered that would affect the “reasonableness” of the data and therefore make it invalid. Another might be that the aircraft ground speed was below 20 knots. Both could be consistent with the aircraft being in the water at 00:19:37.

  98. DennisW says:

    @sk999

    Your refrenced position is the one I also found after reading your post. Of course, I had the same issues with flightglobal. Never did watch the video. Low bandwidth in the boonies. Agree with the need for a document.

  99. Andrew says:

    @Peter Norton

    RE: “sorry, I didn’t know there were substantial differences between the manual versions.
    The reason I chose this version is that it included the exact text Gysbreght quoted.
    Anyway, you all seem to agree that his theory has no merit.
    (PS: seeing my text you quoted, I realize it might be seen as impolite. I hope you didn’t take it that way.)”

    No problem. It’s the different type of engine that causes the difference in speed. The HP compressor on the GE90 has ten stages, whereas the Trent’s has only six stages. Consequently, the Trent HP compressor is lighter and can achieve sufficient RPM for a windmill start at a slightly lower airspeed than the GE90.

  100. Barry Carlson says:

    @David wrote,
    “I suppose there could be extreme acceleration and attitude if 0:19:37 was coincident in the aircraft hitting the water but that is a longshot.”

    Similarly, most other observations made here are “longshots”, so the odds are rather equalized if the above coincidence is included. If and when MH370 is found a few kilometers east of the Arc, those “longshot” odds will have been shortened considerably.

  101. ventus45 says:

    This ATSB Safety issue number: AO-2008-070-SI-01 reference (from 19 December 2011) which arose from the investigation into the violent (with many injuries) QF-72 “Uncommanded Pitch” incident (Report Here).
    It says in part:-
    “The ADIRU manufacturer advised in November 2010 that it had examined a wide range of possible mechanisms within the LTN-101’s central processing unit (CPU) module that may have produced the air data reference data spikes. Although the exact mechanism could not be identified, it was considering options to improve the robustness of some of the CPU module’s processing activities.”
    Action status: Closed
    Current issue status: Partially addressed
    Status justification: The ADIRU manufacturer has reviewed the potential mechanisms that may have led to the data-spike failure mode and was examining options to improve the robustness of some of the CPU modules processing activities. In addition, some aspects of the ADIRU’s BITE have been enhanced to aid in the detection and management of the failure mode (see other related safety issue).

    What I find disturbing is that the internal software (firmware ?) of modern systems has become so complex that even the designers and programmers do not seem to be able to “nail” some faults, determine their cause with certainty, and fix them – with certainty.

    There is also this NASA paper that was written on the QF-72 incident.

    There was also the 8/1/2005 MAS-124 B-777 9M-MRG – Climb – Uncommanded Pitch Upset out of Perth WA (YPPH) incident. an interesting paper listing 5 serious “Uncommanded Pitch” / LOC incidents.

    Then there are these from PPrune. B777 adiru failure, and B777adiru-lrn

    If there were any pre-existing latent degredations in the system (either known – or unknown – to MAS maintenance), then just one more failure / data spike / whatever (random) could have corrupted the data fed to the SDU.

    My question then is, can we, (with VERY high confidence) totally discount the possibility, that the SDU could have received erronious position data at the 7th arc ?

  102. Barry Carlson says:

    ..err um! Should have been “west” of the Arc.

  103. ventus45 says:

    I see that my initial (Part 1) post of March 19, 2018 at 11:16 pm is awaiting moderation – probably because I forgot about the maximum number of links rule.

    Oh well, just one more (which I left out of the first post).
    The QF-72 (7 October 2008) incident was an A330 (same as AF447).

    In AF447 (9 July 2009) – as we are all very well aware, ACARS transmitted many system messages between event onset up to a few seconds before impact.

    INMARSAT did NOT record BTO’s and BFO’s at the time, and only began doing so, afterwards, specifically because of the great diffculty of locating that crash site.

    So, during the QF-72 violent pitch upset, I assume that ACARS would also have sent many system messages, but the BFO’s and BTO’s would not have been recorded, since the incident predated INMARSAT’s actions to begin recording the BTO’s and BFO’s.

    Given all of that, it could be useful perhaps, to pull the INMARSAT satcom / ACARS logs, of ANY flight that has experienced any form of in flight ADIRU failure, since INMARSAT started recording the BTO’s and BFO’s, and put them under the microscope.

    We might learn something new re how far “corrupt” inertial / position data propogates through the systems, and it’s effect on them, and to the SDU in particular.

  104. David says:

    @Victor. Having just had a look at Byron Bailey’s article I suspect he had drafted that before his first posting here 4 days ago and ran his theme here in a trial run. The article had some differences to that posting such as no mention of RAT power-up as a specific reason for nosing down. One day stand?

    Separately, in your theme at the top you say, “In fact, it could prove challenging to complete with even two additional swings, depending on the weather and how well the eight autonomous underwater vehicles (AUVs) perform. Although not publicly stated, there are indications are that at least one of the AUVs is having technical problems.”
    Should it prove that OI needs to reduce the search area which lies beyond the ATSB’s 25,000 sq km hopefully they will not hold off until, necessarily, it is arc length which then is sacrificed. The highest probability is in general closest to the arc so reducing width early, if necessary, would maximise remaining success probability. If there proves to be any capacity left over, they then could use that for supplementary candidates depending on weather and warm spots remaining.

    Not that the above will not have occurred to them.

  105. Don Thompson says:

    @ventus45,

    It is my understanding that the GES delivered by SED Systems, and commissioned at the Perth teleport in 2013, did include the additional logging of BTO but BFO was a log record item available in the previous GES system.

  106. Jack.H says:

    “The investigator (in realities not lead by Malaysian, but AU, ATSB British & the so-called American’experts’) was totally discarding all those eye witnesses at NORTHERN area. ”

    I m sorry, my last comment regarding the searching area is to point out why the NORTHERN (instead of SOUTHERN) area i.e. 7th arch was not seriously considerable by the so-called lead team

    Another doubt wandering in my mind, WHY no such ‘operating procedures’ were implemented by Malaysian Military Authority when detecting the ghost plane was flying hours crossing their land i.e. Butterworth military Radar point rather than just did ‘fate watching’ maybe as a tv show program at that night on their radar screen? This the first event that we should blame them, Malaysian DCA as the first party need to alert the safety order to the military bases for the communication lost to mh370 after certain procedural period. Secondly, without any information by the DCA, their autonomous SOP must be taken place as a national security exercises when look unidentified object was flying unknown at that night. But their reckless after all had thrown everything clueless difficulties. If they acting wisely by sending a jet fighter just to perform an assesment why was the ‘ghost plane’ was at wrong path, we might seen a different senario, at last most probably if being tracked, last known point would be more accurate. Sadly the respective malaysian minister was also cluelesd during the p.c.session. we were not asking to shoot down the plane, but please go n check it out. Like a petrol car looking out suspicious driver. That’s all we meant by.

  107. Jack.H says:

    Hi David,

    “The investigator (in realities not lead by Malaysian, but AU, ATSB British & the so-called American’experts’) was totally discarding all those eye witnesses at NORTHERN area. ”

    I’m sorry, my last comment regarding the searching area is to point out why the NORTHERN (instead of SOUTHERN) area i.e. 7th arch was not seriously considerable by the so-called lead team

    Another doubt wandering in my mind, WHY no such ‘operating procedures’ were implemented by Malaysian Military Authority when detecting the ‘ghost plane’ was flying hours crossing their land i.e. Butterworth military Radar point rather than just did ‘fate watching’ just like as a tv show program at that night on their radar screen? This was the first event that we should blame them, Malaysian DCA as the first party need to alert the safety order signal to the military bases for the communication lost to mh370 after certain procedural period. Secondly, without any information by the DCA, their autonomous military SOP must able be taken place as a national security exercises when looked out unidentified object was flying unknown at that night. But their reckless after all had thrown everything clueless difficulties. If they acting wisely by sending a jet fighter just to perform an assesment why was the ‘ghost plane’ was flown at wrong path, we might seen a different senario instead, at least most probably if crashed after being tracked, last known point would be more accurate. Sadly the respective malaysian minister was also clueless, too emotional during the p.c.session. we were not asking them to shoot down the plane, but please go n check it out there WHAT ACTUALLY HAPPEN. Like a petrol car looking out suspicious driver. That’s all we meant by.

    *My apology if duplicating this comment, due to correction made. Tq David

  108. ventus45 says:

    @ Don Thompson

    You said:-
    “It is my understanding that the GES delivered by SED Systems, and commissioned at the Perth teleport in 2013, did include the additional logging of BTO but BFO was a log record item available in the previous GES system.”

    So are you saying that BFO recording was a standard part of the system from “way back”, and it was only the addition of recording BTO’s in 2013 ?

    If that “is” the case, then there may be the possibility of “data mining” the QF-72 satcom / ACARS logs after all (if they still exist). Since the FDR data is available (at least it SHOULD BE), it should be possible to analyse the BFO’s from the ACARS messages before, during, and after the ADIRU event, to see if they “square up” with the known physical motion of the aircraft from the FDR data, and crew statements.

    If the BFO’s do match the position / velocity of the aircraft before the event (obviously they should), it would become very interesting if they don’t, either “during” the event, or “after”, or “both”.

    If so, we will have a new lead to persue.

  109. Rob says:

    @VictorI

    “@Don Thompson: Has @Byron Bailey, Larry Vance, or anybody else ever explained how the flaps were extended after fuel exhaustion? As we all know, the RAT supplies hydraulic pressure to “center” flight control surfaces, but not to the flaps, as there is a checkvalve that prevents flow in that direction.”

    Victor, doesn’t seem like it was ever explained to him.

    Flaps can only be extended if main engines or APU are running, although there is apparently a backup electrical system that can do the job, but it is slow and would use up a lot of the battery to do so.

    So it seem safe to assume (my opinion only of course) the high energy ditching was on RAT power only, which accords with finding of retracted flaps. If for the sake of argument a pilot wanted a high energy ditching with minimum debris, and he was assuming the APU wouldn’t run for more than a few minutes after MEFE, if at all, then he would have planned from the start, to descend/glide and ditch under RAT only power. Retracted flaps would also lead to a lot less floating debris, anyway, and would be preferable. Those inboard flaps are sizable pieces of carbon and would be very vulnerable in a ditch attempt any energy level, if extended.

    Just thinking.

  110. Don Thompson says:

    @ventus45

    The 7th Oct 2008 QF72 service was operated by VH-QPA, an Airbus A330-303. What benefit might analysis of that event bring to the investigation of the loss of a B777?

  111. Andrew says:

    @Rob

    RE: “…there is apparently a backup electrical system that can do the job, but it is slow and would use up a lot of the battery to do so.”

    The electric motor that drives the slats & flaps in the secondary and alternate modes is powered by 115V AC from the L Main AC bus. That bus cannot be powered by the battery. The slats and flaps are not available If the aircraft is operating on standby power (ie RAT or battery).

  112. Don Thompson says:

    @David,
    Thank for clarifying your use of the word ‘corrupt’: “My use of the word ‘corrupt’ was intended to be a synonym for invalid, as in a message being rendered not fully comprehensible in transmission.

    I hope I have set out that the possibility of a “message being rendered not fully comprehensible in transmission” is considered in the design of the avionics equipment so as to avoid any effect from such an occurence.

    Adding to @Victor’s comment concerning the validity of computed data words, while correctly distributed, from the ADIRU. Given that the flight test envelope of the B777 likely did not explore all roll and pitch attitudes, the actual performance of the ADIRU while encountering extreme attitudes may expose a very remote possibility of an error in the ADIRU’s processing.

    However, no matter how often discussion returns to such ideas they remain ‘very remote possibilities’.

  113. Victor Iannello says:

    @Rob said:Victor, doesn’t seem like it was ever explained to him.

    It never was satisfactorily explained. Larry Vance and @Byron Bailey (notice I did not put your name on that list) claim the flaps were extended at the time the plane touched the water. That’s not possible on RAT power, as most contributors here understand.

  114. Don Thompson says:

    Concerning flaps (High Lift Control System), reminder…

    Cruise Inhibit

    This function prevents flap extension during cruise. This happens when the airspeed is more than 265 knots or the altitude is above 20,300 feet.

    Additionally, the Load Relief schedule inhibits extension according to airspeed limits. All HLCS functions, of course, are dependent on hydraulic power and 115V AC power.

  115. Victor Iannello says:

    @David: Yes, if the extended area can’t be fully searched, there should be some thought about whether to reduce the distance along the arc or the width. Contributors here might care to chime in.

  116. Richard Jones says:

    TC or ex-TC Marcus still looking to be very disruptive in the search area at least till 25 March http://www.bom.gov.au/australia/charts/4day_col.shtml

  117. Don Thompson says:

    @ventus45, now that your post appeared…

    My question then is, can we, (with VERY high confidence) totally discount the possibility, that the SDU could have received erronious position data at the 7th arc ?

    The 00:19:29 LOR burst and 00:19:37 LOA burst were received by the ground station satisfactorily. No concerns for validity of attitude and position data used to derive HGA antenna steering. The GES determination of BTO, and consequently the 7th arc position, is derived from the burst propagation time. VERY high confidence that the SDU operated as expected.

    Correlation of 00:10:59 and 00:19:ss events, and arc positions, is consistent with the possibilities for distance travelled in those 8 minutes.

    As I described in a reply to @David, even during transmission of the R-ch bursts at 00:19:29 and 00:19:37 the SDU would have received multiple updates for aircraft position and attitude via the ‘IRS’ bus resulting in updates to the doppler pre-compensation function.

    Without full and valid ‘IRS’ updates the SDU could not have selected the IOR Region, processed a Class 3 Log On, or maintained beam steering to I3-F1.

    As @Victor replied on Twitter recently, “we can’t change the speed of light”.

    If you’ve been carefully following the conversation the significant factors contributing to the determined BFO are: the fixed frequency bias applicable after the 18:25 Log On and OCXO oscillator drift, both small; and the uncompensated vertical speed.

  118. Ge Rijn says:

    @VictorI

    RE:”..if the extended area can’t be fully searched, there should be some thought about whether to reduce the distance along the arc or the width.”

    My proposal would be; first search the area from the 7th arc between ~32.2S/33S till ~97E then (when still nothing found ofcourse) reduce the width to the north to ~+/-15Nm to extend as far as possible to the north.

    Main reason to me is I see no specific feature/reason north of Broken Ridge that could motivate a possible recovery and glide/ditch. Only the trenches under Broken Ridge could serve such a motivation imo.

  119. DennisW says:

    @Victor

    Re: Search strategy

    I can’t make sense of scanning full width (+/- 25nm) as the search is extended North. Why not scan a selected arc length (say to 26S) along the arc, and sequentially expand the width? If you postulate a bell shaped probability of terminus distance from the arc, that would result in scanning the highest probability area remaining as a function of time.

  120. Victor Iannello says:

    @Ge Rijn: Thank you.
    @DennisW: The question is whether there is also a bell curve along the arc. Do impact points on the 7th arc at 30S and 26S have equal probabilities?

  121. DennisW says:

    @Victor

    Do impact points on the 7th arc at 30S and 26S have equal probabilities?

    Hard to assign probability as a function of arc latitude. In my judgement it is much flatter than the probability distribution orthogonal to the arc.

  122. TBill says:

    @Victor
    Perhaps the idea of truncating the search is part of my frustration, as there are a lot of voices to skip BR, which for some of us is the main hot spot.

    (1) I would be concerned +-25 is needed to cover uncertainty, especially inside the Arc7 seems to get short-changed sometimes

    (2) After 31.5S all I have is hoping Richard’s spot works out, up until 27S I always liked McMurdo. Up until 31.5 I have a lot of interest. If Site2 Areas1-2 does not produce a hit, I now like 31.71S 96.1E out to 25-nm.

    Thus I suggest continue as planned for Site2 1,2, most of 3. Site3 we can shift if Richard is OK to a centerline approach. Some “hot spots” like NcMurdo might be one approach.

    A coincidence I presume, the Z simulator cases, if we say heading was 180S CMH Fair Weather, pass close to DrB’s 31.5S and the birthday points. I don’t think I can make much sense of that, but for what it’s worth.

  123. Ge Rijn says:

    @VictorI

    To shime in; “The question is whether there is also a bell curve along the arc. Do impact points on the 7th arc at 30S and 26S have equal probabilities?”

    I think the best drift-studies (CSIRO/Griffin, Pattiarachi, Godfrey) also make a kind of bell curve along the arc which show probability decreases (sharply) north of ~30S and south of ~35S.
    In this regard I would say 26S has quite lower probabilty than 30S.

  124. Ge Rijn says:

    @VictorI

    A question;
    Do the Inmarsat-data (BFO/BTO best fits) produce some kind of bell curve between 26S and 35S along the arc?

  125. DennisW says:

    @Victor

    Do the Inmarsat-data (BFO/BTO best fits) produce some kind of bell curve between 26S and 35S along the arc?

    Ouch! My advice is to take the day off.

  126. Richard Godfrey says:

    @DennisW

    I am sitting in a chalet in the Swiss mountains, roaring with laughter at your wit!

    Many thanks for your comment to Victor about taking the day off.

  127. Rob says:

    @VictorI

    Agree with you totally on the flap extension issue. If there is only one thing we can be reasonably confident of, it’s that flaps were stowed on impact.

    1) ATSB found the RH outboard flap was stowed on impact. I can see no reason to doubt or question their finding.

    2) Flameout probably occurred at cruising altitude. If it had happened at low altitude, we wouldn’t still be arguing about the most likely crossing point, because the wreckage would have been found by now. (Personal opinion of course)

    3) The APU was unlikely to have kept running for more than a few minutes, and flap extension requires the APU running. Flaps cannot be extended at anything much above approach speed (load relief)

    The lack of debris from either inboard flap and LH outboard flap (more on that in a minute) most of which would float admirably.

    Which prompts a discussion on the debris. Any theory on end of flight and impact energy/dynamics has to be able to account for the unusual nature of the debris. Items recovered include the inboard section of RH outboard flap, together with two of its associated or adjacent flap fairing 7 components (mating components when on the wing, no less) both of which are missing their outboard sides, and the adjacent flaperon. Two flaperon closing panels, one from each wing, A small heavily damaged section of the RH aileron. Fragments from the RH engine cowl.

    From the LHS, only the aforementioned flaperon closing panel, plus a small piece of the trailing edge outboard section of outboard flap and associated fairing tailcone fragment. The tailcone fragment appears to have punched a hole clean through the outboard flap on impact.

    That the impact was high energy is beyond question, as the internal partition panels attest, but a nose-first, high energy impact would not produce this highly unusual, RH wing flaperon/outboard flap (inboard section) RH engine cowl dominant suite of debris. The aircraft could not have impacted at a low angle of incidence, catching the RH wingtip and cartwheeling. On the contrary, the aircraft appears to have impacted in a relatively flat attitude, right wing down, with low forward velocity. When the wreckage is finally located, the images will show the outboard RH wing destroyed up as far as flap fairing 7, the LH wing showing very little damage, the LH flaperon still attached, the fuselage in three separate pieces: forward section, centre section/wing structure, and tail section.

    The argument that other significant debris items did not get washed ashore is simply untenable. The flap fairing 7 mating parts, parts associated with the flap hinge still attached to the outboard flap section, parts much differing in shape and size, but successfully making the journey to the East African coast, clearly tell us that what we have is pretty much all there is.

    Apologies for rambling but the debris is very important evidence. At the very least, it would appear to give very little support to the uncontrolled impact scenario.

  128. Ge Rijn says:

    @DennisW @Richard Godfrey

    Hope you enjoy yourselfs but it was only a question.
    I’m not into the BFO/BTO-data the level you and others are by far.
    I only read often there are best-fit and lesser-fit BFO/BTO’s along certain flight paths and corresponding latitudes.

    My thinking was this could also show some kind of curve between ~26S and ~35S. Like used in determinating the previous high probability areas:

    https://worldairlinenews.com/2014/10/10/atsb-issues-an-update-for-the-search-of-the-missing-malaysia-airlines-flight-mh-370/

  129. Ge Rijn says:

    And are those curves bell shaped or what?

  130. DennisW says:

    @Ge Rijn

    I was not trying to be mean. Actually your question is difficult, and the answer has evolved over the many years of this mystery. Probably everyone would have a slightly different answer, but I will give you mine.

    Because the BFO has indefinite(1) statistics it is not a reliable differentiator of paths at a level of fine granularity i.e. one can derive many paths to the 7th arc with different BFO “error” characteristics. I put error in quotes to emphasize that it would differ (not really fair to call it an error) from the values logged by Inmarsat, but might still be realistic given the range of differences that are deemed acceptable. Even acceptable in this context will vary from person to person doing the analytics.

    The short answer to your question (for me) is that the BTO/BFO cannot be used to build a 7th arc latitude probability distribution. Victor might have a different opinion, and I would not even be prepared to quarrel with him about it.

    (1)

    Indefinite in this case has two interpretations. If I take five hours of BFO data and compare it to a subsequent five hours of data, the statistics will not be the same. (Not stationary.)

    If I take 20 hours of BFO data and compare it to four individual 5 hour samples those statistics will not be the same either (Not ergodic).

  131. Ge Rijn says:

    In fact looking at this early ATSB-update graphic the whole probability area north of ~35S is allready shown till ~25S.
    With the greatest width between ~36S and ~32S(till ~97E) decreasing to the north and south fading sharply north of ~28S and south of ~36S.

  132. Ge Rijn says:

    @DennisW

    Thanks for your kind and thoughtfull response.

  133. DennisW says:

    @Ge Rijn

    Are those curves bell shaped or what?

    The picture in the link below shows 8 random walks. They were all generated by exactly the same process (i.e. flipping a coin and moving up on heads and down on tails) yet the results are very different. You can say they (the ensemble of random walks) have a zero mean and a variance that grows with time. However, you cannot use any one random walk to infer what a subsequent random walk would look like. In the case of the BFO error, any one set of measurements taken from one flight would not be expected to have a zero mean. Just like any one of the random walks in the link does not have a zero mean.

    https://commons.wikimedia.org/wiki/File:Random_Walk_example.svg

  134. Victor Iannello says:

    @DennisW said: The short answer to your question (for me) is that the BTO/BFO cannot be used to build a 7th arc latitude probability distribution. Victor might have a different opinion, and I would not even be prepared to quarrel with him about it.

    I think our best estimators are the final BTO values, which define the 7th arc (with some small uncertainty), drift models, which determine where along the 7th arc the impact might have occurred, and the final BFO values, which tell us that likely the impact was close to the 7th arc. So, if we had to define a probability distribution along the arc, I would base it on drift models, however imprecise that is.

    Considering the unknowns in flight path between 18:22 and 19:41 (and frankly, I think there is considerable uncertainty even in the position at 18:22), even if we assume automated flight after 19:41 (which we can’t be sure about), there are a large range of possible paths that satisfy the allowable BTO error and what could be the allowable BFO error (we’re not even sure about that). So bundle up all that uncertainty, and it’s not surprising we haven’t yet found the plane.

  135. Ge Rijn says:

    @DennisW

    But what I understand is the BFO/BTO values have limited possible errors.
    Then the random walk cannot be completely random but only random within those limited possible errors. With limited possible outcomes.
    Missing the point?

  136. DennisW says:

    @Ge Rijn

    I cannot think of anything that would limit the possible BFO errors (other than growth over time). BTO, on the other hand, is well-behaved.

  137. Marijan says:

    @Ge Rijn @DennisW

    Last time when I read about it, the growth of BFO error over time had to do with the increasing distance between a plane and a satellite. I am not sure if they were able to determine what was the underlying cause of such behavior, but it is for sure a very interesting thing to think about it.

  138. Ge Rijn says:

    @DennisW

    What about the speed of light, the speed of the plane and the distance of the plane to the satelite over time?

  139. Richard Godfrey says:

    @Ge Rijn

    If there was an active pilot and he had a goal in mind (particular end location or soft landing or …), then surely he would execute that goal with sufficient fuel remaining to ensure he succeeded in achieving his goal. However, we know the end was at fuel exhaustion.

    But we also know:
    1. The plane was set on autopilot on a course into the Southern Indian Ocean.
    2. The plane did not deviate from this course between 19:41 UTC and 00:19 UTC.
    3. The plane flew a constant path until fuel exhaustion.
    4. The plane entered a dive after fuel exhaustion, accelerating at over 0.6g, beyond 15,000 fpm.
    5. The flaps were not extended upon impact.
    6. There were a large number of debris items of which over 28 have been found.
    7. Debris items have been found over a wide area, implying a large number at the start point.
    8. Debris items included interior items, implying a severe impact and hull break.

    All of this is what would happen without an active pilot after 19:41 UTC.

    How do you conclude there was an active pilot?

    Your only argument so far, is that there was an active pilot selecting the track into the Southern Indian Ocean and he did not suddenly disappear after 19:41 UTC. Then what was he doing? Just watching it all happen?

  140. DennisW says:

    @Ge Rijn

    None of the factors you mention above are at all significant.

    @Marijan

    I think you are referring to speculation on the part of the DSTG relative to an example flight (Figure 5.4) in their report. My take on that was they were simply observing an extreme case of SDU oscillator drift.

  141. TBill says:

    @Rob
    It is also possible Byron and yourself are correct in principle, but with some variations. An intentional flight, yes indeed, could have had the plan of hitting water surface exactly at sunrise as the goal. But there are many other possible strategies a pilot or hijacker might have had, such that we are only guessing at what could have been the end game in that scenario. Hitting the ocean at sunrise seems less likely right now.

    I do see possible merit in some kind of professional prioritizing of possible intentional strategies, but that’s the whole discussion that is not happening in public, on a professional level. Reasons for this may include aversion to the topic, or maybe behind the scenes, it has been considered.

  142. DennisW says:

    @Richard

    Like you I think fuel exhaustion is a much bigger clue than I had previously given it credit for. It greatly complicates a controlled entry, and it certainly flies in the face of a preselected terminus. It does support a simple path (autopilot mode) selection at the FMT and letting physics take its course.

  143. Marijan says:

    @DennisW

    Yes, you are right, Figure 5.4 in Bayesian Methods. One component of BFO error has a ‘geographical dependency’, rather than distance dependency as I said.

  144. David says:

    @Rob.You included,”Fragments from the RH engine cowl” among items which came from the right side.

    The Malaysian description of item 6 in its summary is, like yours “Right Hand Engine Fan Cowling” which is ambiguous. In its associated debris examination report it describes it as, “Right Fan Cowl of a B777 aircraft”. That is clearer. It could be from either engine.

    Given the damage to items to its rear it could be argued though that it is more likely to be from the right engine than the left.

  145. Perfect Storm says:

    What is OI’s average speed (area per day) so far compared to the previous search ?

    (If no exact numbers are available, rough estimates are also welcome).
    Thank you.

  146. airlandseaman says:

    Rob: Regarding your comment above, “1) ATSB found the RH outboard flap was stowed on impact. I can see no reason to doubt or question their finding.”

    You are inferring something I don’t think ATSB has ever claimed. ATSB concluded, after an extensive investigation of the flap segment, the flaps were retracted at the time the flap segment separated from the wing. It may have separated at impact, but it may have separated in the air prior to impact. Either way, it was retracted at the time of separation (as it would be at high altitude), thus, we know for certain the plane was not configured for a water landing (ditching).

    The distinction is very important. Separation in flight, due to near Mach1 speeds and high aerodynamic forces would be consistent with the 00:19 BFO data and the many post fuel exhaustion simulations done by Boeing (and me). If it separated in flight, it is more evidence pointing to an POI very close to the 7th arc, independent of the BFO and other evidence pointing to that conclusion.

  147. David says:

    @Pilatus. Thanks to @Ventus45 above:

    http://mh370.radiantphysics.com/2018/03/18/mh370-search-update-mar-18-2018/#comment-13410

    and to @Gysbreght on JW, 20th March, 5:22 AM and @buyerninety on JW, March 20th 8:36 AM, a likely sources of the ATSB footnote is the ADIRU failure in MAS-124 (B-777 9M-MRG) on 1st August, 2005.

    The NASA coverage of that which @Ventus45 refers to makes interesting reading as to the possibility of masking of faults in complex electronic systems in the general (even if ‘extremely unlikely’).

    Also, the ATSB MAS report notes, “During certification, the aircraft manufacturer and ADIRU manufacturer conducted validation and verification tests of the ADIRU systems. All features of the ADIRU navigation OPS were checked, but none of the tests duplicated exactly the elements of the occurrence; an accelerometer failure resulting in high value output, followed by a power cycle, followed by a second large-magnitude accelerometer failure, while maintaining the large value on the first accelerometer.”

    @buyerninety and @Gysbreght also discuss some wording issues in that ATSB footnote.

    However as to ADIRU data invalidity in this instance, @Don Thompson has observed, in his response to Ventus, “Without full and valid ‘IRS’ updates the SDU could not have selected the IOR Region, processed a Class 3 Log On, or maintained beam steering to I3-F1.

  148. Perfect Storm says:

    impressive OI video: http://youtu.be/leyjsTTWL1Y

  149. Donald says:

    @Richard Godfrey

    I challenge you to put forward a probable scenario in which the flight deck ‘transitions’ from active to inactive (i.e deceased) presupposing that the diversion was deliberate with an intent to fly into the SIO?

    I will tell you that the idea that the PIC pre-meditated and pre-determined that he would take his own life immediately after the FMT (I assume this is where you believe the ‘transition’ takes place) is contrary to everything inherent in the nature and character of the mission undertaken and contrary to the personality and psyche of the man who undertook said mission.

    The logic and rationale used to support an inactive flight until fuel exhaustion and impact are simply non-existent…Saying he took his own life because he would be bored for 5 or so hours is to demonstrate a profound deficiency in the understanding of the psychology at play.

    Not meaning to offend, of course.

    *I don’t believe that some unplanned for ‘upset’ on the flight deck resulting in an unintentional ‘ghost flight’ is credible whatsoever, and conversations about such a scenario would clearly not be productive.

  150. DennisW says:

    @Donald

    I think the “mission” was over at the FMT. At that point the primary objective had failed, and the pre-planned “or else” was initiated. Just speculation and my unqualified opinion. Under those circumstances I think it is reasonable for the PIC to bail. It is probably what I would have done. There was no possible way to alter the outcome if an autopilot mode was selected, and the PIC departed i.e. PIC input was no longer required under any conceivable circumstances. Why torture yourself?

    On the other hand, if the flight went toward CI as Jean-Luc et. al. have postulated, then there is no question the PIC would have remained engaged.

  151. Richard Godfrey says:

    @Donald

    You stated “The logic and rationale used to support an inactive flight until fuel exhaustion and impact are simply non-existent…Saying he took his own life because he would be bored for 5 or so hours is to demonstrate a profound deficiency in the understanding of the psychology at play.”

    I only said that my 8 points are consistent with an inactive pilot. You may not like the logic and rationale of my 8 points, but logic and rationale does not got away by saying that it is non-existent. Nowhere did I mention the PIC was bored, so please answer the points that I raised, rather than putting words in my mouth.

    You stated “I will tell you that the idea that the PIC pre-meditated and pre-determined that he would take his own life immediately after the FMT (I assume this is where you believe the ‘transition’ takes place) is contrary to everything inherent in the nature and character of the mission undertaken and contrary to the personality and psyche of the man who undertook said mission.”

    The case of Andreas Lubitz and German Wings disproves what you say. Andreas Lubitz pre-meditated and pre-determined that he would take his own life and everyone else on board:

    “By early March, Lubitz’s thoughts drifted toward death. He searched the Internet for the most efficient means of committing suicide: “producing carbon monoxide”; “drinking gasoline”; “Which poison kills without pain?” On March 18, a Düsseldorf physician wrote a sick-leave note for Lubitz, effective for four days, indicating that Lubitz suffered from “a persistent vision disorder with a thus far unknown origin.” A couple of days later, while at home, a new method of self-extinction took shape in his mind. That evening, March 20, he searched the Internet for information about the locking mechanism on an Airbus A320 cockpit door.

    On March 22, the day before returning to work, Lubitz scribbled “Decision Sunday,” along with the flight code BCN, for Barcelona, on a scrap of notebook paper that was later retrieved from the trash in his apartment. Below that heading, Lubitz listed several options: “[find the] inner will to work and continue to live,” “[deal with] stress and sleeplessness,” “let myself go.” On Monday the 23rd, he flew round trip between Düsseldorf and Berlin, and the pilot who traveled with him recalled that his behavior was completely normal. That night, Kathrin—who has said she was unaware of her lover’s disintegrating mental state—came home late from work, and the couple went grocery shopping together, buying food for the rest of the week. Early the next morning, Lubitz parked his Audi in a lot at Düsseldorf Airport and climbed into the cockpit for the 7 A.M. outbound flight to Barcelona. The black box from that flight shows that while Sondenheimer was out of the cockpit for a moment, Lubitz briefly switched the plane’s automatic pilot to 100 feet, the lowest setting—a test run for the return journey. He switched it back again before any air-traffic controllers had taken notice.”

  152. David says:

    @Donald. Do you not think he would have elected a powered dive, like his predecessors? Even a ditching, powered or not, would be too indefinite/indecisive in outcome wouldn’t it?

    Yet the BFOs indicate there was fuel exhaustion unless he chopped AC power then restored it on his way down or just before. A rationale for that?

  153. Ge Rijn says:

    @Richard Godfrey

    The plane had sufficient fuel to reach the area OI is now searching.
    An active pilot would have had control over this fuel and how to use it or not to use it. He could have opted to crash/ditch the plane without fuel at his will at a destination he chose to do so. Maybe to prevent explosion or fire (detection) on impact.

    1. The plane was set on autopilot on a course into the Southern Indian Ocean.

    Yes, this would be the logical thing to do in any case.

    2. The plane did not deviate from this course between 19:41 UTC and 00:19 UTC.

    This would be expected when flying on A/P. But I think it’s not sure how (and if) the plane was flown after 00:19 and there is room in the data some (minor) deviations could have been made along the route after FMT.

    4. The plane entered a dive after fuel exhaustion, accelerating at over 0.6g, beyond 15,000 fpm.

    Yes, this is a very strong indication but it’s not impossible to recover from such a high speed steep descent. We only have a 8 second snapshot. There would have been time enough to recover. No-one knows for sure what happened after this 8 seconds.

    5. The flaps were not extended upon impact.

    Most probably not but still no conclusive statements are made by the officials (ATSB, France) about this. Then a ditch without deployed flaps is still possible. The impact speed would be higher with a higher energy impact as a result. Maybe this explains the debris even better.

    6. There were a large number of debris items of which over 28 have been found.

    28 pieces is no proof of a large debris field imo rather of a relatively small one. Out of 177 historical drifters that passed the search area 30 beached on African shores and islands. This is comparable with the number of pieces found so far and could give an indication of the initial debris field.

    7. Debris items have been found over a wide area, implying a large number at the start point.

    See above.

    8. Debris items included interior items, implying a severe impact and hull break.

    Yes, I agree.

    RE; “How do you conclude there was an active pilot?”

    First, statistically the chance of this skillfull controlled and planned flight turning into a not pilot controlled flight at or after FMT is very, very remote. Why assume a one in a million chance and ignore the most obvious?

    There is this similator flight path found on ZH’s hard disk with two points at different altitudes (high and low) without fuel in the SIO.

    Quite some debris (outboard flap, flaperon, flap fairings, right side engine cowling, nose gear door) can only be explained with a ditch-like nose-up impact unless all those pieces seperated in the air. Which is highly unlikely imo.

  154. Rob says:

    @David

    Re the RH engine cowl:

    This is my take on it: For me it’s a matter of (perceived) statistical probabilities. Perceived being in brackets to indicate my perception, therefore open to argument. I take the evident preponderance of debris items confined to a particular area of the RH wing, the flaperon & inboard fragment of outboard flap with part of hinge 7 still attached, plus two conforming flap fairing components also belonging to hinge 7, to be statistically significant. Then we have three parts from an engine cowl, one of which is confirmed as from the RHS (and proximal to the area of interest) namely the strake baseplate, one probably from the RHS, ie the ROY piece, and a third which could be from either LHS or RHS.

    Then when I take the lack of corresponding parts from the LHS into account, together with lack of identifiable parts from the RH inboard flap, and the tiny, marmalized piece can of RH aileron, I conclude it statistically probable that the right wing came off much worse than the left, damage progressing from outboard to inboard as the aircraft contacted the water, but most importantly, I see the impacting force as acting upwards rather than from front to back. The flap damage (flap was in stowed position), the two flap fairing parts that are both missing their outboard edges, the flaperon trailing edge damaged but leading edge intact tell me that the impact force was predominantly from below to above, rather than from to back.

    This would also explain how the engine cowl came to be shattered into small fragments.

    The aircraft fell out of the sky, onto its belly, right wing down.

    As you said “Given the damage to items to its rear it could be argued though that it is more likely to be from the right engine than the left.”

  155. Don Thompson says:

    @Ge Rijn & @Rob

    Please desist from inferring that there is some ‘statistical’ basis to the opinions espoused above.

    “statistically the chance of this skillfull controlled and planned flight turning into…”

    and

    “it’s a matter of (perceived) statistical probabilities”.

  156. formula says:

    @ Donald – “I will tell you that the idea that the PIC pre-meditated and pre-determined that he would take his own life…is…contrary to the personality and psyche of the man who undertook said mission”.

    Although let us not overlook that we do not know for sure the identity of the PIC at that point in the flight.

    It would be encouraging to think that the ATSB and others forming the official search had consulted psychologists with a view to gaining some appreciation of the probable behaviors of the person(s) in charge of the aircraft. I accept that nothing conclusive could be expected from such an exercise.

  157. Victor Iannello says:

    @formula said: It would be encouraging to think that the ATSB and others forming the official search had consulted psychologists with a view to gaining some appreciation of the probable behaviors of the person(s) in charge of the aircraft.

    I think the ATSB has recommended and conducted its search with no assumptions about psychology and even criminality. They simply don’t have access to that kind of information, nor was that part of their scope. Unfortunately, the party responsible (Malaysia) has no ability to conduct this part of the investigation in a thorough and unbiased manner because of the political and financial ramifications of particular outcomes. I think that’s one of the reasons we are where we are.

  158. Ge Rijn says:

    @Don Thompson

    It’s just about formulating an argument. Seems clear to me.

    Besides. Are you a co-moderator on this blog?

  159. Rob says:

    @Don Thompson

    “Please desist from inferring that there is some ‘statistical’ basis to the opinions espoused above.”

    Sorry to disappoint you Don, but I have absolutely no intention of desisting from inferring or invoking the relevance of considering statistics as they apply to the debris. Nor will I be cowed by bullying. This is not the first time you have used the bully approach. It doesn’t work, it is a sign of desperation, and an embarrassment to all of us.

    Your camp is running out of runway, to use a cliche. Duncan Steel was one of the founding fathers of the incapacitated crew thesis. He his remarkably quiet on the subject these days.

    The flutter theory is bollocks. The debris is localised enough, specific enough to point unequivocally to belly impact, right wing down. But not belly up as in AF447, but belly up as in a Shah engineered impact intended to sink it with the least collateral damage. That’s why he flew in a straight line at M0.82 at 40,000ft, toward S41 E88 into the remotest area he could see on the map that gave him daylight only once the tanks were empty.

    There could not have been any nose-first impact. The ATSB were under the yoke, still are under the yoke of a Malaysian domestic political imperative nanely “the aircraft must not be found, do any delaying tactics you can think of to make sure the aircraft isn’t found” and you are playing unwittingly into their hands.

    The Malaysians think they have been given a get out of jail free card. They must not be allowed to prevail. OI must be given a fair chance to show they can find the plane, with their technology. At the moment, they don’t have a fair chance.

    Take a look in the mirror and ask yourself, “do I really want the plane found, or would I rather not be proven wrong?”

    The unresponsive pilot theory is dead in the water.

    To you other prevaricators and water muddiers out there, the pilot was on a mission. How can you even think about raising the issue of 5 hours of boredom? Can you possibly imagine anyone being bored at the prospect of his impending and certain death? He wanted to see it through with a steely determination. When you embark on a plan such as this, you a committed to seeing it through, or the whole thing could turn out to be pointless.

  160. Richard Godfrey says:

    @Ge Rijn

    Then please share with us all the link to the statistical data you are relying on.

    We will then be able to follow your argument.

  161. TBill says:

    I am not arguing because we all have our own assumptions, beliefs, and attitudes. Thankfully Victor allows us to co-exist. Mainly I like to ask folks to list their assumptions, so Richard has that covered.

  162. Perfect Storm says:

    RE: What is OI’s average speed (area per day) so far compared to the previous search ?

    What I’ve found thus far:
    • “Ocean Infinity aims to cover the ground much faster than Fugro did. In prior cruises in the Atlantic, the firm has, according to Josh Broussard, its technical director, managed to scan 890 km² a day using 6 AUVs. With 8, Mr Broussard thinks that the new mission will be able to manage 1200 km² a day.”¹

    • “In an operation that went on for 1046 days, [Fugro] scoured more than 120,000 km² ”²

    That would be 115 km²/day for Fugro, but I guess the 1046 include days without search?
    Otherwise OI would be about 10 times faster, which would be impressive.

    If someone has real (not projected) numbers for area searched and (net) search days, I would be thankful.


    ¹ http://www.economist.com/news/science-and-technology/21733399-swarm-submarine-drones-will-scour-depths-plane-fantastical-ship
    ² http://www.mtuc.org.my/malaysia-chooses-american-company-ocean-infinity-to-search-for-mh370

  163. DennisW says:

    @Rob

    To you other prevaricators and water muddiers out there, the pilot was on a mission.

    What do you think the mission objective was and why?

  164. Don Thompson says:

    @Rob,

    Bullying: really? Gimme a break! You’re the one emphatically stating you have all the answers. I’m simply callng out your reference to statistical analysis as a fallacy. Great reply, you certainly set out your stall.

    @Ge Rijn,

    Seems clear to me.” The essence of the problem, only you. If I were a moderator, you wouldn’t be commenting. You, alone, have already exceeded 10% of comment submissions to this post: its nothing but pollution.

    There is a view that ignoring comments from @Rob & @Ge Rijn, as many do, is a preferable approach, I don’t agree.

  165. Ge Rijn says:

    @Richard Godfrey

    You only need one link:

    https://en.wikipedia.org/wiki/Helios_Airways_Flight_522

    It’s the only example of a commercial passenger airliner in the history of aviation that turned from a pilot-controlled flight into a not-pilot-controlled flight in a sort of way you and others suppose happened to MH370.
    I don’t know how many millions of commercial passenger flights did took place throughout the history of aviation. But does it matter?
    Maybe this helps regarding statistics:

    https://data.worldbank.org/indicator/IS.AIR.PSGR

    BTW I’m more interested in a reply on my general content instead of catching flies out of it to bring the content down.
    I replied in a thoughtfull/respectfull way to your comment on content.
    But I suspect you never forgave my comment/critique on your drift-analysis or whatever.

    Remember the plane still isn’t found on the assumptions it was not pilot controlled after FMT and the end-of -flight.
    My theory still stands after ~4 years. And north of ~36S along the Broken Ridge trenches around 32.5S since ~2 years.

    Your assumptions still stand also among others.
    We’ll see.

  166. Shadynuk says:

    @Rob You certainly seem committed to your scenario. That is good. Perhaps someone like OI will take an interest.

    Perhaps you will help me understand a few points.

    1. You (and others) believe that the Malaysians do not want the wreckage found. Why not?

    2. You believe that Shah wanted to ‘hide’ the plane. Why? I assume it is something more substantial than just wishing to ‘embarrass’ the Malaysians (they don’t seem embarrassed).

    (Odd that these two parties should have at least partially a common objective.)

    3. You believe that Shah attempted to time fuel exhaustion the arrival of daylight. Do you believe that he attempted to ‘to sink it with the least collateral damage’ by executing a ditching with no power to the engines?

    4. And last – whatever Shah’s mission – do you think he accomplished it?

    Thanks.

  167. Richard Godfrey says:

    @Perfect Storm

    If you accept the latest Malaysian report, then 29,500 sq km have been covered so far in 38 search days, averaging 776 sq km per day. Some days have been lost, due to equipment problems, so this average could improve.

    I estimate Ocean Infinity have covered more search area, but this may be discounted by the Malaysians as outside the agreed search area. At least Ocean Infinity is being very thorough to ensure they cover the agreed search area. Ocean Infinity appear to be only using 7 AUVs currently, but despite this at the peak covered 1,143 sq km per day. Ocean Infinity are able to recycle the AUVs for deployment very fast and are able to keep up a rolling deployment of an AUV launch every 8 to 10 hours.

    Each AUV at 3.7 knots (6.85 kmh) covers a track 404 km in 59 hours. With a track spacing of 1.8 km, this represents a coverage of 728 sq km per AUV. Based on 5 AUVs in the water in parallel, in any 59 hour period, the theoretical maximum is 1,480 sq km per day (728 * 5 / 59 * 24), if I got my sums right. I am sure @Richard Cole has a more accurate and detailed view, than I do.

    The search area up until now has been divided between a N.W. and a S.E. side. After completion of the ATSB First Principles Search Area, the search area is not divided and this should help speed things up. However, the Diamantina Escarpment may require a slower AUV speed, as @Don Thompson has already pointed out.

    In general, Ocean Infinity are on track to achieve their agreed search area in 90 search days, barring major weather or equipment problems that might arise.

  168. Don Thompson says:

    @Perfect Storm

    The MH370 OPERATIONAL SEARCH UPDATE #8 states that the “total area covered as of 18 March 2018 is 29,500 square kilometres“.

    Seabed Constructor’s AUVs have worked 36 productive days up to 18th March. By considering only productive days I exclude the February transit to and from Fremantle, and the subsequent days apparently taken by Constructor to work equipment issues (late on 30 Jan thru 19 Feb).

    By that measure, the average is 819.4km² coverage per day.

    ATSB’s requirement of Fugro was to cover 5000km² per month so as to complete the initial 60,000km² in 12 months. Fugro’s vessels managed about 120km² per day, each. The ATSB Operational Report includes summaries of the work carried out by Fugro’s vessels.

  169. Richard Godfrey says:

    @Ge Rijn

    Many thanks for your various responses. I understand what you are saying is that a pilot murder/suicide is very rare and that a pilot murder/suicide, where there is no active pilot to the end, is almost never heard of in the history of aviation.

    In which case, my 8 points look very thin compared with your point that MH370 without an active pilot to the end would make MH370 a unique case in aviation history.

    I guess we will only know when the plane is found.

  170. Perfect Storm says:

    @Richard Godfrey
    @Don Thompson:

    thank you so much for this very interesting, helpful and incredible detailed information.

    I remember when I first read Victor’s blog article about the possibility of a bunch of AUVs communicating with each other, I thought that would be theoretically possible at a proof-of-concept stage but not yet quite ripe for deployment in real-world operations. It’s humbling to see how wrong I was and stunning to watch science in progress literally before our eyes.

    @Richard Godfrey, @Richard Cole, @Don Thompson and many others: thank you for bringing this faraway operation at the other end of the world a little closer to the rest of us (and also to those who knew the passengers).

  171. HB says:

    Regarding Item 7 of Richard Godfrey’s list, i am wondering whether this is still correct taking into account the time history of debris discovery. One would have expected the discovery frequency to gradually increase and plateau for a while before going down. Instead, there has been a short peak and no recent discovery. As time passes, it is increasingly surprising not to see more debris. Composites must have shattered and even if water logged these should float. A significant number of additional debris should have landed by now. Interested to see the view from others.
    I also think it would be of benefit to organise a systematic search for debris.

  172. Victor Iannello says:

    @Perfect Storm said: I remember when I first read Victor’s blog article about the possibility of a bunch of AUVs communicating with each other, I thought that would be theoretically possible at a proof-of-concept stage but not yet quite ripe for deployment in real-world operations.

    I had the same reaction about using a team of AUVs before learning about Ocean Infinity. The Virginia Tech approach of using small, low-cost AUVs, whether or not the “collaborative navigation algorithms are employed, still offers promise for scanning vast expanses of seabed in the future.

  173. Ge Rijn says:

    @Richard Godfrey

    Yes thanks. That is one of the major points among the others I mentioned.
    And yes, we’ll only know if the plane is found. And probably even more if the plane is not found.

  174. DennisW says:

    @all (I guess)

    Some thoughts on statistics relative to MH370.

    Start with debris finds.

    Some time ago I looked at this process using both Weibull and Poisson stats. Poisson is quite familiar to people across many fields, and is a favorite among professional “sports bets” (people who make money by setting odds and taking bets on sporting events). Weibull is more obscure, and is used mostly by people in the reliability domain. Link below to the Weibull cumulative probability distribution based on the first 10 pieces of debris found after flaperon on 8/15. The picture has sufficient resolution to expand and read the accompanying text.

    https://photos.app.goo.gl/CYxo4DeFyOBNI7xp2

    Obviously Weibull cannot know anything about the level of search activity, and how debris might decay over time. All it knows is what has happened, and what is likely to happen in the future given a relatively constant physical and circumstantial environment.

    Read the text, and draw your own conclusions. Personally I am not terribly surprised by the debris find history. I should update this, but my enthusiasm is low at the moment.

    Flight path statistics

    IMO this is an inappropiate use of Bayesian techniques. Bayes works great if the statistics governing the inputs are well characterized. This was not case for MH370. A lot of guess work went into weighting the statistical inputs both in the case of MH370 as well as AF447. It worked for AF447. It did not work for MH370.

    Pilot vs No Pilot after 19:41

    I put this one in the Taleb “Black Swan” category. It is like the Japanese designing a nuclear power plant to survive the largest magnitude earthquake ever recorded in the region of construction only to have a larger earthquake occur and wreak havoc (a Taleb example). You cannot apply statistics to singularities.

    About all you can do, my opinion of course, is to look at the circumstances and try to make sense of them. I have my biases and others have different biases.

  175. Ge Rijn says:

    @HB

    Then you have to referre to @DennisW’s Weibull-statistics.
    This shows a specific amount of debris to wash ashore over a lenght of time depending on the initial amount of debris from a specific starting point.

    But @Dennis can explain this much better than I can.

  176. Ge Rijn says:

    @DennisW

    Are you telephatic or am I? Our comments crossed.

  177. Victor Iannello says:

    @DennisW said: It [Bayesian analysis] worked for AF447.

    As you know, what worked to find AF447 was to consider that both ULBs either were not functioning or were not heard when that area was initially searched for pings, which some thought was improbable, and therefore not considered. So Metron is the hero for considering this possibility, and to assign probabilities on the failure to detect the ULBs that were not based on empirical evidence.

    Now extend that to MH370. I do agree with you that for MH370 there are so many assumptions with probabilities that can’t be substantiated that Bayesian analysis is of limited value.

  178. Richard Godfrey says:

    @DennisW, @Ge Rijn,

    For points 6 and 7 above, where I conclude a large number of floating debris items at the MH370 point of impact, I was basing this on the total surface area of the items found as a very small proportion of the total surface area of the aircraft, given that a similar distribution as AF447 of floating debris items vs debris items that sunk to the sea floor. There must have been thousands of floating debris items to start with.

    I was also mindful of the probability of a floating debris item actually beaching, from Henrik Rydberg’s excellent paper dated 5th August 2015:

    http://bitmath.org/mh370/debris-origin.pdf

  179. Neville Macaulife says:

    I don’t have the expertise of the other commentators, but the “obvious” explanation for the crash involves the crew leaving the cockpit to deal with a problem that escalates into a full scale emergency, which leaves all on board dead from smoke inhalation and/or a pressure breach. Is there any evidence to support this conjecture?

  180. Ge Rijn says:

    @Richard Godfrey

    Imo only 28 pieces found after 4 years at a tremendous area of possible (and confimed) beechings is rather indicating a small initial debris field.
    If there had been thousands of floating debris at the start, a lot more pieces should have been found by now.

  181. DennisW says:

    @Ge Rijn

    If there had been thousands of floating debris at the start, a lot more pieces should have been found by now.

    I think you need to be careful here. Henrik’s paper (which I did read early on) is indeed very good, but it is hard to infer the size of the initial debris field based on pieces of debris found. The missing variable is the probability that pieces of beached debris will be found over time. Henrik’s paper addressed the fraction that will reach shore over time. You say there are 28 pieces. I won’t check that number, but simply assume it is valid. How many pieces of beached debris do you estimate have not been found? Or said another way, what percentage of beached debris do you think has been found? Found in this context means not only seen by someone, but picked up and associated with 9M-MRO.

  182. Marijan says:

    @all

    Fremantle port webpage states that the next SC port will be Durban, not Fremantle, which implies one more swing.

    https://www3.fremantleports.com.au/VTMIS/dashb.ashx?db=fmp.public&btn=ExpectedMovements

    Does anyone have more info on this and can this be considered reliable?

  183. Victor Iannello says:

    @Marijan: After I you contacted me about this and I asked you to post it, I was advised that Durban was listed as the next port the last time Seabed Constructor left Fremantle. This may not be a reliable indicator.

  184. Marijan says:

    @Victor

    Thank you for the clarification. It is a big relief.

  185. DrB says:

    @Neville Macaulife,

    You said: “I don’t have the expertise of the other commentators, but the “obvious” explanation for the crash involves the crew leaving the cockpit to deal with a problem that escalates into a full scale emergency, which leaves all on board dead from smoke inhalation and/or a pressure breach. Is there any evidence to support this conjecture?”

    The evidence supporting this scenario is:

    1. The aircraft crashed and all aboard are deceased.
    2. The post-Penang route involved multiple turns without a clear destination. This may be evidence of the flight crew being distracted by onboard events.
    3. There appear to be two key points in time at which the “flight strategy” changed (diversion at 17:21 and power restoration at 18:22). This could be interpreted as a cascading sequence of onboard failures.
    4. An oxygen fire and hull breach occurred previously in a B777 while at the gate.

    The evidence against this scenario is:

    1. No descent was made for a landing or for keeping passengers alive after a depressurization.
    2. The aircraft continued past the primary alternate airfield out the Strait of Malacca.

    Others can add to these lists.

  186. airlandseaman says:

    Ignore the Durban Port listed on Voyager. That is old, stale info. The next destination is “SEA” (not Durban). There will be at least one more trip out after this 3rd one, unless they find it during the 3 trip.

  187. Shadynuk says:

    @Neville Macaulife It seems many of the ‘experts’ have long since abandoned any notion that this event could have been caused by a ‘mechanical failure’ – which includes a fire in my view.

    However, since I am not an expert, I am free to speculate and be wrong and subsequently change my views.

    Some have speculated that MH370 was carrying illicit cargo. For example, such cargo could have been lithium batteries. These could have been improperly packaged, been of substandard quality which could have resulted in an intense fire with associated toxic gases.

    This could explain:

    1. The Malaysians seemed to know very soon about what had happened.
    2. The Malaysians don’t seem to want the wreckage found. Surely evidence of such a fire would be present.
    3. The Chinese (the recipients of the speculated illicit cargo) have been less than enthusiastic in resolving the matter.
    4. The aircraft made an abrupt turn about.
    5. Communications and other electronics were lost quickly.
    6. It does not appear that the crew attempted an emergency landing. Perhaps they feared complete lose of control over a populated area. As far as their track is known, it is largely over water.
    7. Someone could have been trying to restore communications and failed except to turn on the SDU which provided the ability to track some aspects of the flight.
    8. The crew was not able to call a Mayday or even squawk 7600 on the transponder. It seems there may have been an attempt at a cell phone contact near Penang.

    How did the aircraft continue to fly for many hours, apparently pretty much due south, after its last known position – no idea.

    We would have to assume that at that time they had no control. With even very marginal control perhaps they could have manoeuvred to a heading that would place them closer to land with the possibility of being seen, possible rescue following a ditching or ultimately at least easier recovery. Perhaps they did not continue with a southerly heading for those hours. The wreckage could be a lot closer to land.

  188. DennisW says:

    @Shadynuk

    The wreckage could be a lot closer to land.

    The wreckage is very close to the 7th arc. There is virtually no doubt about that. It is not even debatable, IMO. The place for that thinking is over on JW. You will not get much of an ear here.

    As far as some sort of mechanical failure is concerned we have been through every conceivable variation on that theme – exploding nose wheels, exploding lithium batteries, melting windshields,… it goes on and on and on… Every one of those scenarios requires such a convoluted sequence of events that it defies credulity.

  189. ventus45 says:

    Seabed Constructor is now secured alongside AMC4 – Moored heading 269 degrees, ie, West, meaning that she is berthed PORT SIDE to the Wharf.

    Do we know which of her two ROV’s was housed on the Port side (the other is housed on the Starboard side).

  190. GordoM says:

    @DennisW “… Or said another way, what percentage of beached debris do you think has been found? Found in this context means not only seen by someone, but picked up and associated with 9M-MRO.”

    Yes, upon the flaperon being found in 2015, other locals on Reunion Island claimed they had previously found debris, including one man whose job it was to burn washed up flotsam along a 2km sector of coastline. Presumably other locals had similar jobs around the island. If so, it is likely smaller unrecognisable pieces were destroyed.

    https://www.brisbanetimes.com.au/world/mh370-search-plane-seat-washed-up-on-reunion-island-in-may-20150802-gipl4t.html

  191. Mick Gilbert says:

    @ventus45

    AMC4 is the HeavyLift/Maintenance Berth on the north side of the Australian Marine Complex facility. I think you’ll find that Seabed Constructor is berthed starboard side to.

  192. Andrew says:

    @Shadynuk

    The fact the aircraft kept flying for many hours counts against a serious fire. In other accidents where lithium batteries are known or suspected to have been a factor, the aircraft suffered severe damage and the crews were overwhelmed very quickly after the first indications of fire. In the UPS6 accident the cockpit filled with smoke and there was severe damage to the flight control systems “less than three minutes after the first warning to the crew”. The aircraft crashed about 25 minutes later. In the OZ991 accident, the aircraft crashed some 17 minutes after the first smoke warning.

    The following reports of those two accidents are somewhat sobering:

    UPS6 Final Report

    OZ991 Final Report

  193. Andrew says:

    @Shadynuk

    Further to my comments above, the generally accepted ‘wisdom’ is that a crew might only have about 15 minutes to get the aircraft on the ground in the event of an uncontrollable fire. That would mean an immediate descent and possibly even a forced landing or a ditching if there were no airfield nearby.

  194. Mick Gilbert says:

    @DennisW

    Dennis, thank you for the precis of applicable statistical methods in your recent post.

    I think that your ‘black swan’ observation is particularly pertinent. Whatever the cause, MH370 represents a first of type (or sub-type) event. The history of aviation safety shows that first of type events are generally confounding even when investigators have a wealth of evidence to pore over; the B737 rudder hard-over accidents of the 1990s immediately spring to mind. For MH370 we have very, very sparse evidence and yet, from some quarters at least, there is a headlong and sometimes breathless rush to unconditionally declarative specificity. At least one contributor here repeatedly restates the same unconditionally declaratively specific statements on matters that are simply unknowable, untestable and unverifiable. And it is interesting, if not instructive, that the authors of the unconditionally declaratively specific generally brook neither dissent nor discussion. I can’t help but be reminded of the sort of discourse that one might have with a parrot.

  195. Mick Gilbert says:

    @ventus45

    My mistake, you’re right, she is berthed port side to. My apologies.

  196. Don Thompson says:

    @Mick Gilbert,

    In order to make a quick turnaround it might be a worthwhile consideration to containerise supplies to be loaded, and waste being offloaded: single lift on, single off. Constructor’s own crane is on the port side, now nearest to the dock.

    Work was carried out into the starboard ROV hangar during Constructor’s first AMC port call.

  197. Andrew says:

    @Mick Gilbert

    RE: “I can’t help but be reminded of the sort of discourse that one might have with a parrot.”

    As my teenage daughter would say, ‘ROTFLMAO’!

  198. Mick Gilbert says:

    @Don Thompson

    Don, do you think that they will be in a hurry to put back to sea given the weather? Tropical cyclone Marcus continues its march down into the SIO. On current forecasts it (or the remnant low pressure region) will be pretty much over the search area in three days time.

  199. Brian Anderson says:

    @ Rob,

    “The flutter theory is bollocks”.

    That sort of statement makes me wonder if this is the same parrot, “bollocks” being the very scientific descriptor that a parrot might use.

    Actually I wonder of the parrot actually knows anything about aerodynamic flutter. [Note, not aeroelastic flutter which manifests itself in a different manner. We are not talking here about flexing of the whole wing in some sort of natural frequency stimulated elastic motion.]

    There are a few examples of aerodynamic flutter that one can find with a G… search, but this one helps illustrate the point quite nicely:

    https://www.ntsb.gov/news/press-releases/Pages/PR20120827.aspx

    This case illustrates how a very small flying surface can begin oscillating rapidly, and at quite a high frequency, to the point of complete destruction in the matter of a few seconds.

    When I first saw the damage to the trailing edge of the flaperon, with the tear right along the line of fasteners attaching the carbon fibre skin to the rear spar, I immediately thought of flutter. The bending moment of a fluttering flaperon would likely be close to a maximum along this line, and it is exactly along this line that one might expect the initial failure to occur. The damage could occur within a few seconds, and be followed by the whole flaperon quickly wrenching itself free of the mounting hinges.

    It is noteable too, that the possibility of flutter occuring to the flaperon is mentioned at least twice in the Boeing 777 training manual, while talking about aileron and flaperon PCU modes. “With the flaperon PCUs in bypass mode, at low airspeed the flaperons droop and can possibly flutter”, and “During an engine test run on the ground, with thrust near takeoff level, the flaperon behind the engine can possibly flutter”.

    If flutter can occur in these situations then I think it is entirely possible that flutter might also occur at some speed outside the normal operating envelope, especially if there was wear in hinge pivots that one might expect on an older airframe.

  200. Barry Carlson says:

    @Mick Gilbert,

    The TC currently slow moving near 16S 106E will deepen and slowly move SxW. Once it becomes ex-tropical at 25S, it will possibly head south and fill/dissipate fairly rapidly.

    That it will expedite the SC’s passage back to the search area, with the SE trades backing to easterlies and may be getting up to 30 knots for a short time on Saturday. By Sunday, weather should be back to normal with an incoming anticyclone.

  201. Mick Gilbert says:

    @Barry Carlson

    Thank you for that clarification, Barry.

  202. HB says:

    @DennisW
    I have not seen Henrik’s paper nor you work. It would be good to have a look. You are correct, the analysis should focus on debris of same size ie same travel speed.

    My thinking is that the Weibull analysis is good to model variation in discovery rates assuming a varying discovery rate over time. A tri weibull parameter analysis will best fit the search strategy ie no strategy initially, had hoc search phase, no search strategy with general awareness to look for debris. Taking this into account, the discovery rate should be at least equal or more that the initial discovery rate that brought up a couple of debris in about one year. Saying that the initial debris travel time should be discounted.
    Saying that, on top of my head, it is not easy to inform the initial number of debris from the Weibull analysis. Happy to see the work done so far.
    An alternative approach would be to use the chi-square approach (one deg of freedom) to inform the initial number of debris with different confidence levels but the time phases would have to be split manually assuming a constant rate for each phase. This approach would be more straight forward, although coarse, to estimate the initial number of debris. At least an order of magnitude could be inferred at different confidence levels. This approach is commonly used in failure statistics where there is a low number of failures or no-failure at all in a given item size x test time. I was thinking the same could apply to discovery rate or debris beaching time. Once this is established a montecarlo simulation with exponentional distribution could be use to predict the time history of future discoveries.
    I personnaly used all methods before. You are right to say it is time consuming.
    Intituively, the rate of discovery now should be similar to the initial rate of discovery which makes a low problability at significant confidence levels of not finding more debris in the recent 2+ year timeframe if the initial debris volume was large.

  203. Shadynuk says:

    @Dennis @Andrew Thanks for the comments guys. More than I expected to get back.

    Two lithium fires is not a very large sample set from which to characterize lithium fires in general. But I hear you – very unlikely.

    When I said ‘closer to land’ I meant somewhere between Christmas Island and West Java – near the 7th arc but not quite in Kazakhstan.

    I think my problem is that I have difficulty accepting scenarios that are presented without any consideration to a plausible motivation or that offer no explanation of the peculiar behavior of some of the parties. I believe motivations and behaviors can indicate something about the flight path.

    I will look through those crash reports. Perhaps I shouldn’t – I may never fly again. I built a sailplane years ago (since sold) and with a friend built a small powered plane so I have a natural interest in this.

  204. ventus45 says:

    @ Mick Gilbert
    No problem.

    @All
    For Info: AMC

  205. DennisW says:

    @HB

    I used a two parameter Weibull not three. It seemed to be a reasonable fit so I left it at that. At the end of the day Weibull is a curve fit. It has no relationship to any underlying physics, so my repsect for it is based purely on its historical accuracy.

    Poisson, on the other hand, feels good to me (and it agrees closely with Weibull). Earlier Ge Rijn claimed 28 pieces of MH370 debris were found. I did not check confirmation status. Poisson (based on the initial discovery rate) predicts the following (~30% chance of finding a piece of debris in any given month and ~10% chance of finding two pieces in any given month).

    https://photos.app.goo.gl/v06vL8AuIcdc0frQ2

    So in the 31 month period following the find on ReUnion we should expect ~3 months of two debris finds and 28 months of one debris finds. That would be a total of 34 pieces found. If Ge Rijn’s 28 piece number is correct, we are off the pace but not drastically off the pace considering Gibson is no longer involved.

    Using my own estimate of the probability of beached debris being found and identified, and Hendrik’s beaching statistics (extended since he only went to 20 months) I estimate an intial debris field of 2000+ pieces. I am not qualified to comment on what that means relative to impact parameters.

  206. William Shea says:

    The South China Sea is contested by many Soverein Nations. Many.

    Each interested Nation has many airborne assets providing surveillance. Without knowledge of ATC.

  207. ventus45 says:

    Seabed Constructor is now refueling from the Tanker Vacamonte which is moored alongside facing east (ships are Starboard to Starboard).

  208. DrB says:

    @Andrew,

    What audible and visual alarms occur in a B777 when there is a significant reduction in cabin air pressure? Do any require a response to be cleared? Could an electronics or power failure disable the alarms?

  209. Ge Rijn says:

    @DennisW

    I took the number of 28 pieces from @Richard Godfrey’s comment; “6. There were a large number of debris items of which over 28 have been found” for convenience in replying to him.
    I think it’s about right depending on what you want to include.

    Your Weibull result is interesting but does it also count with the fact that no debris finds have been reported anymore for more then 24 months now? It seems not to me if you say that in the last 28 months each month one find should be expected.
    Would this change your estimates of an initial debris field?

  210. HB says:

    @Brian Anderson, Rob
    re flutter, without repeating my previous flutter argument like a parrot but assuming flutter as a possibility, it should be clarified that every structural piece has a number of natural resonance modes of oscillation and each mode has a corresponding natural frequency. At low speed, the structural piece can be closed to one oscillation mode and at high speed to another mode. As a matter of fact, it can occur when the speed reduces as it could reach one mode and as the speed increases as it could reach another mode; the fracture line would depend on the mode of oscilation. To bear in mind that, if one assumes flutter, it does not necessarily mean high speed it could be low speed too. Based on the manual reference, the flaperon has clearly a natural frequency mode to avoid at low speed too. Low speed mode have also inherently higher amplitudes which drives the failure mechanism and high speed modes have higher frequency with lower amplitudes which are more prone to fatigue. The apparent plastic torsional deformation of the hinges and the apparent fibre pull out/ lack of delamination may be counter to a fatigue argument (TBC).

    The other clarificaton is that if one assumes flutter, the speed needs to be relatively constant to remain in the natural frequency range. If the speed increases or decreases passing a natural frequency range relatively quickly, flutter damage would not be expected too.

  211. Brian Anderson says:

    @HB,

    I agree. I have observed flutter testing on an aircraft and I have to say I was surprised how various parts of the structure resonated at different frequencies.

    I quoted the reference to illustrate that flutter at high speed is exceedingly dangerous. It can occur very rapidly and create significant structural damage in just a few seconds.

  212. HB says:

    @DennisW
    As a matter of interest, was the initial travelling time discounted? Would you mind posting the input table used and the weibull parameters obtained?

  213. HB says:

    @Brian Anderson,
    Hopefully we will find more about that in the final report. Shame that this information cannot be released earlier.

  214. David says:

    @Rob. About engine cowl debris, you commented that the strake piece came from the right engine, the ‘Roy’ piece, “probably” from there and implied that therefore the fan cowl piece probably would have also.

    That may well be so but of ‘Roy’ the ATSB said, “..there were no significant differentiators on the cowling segment to assist in determining whether the item of debris was from the left or right side of the aircraft, or the inboard or outboard side the cowling.”

    So I do not think that you can use this in support.

  215. Ge Rijn says:

    @Rob

    I was thinking about this also (what @David suggested). The Roy-piece and the fan-cowling piece could as wel come from the left engine.
    Then we have the No 1 flap-fairing-piece which comes from the outer side of the left wing, the left wing outboard-flap trailing edge piece and the lef wing flaperon closing panel.

    Imo this makes it impossible to conclude on which wing maybe impacted first. It rather shows symmetry to me in positions of the pieces over both wings.

  216. Andrew says:

    @DrB

    RE: “What audible and visual alarms occur in a B777 when there is a significant reduction in cabin air pressure? Do any require a response to be cleared? Could an electronics or power failure disable the alarms?”

    There are three warnings if cabin pressure is lost:

    1. A Master WARNING light in front of both the Captain and First Officer.

    2. A ‘CABIN ALTITUDE’ EICAS alert on the upper EICAS.

    3. An aural siren.

    The warnings will not clear unless the pilot(s) press the Master WARNING/CAUTION reset to clear the Master WARNING lights and siren, and the CANC/RCL switch to clear the EICAS warning.

    The Master WARNING lights and aural siren are generated by the warning electronic system (WES), while the EICAS alert is generated by the AIMS cabinets. I think the only common point of failure that might stop all warnings is the remote cabin pressure sensor that supplies the pressure signal.

  217. DennisW says:

    @Ge Rijn

    Yes, my last post was incorrect. Please provide your list of confirmed (and highly likely) finds, and we will start over. My initial Poisson work was done in December of 2016. At that time I counted 8 pieces of debris (3 confirmed and five likely) in the 16 months since the flaperon was found.

    I took my blog down since I transitioned it to personal stuff unrelated to MH370. I copied the Poisson stuff to Google docs. It is pretty trivial. See if this works for you.

    https://docs.google.com/document/d/1ofiQ19Nr1dU23122Rf2g5nFUcnsiLBpg-ngfeuijewY/edit?usp=sharing

  218. Don Thompson says:

    @DrB asked “Could an electronics or power failure disable the alarms?

    The aural alarm, generated by the WES, may be readily inhibited by pulling the circuit breakers for the integrated speaker-amplifier used to sound the aural alerts. The cct breakers for each of the two speakers are located on the flight compartment overhead panel.

  219. Andrew says:

    @Shadynuk

    RE: “Two lithium fires is not a very large sample set from which to characterize lithium fires in general.”

    Fortunately, uncontrolled fires of that nature are rare. Nevertheless, research conducted by the FAA and other organisations suggests that such fires are likely to be catastrophic. A test conducted by the FAA showed the ignited gases from as few as eight 18650* lithium-ion batteries at 50% charge is sufficient to compromise the cargo compartment fire suppression system of an airliner. It’s little wonder that ICAO has now banned the carriage of bulk lithium-ion battery shipments by passenger aircraft.

    *An 18650 cell is slightly bigger than a standard AA battery.

  220. DennisW says:

    @HB

    @DennisW
    As a matter of interest, was the initial travelling time discounted? Would you mind posting the input table used and the weibull parameters obtained?

    HB, I don’t know what you are referring to as “traveling time”. Maybe I am being dense here. Could you rephrase your question? BTW, I would be happy to give you everything I have. The calculations were done in Python.

    The “find array” used is:

    data = np.array([1,4,6,7,7,8,9,10,11,13])

    where the numbers refer to month starting at 8/15 (flaperon) for the 10 pieces of confirmed and likely debris I used at that time.

    The slope/shape parameter obtained is 1.202 and the scale parameter 66.12.

  221. Kenyon says:

    @HB,
    The third parameter for Weibull analysis might be handy if we did not know what season MH370 disappeared in 2014. @DennisW application of Weibull analysis to the discovered debris available at that the time of analysis seems to be spot on to me. Additionally the drop off of debris discovery is measured in week/months. Therefore, after several years have expired, offsetting the Weibull curves to account for debris travel time and or shape coefficients to account for available beach areas vs. human interaction probabilities is beyond the scope of the exercise. Updating a Weibull analysis to include all debris discoveries would be nice to have for the files, please post curve results it you get around to putting it together.

    Tom

  222. Ge Rijn says:

    @DennisW

    I’ll take the MoT report ‘Summery of possible MH370 debris recovered” from 30 april 2017. This counts 27 items. And I would add the small honeycomb panel item found by Blaine Gibson and the vortex generator piece. These are also the final pieces reported to be found (august 2016). So not +24 months but ~20 months ago. This makes 31 total.

    So the Weibull-curve seems to have stopped after august 2016.
    I don’t think an inactive Blaine Gibson can be the only reason given the population density along the coasts of east and south Africa, Madagaskar and the other islands.

    I rather think it indicates not that much more MH370 aircraft items have beached (not considering possible very small items). I would estimate maybe 10 or 20 are still lying around somewhere.

  223. Ge Rijn says:

    small correction.. 29 pieces total (not 31)

  224. TBill says:

    @Andrew @HB
    If flutter damage can happen in a few seconds and at slower speeds, it opens up a lower altitude descent as a possibility.

  225. DennisW says:

    @Kenyon

    Updating a Weibull analysis to include all debris discoveries would be nice to have for the files, please post curve results it you get around to putting it together.

    I’ll do that just for completeness. Won’t have any effect on the search, but I am curious about the debris physics beyond just drift trajectories. I’ll be heading to the Bay Area for the next several days, and will be engaged in serious partying with my daughter and her friends. No computers in the travel kit.

  226. HB says:

    @DennisW
    You answered my question by stating the starting month.

  227. DennisW says:

    @HB

    Now I understand what your question was about. 🙂

  228. Rob says:

    @Brian Anderson

    Thank you for your words of support concerning flutter. These are graciously received. I do have some basic awareness of the mechanisms involved, and the potential results. The parrot has absolutely no knowledge of flutter. His primary and overriding concerns revolve around when the next grape is going to materialise.

    Flutter can be extremely destructive if the initial design didn’t take it duly into account. Aircraft makers have learned lessons from previous mistakes.

    However, I remain unconvinced about flutter being involved in this case. As you know, flutter is taken into account by the manufacturer, in both the airframe and hydraulic actuation system design stages. Also, as you know, the flaperons are worked hard in flight, and are put under stress during takeoff. They are important components of the control system, and hinge/actuator wear is a recognised potential problem area. Accordingly, Boeing have included an adequate factor of safety in the design. Do you know of any Boeing airworthiness directive being issued subsequent to MH370, relating specifically to flaperon wear?

    It’s also been established on this forum that the hydraulic system would remain sufficiently pressurised during the period between engine run down and RAT run up. Flutter should not be expected to have manifested at any stage of this aircraft’s journey between flameout and impact.

    My earlier assessment stands, as a certain well known and highly experienced pilot would say, albeit without the entirely inappropriate and uncalled-for expletive.

  229. HB says:

    @TBill
    I dont know if there us a credible scenario. Pilots would probably know. I would think unlikely.

  230. Richard Godfrey says:

    SC is leaving Fremantle.

  231. Richard Godfrey says:

    SC destination is stated as Offshore.

  232. Kirill Prostyakov says:

    @BFO model users,

    As DennisW arguments about ocxo drift are finally gaining ground, I can’t refrain from asking, why is pair-wise difference between subsequent BFOS not (ever) modelled instead of actual BFO values? Presumably that would kill most of multi-hour-scale random walked bias?

  233. airlandseaman says:

    TBill: Re: “If flutter damage can happen in a few seconds and at slower speeds, it opens up a lower altitude descent as a possibility.” Let’s be clear. The low speed flaperon flutter that the 777 manual refers to is something that can happen during the takeoff roll. It does not pose a significant risk of catastrophic damage. OTOH, the flutter modes that can occur when the airspeed approaches mach1 are potentially catostofic, and the damage can occur within a few seconds of the onset, exactly as Brian and I have described for the last 2 years.

    Rob: Re: “It’s also been established on this forum that the hydraulic system would remain sufficiently pressurised during the period between engine run down and RAT run up. Flutter should not be expected to have manifested at any stage of this aircraft’s journey between flameout and impact.” This statement is incorrect in the premise and the conclusion.

  234. Ge Rijn says:

    @Brian Anderson @Rob

    A parrot is a creature who simply uses and repeats the words of others in a senceless way to please his masters or whatever.
    You can call me anything but a parrot. You should all know by now I’m not interested in parroting anyone. And the fact that I get so much opposition is kind of proof of this. A parrot never gets opposition this way.

    @Rob, your opportunism shows heavily again. I learned to smile about it.

    Back to business.

  235. airlandseaman says:

    Kirill Prostyakov:

    It is important to distinguish between the two types of BFO Bias change that could occur after a prolonged (>1hr) AES power off period.

    One is “OCXO retrace error”, which is a bias error resulting from the power cycle and imperfect return to the nominal frequency after the warm-up transient, and the oven comes into equilibrium. We have very little evidence to quantify the retrace error for the 9M-MRO OCXO, but as Victor found in the MH371 data set, it is possible that an error OTOO 4 Hz could be expected. IF there was some retrace error after the 1825 power cycle, it would result in an offset, more or less constant, to the calibrated BFO Bias (150 Hz), not random walks.

    The other is medium term drift, a random error, which occurs continuously, and independently, from any power cycle. Medium term drift is very small, typically <10 Hz over 24 hrs (referred to L band). This part has been in the math from day one, although estimates of the magnitude have evolved as we learned more. This part leads to the SD estimates.

    Whereas the random component effectively fuzzes up the POI on the 7th arc more or less symmetrically, the bias moves the center of the fuzz up (or down) the arc.

  236. Ge Rijn says:

    @DennisW

    From the MoT report 30 April 2017 I see the final piece found and reported is from october 2016 also found by Blaine Gibson.
    So this reduces the gap of no finds reported to 18/19 months till present.

  237. DennisW says:

    @kirill

    As DennisW arguments about ocxo drift are finally gaining ground, I can’t refrain from asking, why is pair-wise difference between subsequent BFOS not (ever) modelled instead of actual BFO values? Presumably that would kill most of multi-hour-scale random walked bias?

    BFO depends far more heavily on satellite and aircraft motion than oscillator drift.

  238. Ge Rijn says:

    ..ofcourse this an open door to jump into for some..

  239. Ge Rijn says:

    @DennisW

    an add to my previous comment

  240. Rob says:

    @ALSM

    “Whereas the random component effectively fuzzes up the POI on the 7th arc more or less symmetrically, the bias moves the center of the fuzz up (or down) the arc.”

    Absolutely. A perfect summary of the situation. The random component, ie the medium term FFB drift, fuzzes up the POI. Better by far then to rely on the fuzz proof BTO as an indicator of POI. But is the random component really random, or is the FFB more appropriately described as non ergodic? The spread of values as measured over time do not sit neatly under a bell curve. I think Dennis would agree with me on this point.

    Is not the take away from this to treat the BFO values with caution, and to err on the side of caution and give more weight to BTO?

    Just thinking, as John Young would have said.

  241. DrB says:

    @Andrew, @Don Thompson,

    Many thanks for your helpful responses.

    Could I ask you to comment on the accuracy of the following statements regarding the circumstances under which the low-cabin-pressure warnings could be missing altogether or less obvious than normal, so that they might be missed, allowing the flight crew to become hypoxic without being aware of it:

    1. A person said: “Without a Left AIMS Cabinet, 4 of 6 display screens blank. IN reversion mode Pressurization screen is NOT available. Crew can be unaware.”
    2. A different person said: “When smoke develops in the MEC the Equip Cooling Override valve opens cabin pressure duct to outside & silences cockpit alarm. Electrical failure prevents the valve from closing again. Hypoxia is gradual.”
    3. “a Boeing 777 ventilates smoke from electrical fire in the MEC by opening a valve (MEC override valve) which inhibits alarms. Such a fire was more likely in the form of melting contacts and smoke than a blaze, but pilots may well have turned back before hypoxia.”
    4. “. . . in the event of smoke in the MEC when pilots get an ECIAS alert “EQUIP COOLING OVRD” – donning O2 mask is optional & not required yet override valve can cause hypoxia”

  242. Ge Rijn says:

    @Rob

    I would change your last frase to ALSM:

    Just parroting, as @Brian Anderson would have said.

  243. DrB says:

    @All,

    The “flight strategy” changed at 2 or 3 specific times during MH370. There are some surprising (to me, at least) similarities in the events occurring circa 17:21 and circa 18:24. Here is a summary:

    A. What happened at 17:21 Diversion:
    1. Electrical power was lost to the SDU and probably to left AC bus.
    2. Numerous equipment items (transponder, radios, etc.) ceased to function.
    3. Turned around to return to Malaysia.
    4. Altitude changed to even Flight Level.
    5. Speed increased to M0.84.
    6. Navigation mode changed from waypoints to either flying manually or using MCP constant headings.

    B. What happened at 18:24:
    1. Power was restored to SDU and IFE, and possibly to left AC bus.
    2. Contingency Procedures (with no ATC Clearance) implemented:
    a. Establish 15 NM lateral offset.
    b. Probably reduced altitude by 500 feet.
    3. Shortly thereafter, speed was reduced, probably to MRC.
    4. Now navigating normally by airway/waypoints.

    The 17:21 events appear to me to be fairly (but not completely) consistent with either a hijacking or a serious onboard system failure. One exception is the A.6 item. Why would a hijacker pilot revert to flying manually or using the MCP, assuming all the equipment was working normally?

    The 18:24 events all appear to me to be consistent with an accident scenario. B.1 and B.2 do not appear to me to be consistent with a hijacking. Why follow contingency procedures? Why turn the power back on to the SDU?

    These discrepancies appear to me to be serious impediments to the hijacking theory.

    For the accident scenario, the biggest discrepancy is not at 17:21 nor at 18:24. It is at 17:52. Why not land at Penang? So far, the only theory for this lapse is hypoxia. Hypoxia that would cloud judgment at 17:52, sufficient to decide not to descend or land then, but not totally incapacitate the pilots until after circa 19:00. If this is possible, then I would say the evidence we have is more consistent with an accident scenario than a hijacking scenario. Perhaps I have missed plausible explanations for A.6, B.1, and B.2 within the hijacking scenario. I am sure I will get some differing opinions.

  244. airlandseaman says:

    Rob: Re “Is not the take away from this to treat the BFO values with caution, and to err on the side of caution and give more weight to BTO?”.

    I would not characterize it that way. The BTO and BFO values contain different types of information. The BTO values require no assumptions to derive the arcs. Pure math and known facts give accurate arcs with small, well characterized error bars. Nice and easy to calculate, check and verify.

    OTOH, the BFO values represent a sum, the terms of which include (1) a preflight calibrated bias term (150 Hz), (2) a weak but complex horizontal motion residual (due to incomplete aircraft Doppler suppression), (3) a strong vertical motion signal and (4) a combination of small, but important random drift and bias errors. Due to the complexity of the BFO interpretation, analysis is tedious, but doable. I would say we need to exercise special care, not so much caution. To me, “caution” implies that the BFO values may not be trustworthy, and I definitely do not believe that to be the case. They are trustworthy, but the investigator needs to understand the limitations. So, for example, the BFO data provides a very strong, 100% reliable indicator that the plane flew south and not north. It also provides a very strong, unambiguous indicator that the plane was descending very fast at 00:19. But beyond those two relatively certain conclusions, the BFO data can only provide a rough idea of where on the 7th arc to look. 3 years ago, the assumed error bars were smaller than they are today. When we trow in the possibility of a little post 1825 retrace error and some random noise (whether truly random or non ergodic), the BFO guidance on horizontal motion and the POI on the 7th arc is pretty fuzzy…about like the drift analysis.

    As to the question: Is the OCXO noise random or non ergodic, the answer is really dependent on the time frame. That is why XTAL oscillator gurus prefer to talk about short term (seconds to minutes), medium term (hours to a day or so) and long term (many days to years). Short term XTAL oscillator drift is virtually pure random noise. Long term drift tends to go in one direction (say, 1ppm/yr) over several years, usually tapering off to a very slow drift rate after a few years. The medium term is (as the name implies) somewhere in-between. Dennis seems to be emphatic that the process is non ergodic, and I don’t disagree, but for our purposes here, I’m not sure the distinction is all that important. Either way, the “noise” prevents us from saying with any certainty where on the 7th arc MH370 is. OTOH, regardless of whether the noise is random or non ergodic, it is small compared to the signal telling us the plane went south and it went down fast in the end.

  245. Brian Anderson says:

    @Ge Rijn

    My comments about he parrot were not directed at you.

  246. Ge Rijn says:

    @DrB

    Just mu thought. If hypoxia happened before or just after Penang there would not have been a re-logon at ~18:25 and no FMT to the SIO.

  247. DrB says:

    @All,

    Another point I have thought about is the First Officer’s phone connecting to a cell tower at Penang. If it were turned on and located in his shirt or jacket pocket, it would have had a good line of sight path to the ground through the right-hand flight deck windows as the aircraft rolled to the right as it rounded the island. So, I would say that brief connection (with no successful attempt to make a call or to send a SMS message) is consistent with the First Officer being in his normal right seat on the flight deck. If he had been locked out of the cabin (and if we were still alive at 17:52, which he should have been even if the plane were depressurized circa 17:21 because he would know to use an O2 bottle in the cabin), and failing to breach the cockpit door, don’t you think he and any still-alert passengers would be frantically trying to make calls and send text messages while holding their phones to the windows? And wouldn’t this be done when the lights from Penang were visible on the right side of the aircraft? I would add that I would expect the same behavior of the passengers if they felt the plane was about to crash because of problems with the aircraft itself. To me the lack of cell phone connections implies the passengers felt no need to violate the normal rules (and thus the only mobile phone turned on was the First Officer’s). Perhaps they had been told the aircraft was being diverted due to technical issues but everything was under control. Since they could see land beneath them at that time, they knew the plane had already brought them close to an airport, and they remained calm and obedient with their cell phones turned off. In my opinion, the single tower connection is consistent with the First Officer’s phone being turned on and with the passengers’ phones being turned off, but no phones were in active use at 17:52 attempting to make calls or send text messages. Otherwise there would have been more logged traffic. That set of circumstances does not appear to me consistent with a hijacking by the Pilot. If I have got this wrong, please tell me where my logic has failed.

  248. DrB says:

    @airlandseaman,

    Mike, that was an excellent summary of our current understanding of the BFO data.

  249. DrB says:

    @GeRijn,

    You said: “Just mu thought. If hypoxia happened before or just after Penang there would not have been a re-logon at ~18:25 and no FMT to the SIO.”

    First, hypoxia affects people differently, so not everyone is incapacitated at the same time. Second, the onset can be gradual and take quite a while, especially so if the cabin pressure is declining gradually over tens of minutes. That was the heart of my question: Can judgment be impaired at 17:52 but incapacity delayed until roughly 19:00? I don’t know the answer, but I think it is an important question, since the accident scenario seems to hinge on the answer.

  250. Victor Iannello says:

    DrB asked: Why would a hijacker pilot revert to flying manually or using the MCP, assuming all the equipment was working normally?

    The pilot was not flying an LNAV flight plan. Rather, to me it looks like he was navigating by airport using the ND and SEL HDG, while avoiding flying directly over the airports. The pilot was not constrained to fly within a small distance of an airway. Not using waypoints doesn’t necessarily imply inexperience.

    The 18:24 events all appear to me to be consistent with an accident scenario. B.1 and B.2 do not appear to me to be consistent with a hijacking. Why follow contingency procedures?

    You have made numerous assumptions here that could be false. For instance:

    1. We have no confirmation from the Malaysians that the Lido Hotel image captured MH370. In fact, they have explicitly said it does not.

    2. The plane might not have flown a lateral offset. Assuming the Lido Hotel image is correct, there are paths that satisfy the BTO and BFO values for the log-on values that involve turns and changes in speed.

    3. If the Lido Hotel image is not correct, there are straight paths starting from the position at 18:02 that satisfy the BTO and BFO values for the log-on.

    Why turn the power back on to the SDU?

    Many answers have been given before. One that has not gotten much attention is the option to for an fully automated landing.

    Hypoxia that would cloud judgment at 17:52, sufficient to decide not to descend or land then, but not totally incapacitate the pilots until after circa 19:00. If this is possible, then I would say the evidence we have is more consistent with an accident scenario than a hijacking scenario.

    I agree. However, I seriously doubt it’s possible. I’d be interested in any data.

    Of course, there is also the extraordinary coincidence of the simulator data, which has captured snapshots of a flight that proceeds from KLIA, up the Malacca Strait, past the Andamans, and turns to the south, and ends in fuel exhaustion in the deep SIO.

    I have more observations and theories about the simulator data that I might publish at some point.

  251. DennisW says:

    @ALSM

    +1

  252. sk999 says:

    Victor: “We have no confirmation from the Malaysians that the Lido Hotel image captured MH370. In fact, they have explicitly said it does not.”

    Can you provide a reference to where “they” explicitly said that the Lido Hotel image does not capture MH370?

  253. sk999 says:

    R. E. a change in the OCXO FFB after the SDU restart at 18:25 due to retrace error, the possibility must be considered. More than two years ago I generated a set of models that were designed to match the ping rings (continuously adjusting the heading in a smooth fashion, not with heading discontinuities as was done by Inmarsat) without regard for the BFO. I kept a fixed turn at IGOGU at 18:39 and allowed the initial speed to vary. These are not routes that can be flown in any MCP autopilot mode, but can serve a pedagogical purpose. I went back and solved for the FFB for a range of routes, minimizing the overall BFO rms, to see what would happen. For speeds after IGOGU ranging from 360 knots (ending at latitude -26 deg) to 480 knots (ending at latitude -38 deg), the FFB varies from 148.8 to 150.4 hz. That’s a peak-peak change of 1.6 hz. My FFB at the gate is 149.7 hz, right in the middle of the range. So for this particular family of models, the change in FFB after the 18:25 restart from the value at the gate is less than 1 hz.

    One can come up with other routes (whether more whacko than mine is a matter of taste) where the FFB would need to change more.

    One can look at other statistics, such as the slope and/or curvature of BFO residuals v. time, and I did so as well in the prior study. The problem is one of (a) establishing robust criteria for the significance of any measurement, and (b) determining whether any anomaly is caused by a choice of bad route or a flaky oscillator.

  254. Don Thompson says:

    DrB.

    Concerning ‘accuracy

    1) the Primary Display System, controlling all six flight Display Units and the two Cursor Control Devices, relies on only one of the two AIMS cabinets. The alternate AIMS cabinet provides full redundancy in case of failure.

    2, 3 and 4) EQUIP COOLING OVERRIDE:

    The Fwd Equipment Cooling Override Valve is set to ‘override’ so shutting the normal cooling supply path but opening the smoke removal valve allowing airflow to exhaust smoke overboard.

    The override mode cools the equipment without the supply and vent fans. This mode uses cabin differential pressure to cause airflow through the equipment. The override mode gives sufficient cooling when the airplane is pressurized and above 25,000 feet (7625 meters). The flight crew also selects this mode when they want to clear smoke from the flight compartment.

    An important additional effect is – “Cabin differential pressure pushes cabin air through the components and out of the airplane

    While EQUIP COOLING OVERRIDE stops the vent fans provided for the equipment cooling, it does NOT stop the aircon supply to the flight deck. The aircon supply, directly from the left pack, ensures the airflow through the equipment to clear smoke via the smoke removal valve.

    The Override Valve is powered by 28VDC from the Capts Flt Inst bus (redundancy via main battery). Electrical failure of this 28VDC supply is catastrophic, no chance of many hours continued flight.

    The EQUIP COOLING OVERRIDE procedure detailed in the QRH makes no reference to failure of flight deck conditioned air supply nor reference to the Fire, Smoke or Fumes checklist.

    (I detect Gunsian nonsense.)

  255. Don Thompson says:

    @DrB, again

    Concerning

    A. What happened at 17:21 Diversion:
    1. Electrical power was lost to the SDU and probably to left AC bus.
    2. Numerous equipment items (transponder, radios, etc.) ceased to function.

    A.1 may have occurred at anytime after 17:08; the final ACARS communication of the FMS Progress reports.
    A.2 only the transponder ceased to function. The radios may well have been receiving satifactorily.

    You later pose the question, “Why turn the power back on to the SDU?” A: in order to receive a message. It’s possible that the implicit procedures to establish the end-to-end ACARS connectivity and thus enable a simple flight deck comms message to be received by the aircraft was not understood.

  256. airlandseaman says:

    Re retrace error magnitude: Before we get too far out in the weeds over the possibility of some retrace error, I want to report that after some further reading on the subject, mainly papers back in the 80’s (i.e., about the time the AES equipment was being designed), it looks like the retrace error for a well aged OCXO (>1 year old) is likely to be in the range of 2X10^-9 to 5X10^-10. IOW…about 1-3 Hz at L band. So, while it may be significant at the margins to path studies, it has no significance to the determination that the FMT was to the south, or the determination that there was a steep final descent. The retrace error on a 10 year old OCXO is probably closer to 1 Hz.

  257. Andrew says:

    @DrB

    RE: ”Could I ask you to comment on the accuracy of the following statements…”

    (Don has beaten me to it with his reply. My thoughts below.)

    1. A person said: “Without a Left AIMS Cabinet, 4 of 6 display screens blank. In reversion mode Pressurization screen is NOT available. Crew can be unaware.”

    The primary display function that controls the displays is duplicated between the L & R AIMS cabinets. Both AIMS cabinets drive all six displays. The loss of the L AIMS cabinet should not affect the displays.

    2. A different person said: “When smoke develops in the MEC the Equip Cooling Override valve opens cabin pressure duct to outside & silences cockpit alarm. Electrical failure prevents the valve from closing again. Hypoxia is gradual.”

    3. “a Boeing 777 ventilates smoke from electrical fire in the MEC by opening a valve (MEC override valve) which inhibits alarms. Such a fire was more likely in the form of melting contacts and smoke than a blaze, but pilots may well have turned back before hypoxia.”

    This sounds like some nonsense that was peddled by Simon Gunson. The equipment cooling override valve automatically opens if smoke is detected in either the forward equipment cooling system or the forward equipment ventilation system. The smoke then vents overboard due to the cabin differential pressure. However, the air conditioning packs supply more than enough air to replace that escaping through the override valve and the outflow valves continue to regulate the cabin pressure. In short, the cabin will NOT depressurise (Boeing aren’t that stupid!). I don’t know of any alarms that would be silenced (or inhibited) without action by the crew.

    4. “. . . in the event of smoke in the MEC when pilots get an EICAS alert “EQUIP COOLING OVRD” – donning O2 mask is optional & not required yet override valve can cause hypoxia”

    The EQUIP COOLING OVRD checklist does not call for the pilots to don their oxygen masks. However, the cabin will not depressurise, as discussed above. If the pilots were to detect smoke in the cockpit, their first action should be to don their oxygen masks.

  258. Andrew says:

    RE: “For the accident scenario, the biggest discrepancy is not at 17:21 nor at 18:24. It is at 17:52. Why not land at Penang? So far, the only theory for this lapse is hypoxia. Hypoxia that would cloud judgment at 17:52, sufficient to decide not to descend or land then, but not totally incapacitate the pilots until after circa 19:00.”

    If the crew turned back with the intention of landing at Penang, then they should have started descending at about the time the aircraft crossed the coast near Kota Bharu (17:38). A hypoxic crew might explain the failure to descend, how did the aircraft then turn and fly up the Malacca Strait and then turn south? I’d be much more inclined to believe a hypoxia theory if the aircraft continued to fly south-west past Penang and across Sumatra into the Indian Ocean, but that doesn’t appear to have been the case.

  259. Mick Gilbert says:

    @DrB

    EQUIP COOLING OVRD

    If I’m not very much mistaken that is the mating call of the New Zealand ground-handler parrot (Strigops habroptila Gunsonii). A showy and bombastic little fellow that is generally untroubled by facts, logic or reason.

  260. Victor Iannello says:

    @sk999: The statement was made by a Malaysian official in direct response to questions I posed about the image. The claim was the captures represent a composite of other aircraft that were not MH370. When I asked how the timestamps could match that of MH370, the official simply shrugged. (@Don Thompson and @airlandseaman were also in the room.) You can draw your own conclusions about what the truth is. I put the data in the category of uncertain.

  261. DennisW says:

    @ALSM

    The problem with OXCO retrace papers is that ALL of them use a time scale of days not hours. People don’t power cycle OCXO’s in a typical use profile. So the papers are not bad, just not directly related to our situation. I would really like to see some SDU power cycle tests both with respect to retrace and warm up characteristics (per our earlier discussions), but it is not be. In a previous post I used 1X10^-8 as a worst case retrace estimate which would be some 15Hz at L band. It does not change the conclusions relative to direction after the FMT or the rapid descent at the end.

    Graphic linked below from John Vig.

    https://photos.app.goo.gl/glsr9XTLdAbLf3Jr2

  262. Don Thompson says:

    @sk999 & @Victor

    Altho, it appears the same guy was quite satisfied to stand by a similar depiction description of the track on 29th April 2014 in the Metro Lido Hotel to next-of-kin.

  263. airlandseaman says:

    Dennis:

    I found plenty of papers with relevant test data. From one paper (Frequency Accuracy & Stability Dependencies of Crystal Oscillators, Hui Zhou, et.al, 2008) that included the same graphic you cite:

    “3.1.4 Retrace
    When power is removed from an oscillator for several hours, then re-applied on it again, the frequency of this oscillator will stabilize at a slightly different value. This frequency variation error is called retrace error. It is usually occurring for twenty four or more hours off-time followed by a warm-up time which is enough to complete thermal equilibrium. Retrace errors will reduce after warming. The shape of the error curve is like that the crystal walks back down its aging curve when cold and then moves toward the prior drift curve when activated. If the resonator is in its aging phase, the retrace error will be added to the aging drift, while with wellaged resonators the frequency will look for a new level characteristic for alternating operation.

    Usually, retrace errors show clearly less spread with SC cut than with AT cut resonators. Careful selection of crystals, oscillators can decrease the influence from retrace effect which is as close as a few parts in 10 10 [6]. Retrace is one of factors that affect frequency accuracy of OCXO. For TCXO or other oscillators, retrace is usually not considered as a significant affecting factor to frequency accuracy [11].

    Figure 6 shows how OCXO retrace influences oscillator frequency accuracy. The x axis represents time and the y axis represents frequency accuracy. In (a), the oscillator was kept on continuously while the oven was cycled off and on. In (b), the oven was kept on continuously while the oscillator was cycled off and on.”

    Reference [6] is particularly relevant (Frequency Retrace of Quartz Oscillators-Euler, 1981). I clipped it from the original 1981 Symposium records and it is available here: https://goo.gl/V4qmpr

  264. Victor Iannello says:

    @Don Thompson: In that slide, a straight line is depicted between 18:02 and 18:22. The Lido slide shows a slightly different path.

  265. sk999 says:

    Victor: “You can draw your own conclusions about what the truth is.”

    No problem.

    “There has NEVER been an explicit statement that the Lido Hotel image does not capture MH370.”

  266. Barry Carlson says:

    @Mick Gilbert,

    The Kakapo (Māori: kākāpō) or night parrot, also known as owl parrot (Strigops habroptila), would due to its insatiable appetite for something to do, find the dimmed night time flight deck of a B777 a magical place to practice its inexhaustible stable of tricks.

    Someone needs to let it know of its elevated status!

  267. Victor Iannello says:

    @sk999: I don’t know the point you are trying to make, but your statement in quotes is FALSE. The official was quite clear.

  268. airlandseaman says:

    sk999: As Victor noted above, I was there and I heard the same statement re Lido Hotel image. Separately, the same person told me that many people in Malaysia (he was not specific) did not believe the plane was in the SIO. I found that statement incredible coming from an official representing the Malaysian Government.

  269. sk999 says:

    All: I wasn’t there. I do not know how to read minds.

    Victor: “the official simply shrugged”

    ALSM: “(he was not specific)”

    Apologies for cutting and pasting, but “EXPLICIT”? No.

    You know better. When? Where? Who? Circumstances? Give us the story.

  270. Victor Iannello says:

    @sk999: No, you can’t read minds. You also seem not to be able to read words. I don’t know how to be more clear.

    He was EXPLICIT. The official EXPLICITLY said that the captures were not MH370. He provided no further explanation when I pressed him for details about the coincident timestamps, as evidenced by his shrug. He obviously did not want to say more.

  271. airlandseaman says:

    sk999: Re “(he was not specific)”, I was referring only to the obvious…He did not name any specific people that said the plane was not in the SIO. He was expressing a view, apparently held by some people in his circle, that that the plane was not where we are looking. I can’t be more clear.

  272. sk999 says:

    Victor,

    My original question was as follows:

    “Can you provide a reference to where ‘they’ explicitly said that the Lido Hotel image does not capture MH370?”

    Your most immediate response was “The official EXPLICITLY said that the captures were not MH370.”

    Is that all? Circumstances? Date? Who? Where? Clearly no video recording. Were you writing down notes? Are you recalling from memory?

    Here’s an example of a “reference”.

    On Day 5, (Wednesday, Mar 12, 2014, if I have my dates correct), Daud said the following at 1:48:
    https://www.youtube.com/watch?v=Pl0CW5m9OX0

    “The last plot, after several intermediate primary plot … the last
    plot have a … at 0215 radial 295, 200 miles Northwest of Penang.

    That is a source I can quote with authority. I am looking for a comparable source for your statement.

  273. TBill says:

    @DrB
    “If I have got this wrong, please tell me where my logic has failed.”

    DrB you are obviously working very hard to develop an innocent explanation. The non-innocent hypothesis has proposed answers to some of your questions, and they are not pleasant. I have no problem if you try to develop an innocent explanation.

    The problem comes if you then try to say there is some weakness with the non-innocent explanation(s). The latter requires Victor’s buy-in as a bare minimum, or else it is a suspect argument. If Victor buys it, then I want former NTSB Greg Feith as my second referee.

  274. David says:

    @Ge Rijn. For interest, paint removal. See item 18, photo in 1.0 Introduction of the outside and its paint. You get an idea of scale vs the right door in the 3.0 Identification photo.
    http://mh370.gov.my/phocadownload/3rd_IS/Debris%20Examination%20300417.pdf

    Compare that to the right nose door here, a year before its loss. https://www.airteamimages.com/boeing-777_9M-MRO_malaysia-airlines_196310.html

    Note the position of the “O”. One would expect a vestige of part of that to be evident on the recovered part. There not seem to be abrasion damage as there was to Item 6, the fan cowl. So, lost on impact? Yet the decal on “Roy” shows no deterioration. Conclusions cannot be drawn IMO but this should be put into the “of interest” locker when looking at debris condition in detail.

  275. Brian Anderson says:

    @Mick Gilbert,

    “the New Zealand ground-handler parrot (Strigops habroptila Gunsonii)”

    Yes, we have then too 🙂

  276. TBill says:

    @ALSM
    “Separately, the same person told me that many people in Malaysia (he was not specific) did not believe the plane was in the SIO”

    I wonder if that is similar to what @TimR keeps saying, there is some belief in an Xmas Island/Jakarta planned destination, but strictly speaking that is still SIO.

    “…the flutter modes that can occur when the airspeed approaches mach1 are potentially catostofic, and the damage can occur within a few seconds of the onset, exactly as Brian and I have described for the last 2 years.”

    So your flutter-with-damage requirement is hitting M1.0 sometime in the descent. So my question boils down to, what is the lowest altitude that M1.0 can be reached? The BFO math seems to say that starting descent from a lower altitude (say FL100-FL150) requires a very steep pitch down to meet the BFO, so that works in favor of a lower altitude case working. Would also tend to suggest Arc7 is the final point for that case.

  277. David says:

    @Dr B. As you know there is plentiful data and research into time-of-useful-consciousness.

    What is missing is what you have in mind I think, the time from loss of useful consciousness to the time of no activity, during which there could have been misdirected activity, presumably in this case by the pilot less susceptible.

    I raised the issue some time back with a retired air force doctor trained in aviation medicine. He doubted that the onset in hypoxia could plateau over long periods.

    Still, here is an account many have seen of erratic behaviour over a some time. See the footnotes of:
    https://www.theairlinepilots.com/forumarchive/aeromedical/decompressionandhypoxia.php

    What also may be of interest to you as to possible causes, again raised before I think, is this: http://www.airweb.faa.gov/Regulatory_and_Guidance_Library/rgAD.nsf/0/98f7d563caeb94b7862572b2005129b3/$FILE/2007-07-05.pdf

    Note that the cabin altitude warning message was inhibited.

    In the context you explore, the turn back suggests being awake to a problem, slow hypoxia not having concealed it, though all the same the unnecessary double reporting of altitude beforehand is seen by some as odd. Whatever was perceived it would have required various switching and no obvious call to don oxygen masks. Presumably there was a reluctance to descend much, maybe the pilot wanting to retain glide potential…..

    Back to duration, the P3 experience does suggest there can be prolonged consciousness after behaviour becomes illogical. I doubt there to be any research data about how long.

  278. Ge Rijn says:

    @David

    On the nose gear doors/item18
    I think you made a slight observation mistake here.
    The gear doors depicted in your linked 9M-MRO picture are the rear nose gear doors.
    The forward nose gear doors are not visible because they are closed:

    http://www.flickriver.com/photos/wbaiv/2720148306/

  279. Sabine Lechtenfeld says:

    @ Dr.B
    I’m not saying that the plane isn’t at your meticulously calculated crash location. We will soon know. But to be very frank: I consider it a waste of time and energy still trying to construct disaster/accident scenarios.
    There’s a rule, we call Occam’s Razor, and it actually suggests that the plane was deliberately diverted with criminal intent. Others have pointed out in great detail where you made chains of untenable assumptions. Therefore I won’t repeat them.

  280. Ge Rijn says:

    @David

    To clear this up more sufficiently, the next picture shows the forward nose gear doors have no decoys or something else painted on the outside.
    It’s 9M-MRG but I assume 9M-MRO would show the same:

    https://www.airplane-pictures.net/photo/106118/9m-mrg-malaysia-airlines-boeing-777-200/

  281. Rob says:

    @ALSM
    @TBill
    @Sabine

    “…the flutter modes that can occur when the airspeed approaches mach1 are potentially catostofic, and the damage can occur within a few seconds of the onset, exactly as Brian and I have described for the last 2 years.”

    I’m not denying per se, that flutter can be a very damaging and destructive failure mode. That is not the issue can here. What I am questioning is the proposal that flutter explains how the MH370 flaperon got detached. The flutter theory fails on a number of counts:

    Firstly, it fails to account for how the adjacent (touching in fact) outboard flap was so violently ripped away from the wing while in the stowed position,ripped away so violently that it took its two hinge fairings with it, and in the process, robbing them of their outboard edges. And all that remained of the outboard flap to make the journey to Africa, was the part that had been closest to the flaperon. The flap was shattered into pieces, the flaperon only lost its relatively vulnerable trailing edge outboard of its rear spar. A more satisfactory explanation for the damage is impact with the water as the aircraft belly flopped, right wing down.

    The flaperon succumbing to flutter fails to explain this region of the wing’s trailing edge was so violently and extensively damaged, while the flaperon itself was broken cleanly away from its two underside hinges, in such a gentle (compared to the flap) manner. And where is the other flaperon? Don’t tell me, that was so violently damaged by flutter that it sank to the bottom, while it’s companion outboard flap stayed on the wing, totally intact?

    Secondly, previous cases of aircraft sustaining structural damage while exceeding the envelope in a dive, or even reaching or exceeding M1, were all under engine power during the descent, and flutter didn’t figure in any of them. The damage was focused on the empenage structure, and wing structure generally.

    Thirdly, and most tellingly, it fails to explain why the aircraft wasn’t found close to the 7th arc as predicted. The aircraft should have been found close to the 7th arc, near or at S38. It wasn’t. This is a damming finding, the real significance of which seems to have been lost on those responsible for defining search areas.

    It seems to have been lost in the panic that ensued after the ATSB search failure, that there is actually no justification in rejecting/suspecting the DSTG primary finding that the aircraft crossed the arc at or near S38. The DSTG got it right first time. One particular criticism of the DSTG Bayesian approach is about the assumptions they took concerning priors; the minimum manoeuvres before = minimum manoeuvres after argument. Well, this criticism is totally unfounded and unjustified. DSTG carried out 6 validation flight exercises on previous MH370 flights, comparing Bayesian PDFs to ACARS path reconstructions, and they were comfortably in the ball park every time. Actual flights, with actual manoeuvring, and they were within the 85℅ of the peak of the PDF each time. They also ran the MH370 analysis with BTO alone and with BFO included, and the difference in the outcome was minimal.

    My original assessment was right on target. The flutter theory is b……s, sorry, baseless (the parrot has just given me a dirty look)

    Nice one Sabine (re DrBs ridiculous, contorted and contrived path fitting exercise, a final desperate effort at defending the reputation of a murderer) You have put into words what I and others were unwilling to. 😉

  282. David says:

    @Ge Rijn. You are right thank you.

  283. Ge Rijn says:

    @David

    You’re welcome. Just one thing sorted out again.
    And thanks to your misguided observation we know now for sure the forward gear doors have no decoys or other painting.

  284. Ge Rijn says:

    @Rob

    It shocks me you publicly accuse ZH of having the reputation of a murderer.
    I think it’s heartless to his relatives and the NoK and a completly irresponsable statement in general.
    This should not be allowed here imo.
    It not only makes you look dirty but also @VictorI’s well respected blog.

  285. Sabine Lechtenfeld says:

    @Ge Rijn, calm down! There’s nothing wrong with Rob’s assertion. Whoever diverted the plane (most likely Zaharie Shah), is ultimately responsible for the death of 200+ people. I don’t think at all, it’s disrespectful towards the NOK to state the obvious.

  286. TBill says:

    @Rob
    As far as flutter damage, I am open on debris interpretation. I assume the French know if it was air vs. water damage. I suppose it is flutter damage, it could have happened before end-of-flight. To me, debris and drift are secondary to BTO/BFO.

    As far as 38S: your argument hinges on that the DTSG Bayesian analysis was some kind of immaculate conception. But ATSB has moved on to 32-36S, presumably recognizing that the Bayesian analysis had weaknesses.

  287. airlandseaman says:

    Rob: Re “, and most tellingly, it fails to explain why the aircraft wasn’t found close to the 7th arc as predicted. The aircraft should have been found close to the 7th arc, near or at S38. It wasn’t. This is a damming finding, the real significance of which seems to have been lost on those responsible for defining search areas”

    This statement shows how conflated and flat wrong your understanding is. The POI location has nothing to do with the descent rate. Moreove, they have not finished searching the 7th arc. More false logic.

  288. DennisW says:

    @Rob

    One particular criticism of the DSTG Bayesian approach is about the assumptions they took concerning priors; the minimum manoeuvres before = minimum manoeuvres after argument. Well, this criticism is totally unfounded and unjustified. DSTG carried out 6 validation flight exercises on previous MH370 flights, comparing Bayesian PDFs to ACARS path reconstructions, and they were comfortably in the ball park every time. Actual flights, with actual manoeuvring, and they were within the 85℅ of the peak of the PDF each time.

    Yes, I read the DSTG book. Do you think it is realistic to test the accuracy of the model using successfully completed scheduled flights? None of the modeled flights had an abrupt early turn from the scheduled flight path. None of the modeled flights had an FMT. If the DSTG were truly serious and wanted to be credible they should have conducted some model testing using diverted aircraft on paths unknown to them to see if they could find them. How expensive could that have been in contrast to the expense of the search? That is what I would have done to test the model. Likewise extensive testing of the SDU reboot was not done. Another cheap set of tests that has been ignored.

    You don’t test things sitting in your office and playing at the keyboard.

    My heartburn with the ATSB has nothing to do with what they did. It has everything to do with what they did not do, and still have not done.

  289. Don Thompson says:

    @DennisW,

    I ought to engage more on comments with which I can agree. Your retort, above is a case in point.

    There is scope for further investigation to refine and confirm the understanding of particular phases of the flight. Holland’s work on the SDU went some considerable way toward completing the understanding the 18:25 reboot but it took 3yrs to produce. Somewhere above I described how data is distributed from the ADIRU, via AIMS, and to the SDU so as to generate the doppler pre-compensation in the AES IF-RF upconvert process. Validating the operation of the SDU through a dynamic regime that may plausibly have occurred during the end of flight phase would be worthwhile.

  290. Ge Rijn says:

    @Sabine Lechtenfeld

    I agree the captain still can be called a prime suspect. But still nobody knows what happened exactly and why. Imo it’s way off limit to accuse him of having the reputation of a murderer.
    The man had a flawless reputation. We are not here to judge and convict someone.
    This statement is way beyond responsable arguing about all that can help finding the plane.
    It sets a tone, when not adressed, of agreement on this blog.
    Which does not serve the blog in a possitive way at all imo.

  291. Victor Iannello says:

    @sk999: Thank you for informing me about what you consider to be a valid reference and why you believe my reference about what was said by a Malaysian official should be ignored. I gave as many details as I am at liberty to give. People here can form their own opinions about my truthfulness about what I reported.

    @All: For those that consider with 100% certainty that the Lido Hotel radar data to be valid, I ask the simple question: After showing the slide to the NOK in Beijing on March 21, 2014, there have been four reports that present the radar data in various formats. Those are the RMP report (not meant for public consumption), the one-year Factual Information (FI), the ATSB report from June 2014, and the DSTG report on using Bayesian analysis to define the search area. In three of those reports, the path from 18:02 to 18:22 is not consistent with the Lido Hotel image. (For the FI, the path in the Malacca Strait from derived from the radar data is only shown as a cartoon. For the ASTB and DSTG reports, the path is shown as a straight line from 18:02 to 18:22.) In one of the reports (the RMP report), no data is shown after 18:02. If the Lido Hotel slide really does show MH370 captures, why do you think subsequent reports neglect to present the captures between 18:02 and 18:22 that are shown in the Lido Hotel slide?

  292. TBill says:

    @Victor
    I do not have a strong opinion, but:

    (1) 1822 point may have been visual (not recorded) as the military was just finally tuning in to find MH370 as it left the MY airspace. BTW at this point they, couldn’t MY have contacted Indonesia to see if Sabang could check their radar? Just saying presumably MH370 had to be aware Indonesia *might* be looking.
    (2) Your SLOP/CP fits like a glove almost…except I favor NILAM to 0796E with similar path
    (3) With reference to the shipping lanes, perhaps Singapore airborne radar captured MH370 as it was possibly in their line of site
    (4) As some point, the working group apparently told Keith Legerwood that his SIA68 hypothesis was not working as the flight went up around N571. So it seems like the working group knew LIDO slide was reasonable.
    (5) I do not discount your discussion, but it is perhaps symptomatic of an urban myth that won’t go away in MY. Possibly they are less able to view MH370 with an open mind, as it hurts the national pride too much over there.

  293. Victor Iannello says:

    @TBill: To go back in a complete circle, my first blog post discussed possible paths over the Malacca Strait. I found that starting at the position and time of the radar capture at 18:02, there are constant speed, straight paths that satisfy the BTO data for the log-on at 18:25. In fact, one path goes through the questionable radar position at 18:22 and has a (18:28) BFO error (calculated minus measured) of -6 Hz, which can be explained by a retrace shift of the FFB by +6 Hz. That shift of FFB to 156 Hz is also consistent with low BFO error for a great circle path crossing the 7th arc around 28S latitude, and would extend to the 45S1 sim point at 45S,104E.

    My main point is that we can cherry-pick data resulting in scenarios that are consistent with the Lido data set being wrong as well as cherry-pick data that are consistent with the Lido data being correct.

  294. DrB says:

    @VictorI,

    You said: “ We have no confirmation from the Malaysians that the Lido Hotel image captured MH370. In fact, they have explicitly said it does not.”

    To my knowledge, there is no public official statement of the invalidity of the military radar track data. I choose to ignore all anonymous sources, particularly when there is no corroborating evidence whatsoever that the radar positions are wrong.

    You said: “Not using waypoints doesn’t necessarily imply inexperience.” I was not implying inexperience. I am just saying that it seems odd to me that a trained pilot would abandon waypoint flight after a quick turnaround when still far out to sea, if all the equipment was working properly. Possibly that choice was affected by some malfunctioning equipment. Maybe Andrew will address this point.

    You said: “The plane might not have flown a lateral offset. Assuming the Lido Hotel image is correct, there are paths that satisfy the BTO and BFO values for the log-on values that involve turns and changes in speed.”

    I think the speed changes for this case may be rather large, and this turn/speed change may not be possible using the full BFO set corrected for OCXO warm-up. It may be possible if you ignore some of the BFOs. Can you provide an example set of values so I can evaluate the possibility?

  295. Victor Iannello says:

    @DrB said: I choose to ignore all anonymous sources, particularly when there is no corroborating evidence whatsoever that the radar positions are wrong.

    The source is anonymous to you, not to me or others here. He is a high-ranking official that has a high level of aviation expertise. Again, you may choose to believe I am not being truthful. You go right ahead. But then your statement that “there is no public official statement of the invalidity of the military radar track data” is also predicated on your belief that I am lying.

    As for corroborating evidence, I asked previously why the subsequent RMP, DCA, ATSB, and DSTG reports do not present the Lido data. No response yet. So tell me, what evidence CORROBORATES the data? I am willing to say the data’s validity is uncertain. You seem quite sure it is valid. Why?

    I am just saying that it seems odd to me that a trained pilot would abandon waypoint flight after a quick turnaround when still far out to sea, if all the equipment was working properly.

    We can be sure the turn was not initiated using waypoints based on the shape of the turn. It was either manually flown or using HDG/TRK SEL. If a pilot wanted to fly near airport waypoints, but didn’t care to fly exactly along airways or over waypoints (why would he?), using HDG/TRK SEL along with the ND is perfectly acceptable. In fact, it would save him from having to fiddle with the CDU, which takes a bit more time than using HDG/TRK SEL. I don’t associate the use of this A/P mode as meaning the pilot was untrained.

    It may be possible if you ignore some of the BFOs.

    I don’t believe it is possible to accurately correct the BFOs during the transient warm up. The errors introduced are too large. Also, if you are using the log-on at 12:50 log-on as a reference, the measured BFO transient errors have to be reduced by one-half, as the BFO error is twice the drift when the log-on proceeds as Class 1 using the closed-loop Doppler compensation.

  296. Don Thompson says:

    @Victor asked “why do you think subsequent reports neglect to present the captures between 18:02 and 18:22 that are shown in the Lido Hotel slide?

    I am confident that the Lido Hotel slide is valid, not 100%, but little in this endeavour is conclusive.

    I will suggest the Lido Hotel slide presented as too difficult to explain: there’s a hole in the middle, the targets are sporadic, the track leading into hole doesn’t line up with the track leading out of the hole, it’s of military provenance, its precise accuracy may be questionable (as a consequence of slant range from radar head and atmospheric effects).

    The crisis management consultancy team from John Bailey’s Ketchum-ICON office in Singapore, engaged on 15th March, may have given advice about its use, ie don’t, it doesn’t fit ELI5 criteria. The entire involvement of the air force was an unmitigated embarrassment. The subsequent releases dealing with ‘what the radar saw’: the FI, the ATSB and DSTG documents referred to only the final contact depicted on the ‘Beijing Lido’ track, so the content of the image cannot be discarded entirely.

    Despite various attempts, I have had no success in getting further details of the briefing scope for that day (21 Mar 2014) in Beijing. The leader of the delegation Lt-General Ackbal, Chief of Air Operations, may even have presented a previous slide showing the ‘air turn-back’ alleged to have been recorded by military radar in the vicinity of IGARI.

  297. DrB says:

    @VictorI,

    You said: “If the Lido Hotel slide really does show MH370 captures, why do you think subsequent reports neglect to present the captures between 18:02 and 18:22 that are shown in the Lido Hotel slide?”

    Based on what DSTG said, they were not provided all the radar positions, and it seems the same is true for ATSB. Perhaps Malaysia, or more specifically the Malaysia military, decided that the intermediate points were not necessary to define the connecting path, and they simplified the data set they gave ATSB. That would reveal less about their radar capabilities, which might have been a national security concern for them. The Lido slide might have inadvertently exposed more information on capabilities than they were comfortable admitting to later on.

  298. Victor Iannello says:

    @DonB, @DrB: If you two are trying to persuade me that the data is uncertain, you are doing well. That’s not to say the plane did not fly up the Malacca Strait, and even the target at 18:22 might be valid but with considerable position error, which is what the DSTG assumed. However, I simply can’t assign the certainty on the N571 targets between VAMPI and MEKAR that others do.

  299. Niu Yunu says:

    @Victor:
    I may be wrong, but I think sk999 and DrB are not attacking or even question your truthfulness (much less accusing you of lying). I believe nobody here does. I think they are just not comfortable – as a matter of principle and professional/scientific/journalistic diligence – with using anonymous sources*. I am sure that is their concern. I don’t think their problem is that you are quoting an anonymous source – they would have the same problem with anyone else. In other words: To them it’s not you, who is the problem, but the use of anonymous sources as a proof for something. sk999 and DrB can correct me, if I am wrong.
    Just an attempt to mediate.

    * (Of course an anonymous source is known to the journalist, obviously, but if he/she cannot mention the name in the public domain, it’s called “anonymous source”, it’s a technical term.)

  300. DrB says:

    @VictorI,

    You need to take a rest. You just said: “But then your statement that “there is no public official statement of the invalidity of the military radar track data” is also predicated on your belief that I am lying.”

    I never said or implied any such thing, Victor. There is no public statement. What one person said in a room to a few people is not a public statement from an official source. I am not doubting you heard what you described. I’m just saying that I don’t use verbal, non-public sources of information. That’s all. It has nothing to do with you personally.

  301. Niu Yunu says:

    DrB’s second paragraph –> that’s exactly what I was trying to say

  302. Victor Iannello says:

    @DrB said: I am not doubting you heard what you described.

    Thank you for saying that. I think we understand each other. Let’s move on.

  303. DrB says:

    @Andrew, @Don Thompson,

    Many thanks for debunking the claims made elsewhere regarding depressurization and warning lights. Your technical expertise is invaluable, as always.

    Yes, Don, you are correct that we only know that no radio calls were detected from 9M-MRO, not that the radios were non-functioning.

    Don, I didn’t fully follow your comment about receiving messages as a reason for turning the SDU back on. Are you saying that to get ACARS running the SDU has to be on, and probably no one knew in addition that, if the SDU were turned on, the BTOs/BFOs would be recorded? But no additional ACARS messages were recorded, and if ACARS were up and running again, wouldn’t outgoing status reports automatically be issued?

    Don or Andrew, what is the power source for the integrated speaker-amplifier used to sound the aural alerts for depressurization? Could it have lost power at the same time as the SDU?

    @Mick Gilbert,

    You and Don are both (mostly) correct on the source of the debunked claims.

  304. DrB says:

    @Sabine Lechtenfeld,

    Thank you for commenting. I can’t agree with you that Occam’s Razor has a strong preference for a hijacking scenario. I have continued to point out the several inconsistencies in both hijacking and accident scenarios, and I don’t see a clear preponderance of evidence on this issue. I am on the fence, as they say, and I think it is a mistake to rule out an accidental cause or a hijacking cause without strong evidence. At this point no one knows what happened, but this might change in the near term. If it does, there will be a lot of crow to eat for those who have already (incorrectly) made up their minds.

    @All,

    Does anyone care to respond to my posting yesterday on why the single cell tower connection may be evidence against a hijacking scenario?

  305. DrB says:

    @Andrew,
    @David,

    Thank you for your comments on the possibility of an undetected and extended hypoxic condition.

    David, your second reference said: “. . . we are issuing this AD to prevent an ASCPC (air supply and cabin pressure controller) failure that could stop airflow into the airplane, inhibit the cabin altitude warning message, and cause an incorrect display of cabin altitude. These failures could result in depressurization of the airplane without warning.”

    So the “undetected” decompression part has happened before in a B777 in 2007, although presumably some corrective actions were taken for 9M-MRO.

    Regarding the “extended” hypoxia, your first reference was enlightening. Basically, for a person to function for 30 minutes or longer with no supplemental oxygen, the cabin altitude for a rapid decompression would be 15,000 feet or less. I have not found similar data for a slow decompression, but the active period must be longer than for rapid decompression.

    For a pilot to function at a minimal level for an hour or more after decompression, the cabin altitude must be very much lower than FL340 (i.e., the decompression must be partial), and a slow decompression is probably needed.

  306. TBill says:

    @DrB
    The cell phone capture has the appearance that the FO left his cell phone on and he may have been incapacitated, or trying to signal for help at a window, locked outside the cockpit, in the cabin. Or some feel the FO could have been incapacitated in the cockpit, and he left his cell phone on.

    The possibility of using non-pressurized mask/O2 bottle depends on the altitude and abruptness of any depressuring event. Conceivably an intentional depressure with intent to incapacitate could be difficult to defend against. Possibly the FO was in the best position to grasp what was happening and make a mayday signal. There have been non-substantiated accounts of other cell phone captures. It is also very hard to connect at FL350 or thereabouts, that was one of the fortuitous events that happened that perhaps tell us more than we otherwise would know, right up there with the final BFO burst.

    Now then if the scenario above did not happen, others have argued that the PAX could have been told everything was OK as you stated.

  307. Rob says:

    @DennisW

    “If the DSTG were truly serious and wanted to be credible they should have conducted some model testing using diverted aircraft on paths unknown to them to see if they could find them. How expensive could that have been in contrast to the expense of the search?”

    I agree, that would have been a more productive way of going about it. But we are criticising the ATSB and DSTG with the benefit of hindsight. When this all kicked off, decisions about scope and degree had to be taken, and kept to, for expediency if nothing else. The ATSB were responsible for searching on behalf of Malaysia. Malaysia were probably not the most encouraging or supportive of partners, then there was the cost element. The ATSB were operating with one hand tied behind their backs.

    In theory, they could have gone overboard and commissioned a test flight with a B777 LR and reflown the possible routes up to the 3rd arc, breaking off to refuel, repeating time and time again, recording the BTO and BFO (problem there straight away, ever oscillator behaves differently of course) but with the ACARS switched on.

    It might have allowed a better appreciation of how much reliance could be placed on BFO.

  308. Don Thompson says:

    @DrB,

    The overhead speaker/integrated amplifier, and the Warning Electronics Unit, are powered from 28V DC sources. Left side from 28V DC Battery Bus, Right side from Right 28V DC Right.

    ACARS and SDU: the SDU must be powered, Logged On, and SATCOM selected in ACARS Manager for initiation of air-to-ground messages.

    However, the FCOM states: “If both boxes [in MFD ACARS Mgr screen indicating VHF and SATCOM carrier] are deselected, ACARS loses the capability to send downlink messages, but can receive and display uplink messages“.

    That FCOM statement is only true if an active ‘session’ exists between the aircraft and the ACARS server on the ground. At 1809UTC that ‘session’ was lost due to the failed ground-to-air message from MAS ODC & no ‘session’ was reestablished regardless that the satcom datalink was re-established at 18:25.

  309. Don Thompson says:

    @Victor

    Beijing Lido: since 21 Mar 2014 Malaysia has been 100% effective in evading any clarification and any possibility for certainty of the detail described by Ackbal and his team.

    The assets capable of surveilling the area have been independently identified; atmospheric conditions that likely affected the performance of the surveillance assets can be determined from contemporaneous atmospheric soundings at Medan, Phuket, Penang and Kota Bharu; 9M-MRO flew through the area; the information passed to the Australian agencies was derived from one target plot depicted in the Beijing Lido image (and those agencies accepted the uncertainty for the accuracy of the 18:22 position).

  310. DennisW says:

    @Rob

    I agree, that would have been a more productive way of going about it. But we are criticising the ATSB and DSTG with the benefit of hindsight.

    I don’t think it is hindsight. Model verification is extremely important, and money well spent (a relatively small amount of money compared to the search cost). That is in the axiomatic domain not hindsight.

    I compare it my practice of going to shipping and pulling a packaged first article and taking it home, inviting a techie neighbor over, and downing a twelve pack while seeing if he can get it working using the included operator manual. More than half the time something was missing, cable, connector,…vague instructions,…

    There is absolutely nothing like real world physical testing to understand boundary conditions.

  311. Victor Iannello says:

    @TBill: We don’t know that other cell phones on the aircraft did not register on Penang Island or elsewhere. In the RMP report, there is some data pertaining to cell phone activity of the crew. I have not seen any statements about cell phone registrations for passengers. I’ve asked two relatives of passengers whether Malaysia contacted them for passenger cell phone numbers. One, a non-Malaysian, said never. The other, a Malaysian, said that other Malaysians were asked for cell phone numbers of passengers, but it was not compulsory to provide those numbers. Be careful about assuming that there were no other cell phone registrations. We simply don’t know.

  312. Andrew says:

    @DrB

    RE: “So the “undetected” decompression part has happened before in a B777 in 2007, although presumably some corrective actions were taken for 9M-MRO.”

    Also note that this problem only affected ASCPCs with a certain part number. The issue might never have affected 9M-MRO, given that the aircraft did not have the APU-to-Pack takeoff mode mentioned in the AD.

  313. sk999 says:

    DennisW,

    There is more to route reconstruction than just sending a plane South. The phase of the satellite in its orbit is important. Does the specific piece of equipment matter? You are not going to get the SATCOM on 9M-MRO back. You can never reproduce the wind pattern of 7 Mar 2014. Magnetic precession has marched forward inexorably. Do you need to fly an actual plane, or are simulations good enough?

    If I could ever get my 777 out of the shop, I would fly in constant track and heading modes (both true and magnetic) to see how accurately the aircraft follows the nominal routes. As far as we know, the DSTG has zero data on flights flown for extended lengths of time in those modes, and none of the data used to calibrate the Orstein-Uhlenbeck processes in LNAV mode are really applicable to them.

  314. Sabine Lechtenfeld says:

    @Dr.B,
    Thank you for answering me.
    I do not think that you have pointed out any inconsistencies with a hijack scenario. You pointed out several things and claimed that they are inconsistent with a hijack scenario. That’s big difference. Especially your understanding of hypoxia and it’s consequences seems not to be realistic.

    But let’s agree that we disagree and move on The important point is if the plane went down where you think it went down. And we will soon know. We will all be very happy if the plane will be found – no matter which scenario we have favored.

  315. David says:

    @Dr B. “For a pilot to function at a minimal level for an hour or more after decompression, the cabin altitude must be very much lower than FL340 (i.e., the decompression must be partial), and a slow decompression is probably needed.”

    Sounds right (though @ Sabine Lechtenfeld might disagree it seems).
    Getting to possibilities, some musings:
    • Assume a cabin altitude high enough to induce mild hypoxia yet cabin altitude stable at that point, so the hypoxia not worsening from that rising.
    • This would not occur if bleed air were turned off, when cabin altitude would continue to increase past that stage and mild hypoxia would get worse.
    • So assuming air supply stays, were there then a small hole* in the hull the outflow valves would close by the area of the hole and thus there would be no appreciable increase in cabin altitude. (*By hole above I mean increased air discharge other than through outflow valves.)
    • Were the hole big, the valves would close automatically at 11,000 ft cabin altitude. Cabin altitude would continue rising. However with outflow rate falling as the cabin altitude rises and its pressure difference to ambient is reduced, it would drop no further when outflow no longer exceeded inflow.
    • Assuming bleed mass flow rate remained approximately constant and disregarding the effect on flow of aperture shape, for cabin altitude to exceed 11,000 ft obviously the hole area would need to be bigger than the previous outlet area of the outlet valves.
    • However it would not just be the hole area which would control the cabin altitude at which outflow rate restabilised but the also the local total pressure at the outlet. Obviously facing forward would bring a ram effect. In the ‘taper’ towards the tail the converse would apply. The stabilised cabin pressure would rise or fall by that delta P. Thus stabilised cabin altitude would depend on both hole size and position. That complication might be reduced in this case since a forward hole might be too draggy for consequent fuel consumption increase not to weigh against that.
    • (An idle question which arises is what commonly is the opening of the outlet valves at altitude? Given roughly constant bleed flow rate with changing altitude and an increasing cabin-to-ambient pressure differential I would expect the opening to close with altitude. That of course would reduce ‘minimum’ hole size)
    • Changing the scenario now, if for any reason the outflow valves opened fully and nothing else changed much the cabin altitude may well stabilise above ambient because of their continuing choke effect. Walking back from that, their opening but not fully would lead to a stabilised cabin altitude but lower, that is without any hole.

    The above may not help you much other than painting scenarios where it may be that an hypoxic pilot could last some time longer than under the standard supposition that cabin altitude would increase. Summarising, it might take a substantial increase in outflow to overwhelm automatic flow reduction through outlet valves if they were operating properly but obviously were they open more than they would be under normal operation there would not be the same compensation.If opened wider somehow than normally they would be, cabin altitude could be stabilised as in the above but without a hole.

    Broadly then there is no reason to suppose that cabin altitude could not be stabilised at a mildly hypoxic level, extending its duration. Where this might differ from the standard approach is that such as ‘useful time of consciousness’ is based on cabin/cockpit pressures continuing to decrease and naturally there has been little interest in human performance once the useful consciousness point is passed.

    An accompanying explanation would be needed for alarms and warnings not to have had effect. (Deployment of masks in the cabin could be delayed from 13,500 ft to up to 15,000 ft by landing altitude setting).

    The FAA example is a reminder that robust systems like this and the MAS ADIRU, can fail as can such carefully designed and tested components as the QF-32 A380 turbine disc in 2010.

  316. Rob says:

    @Andrew

    I would be grateful if you could check something for me.

    On the B777, do you know which bus/buses power the outflow valves and aircon pack inlet valves?

    Thanks

    Rob

  317. Rob says:

    @David

    Our two posts were whizzed off at the same time, purely by coincidence both refer to pressurisation system valves. And some people think synchronicity is all in the mind.

  318. David says:

    Dr B. Above, 5th dot point I said, “Assuming bleed mass flow rate remained approximately constant…”, I think volumetric would be more constant but that does not affect the principles I aimed to address.

  319. Richard Godfrey says:

    I have been following the debate regarding the Lido slide over the past 2 days.

    I agree with Victor that “We have no confirmation from the Malaysians that the Lido Hotel image captured MH370. In fact, they have explicitly said it does not.” Personally, I think the Lido slide was faked.

    When pressed by @sk999 and @DrB, Victor stated “I gave as many details as I am at liberty to give”.

    Mike has corroborated Victor’s statement “I was there and I heard the same statement re Lido Hotel image.”

    I agree the statement about the Lido image reported by Victor was not a public statement, but it is an official statement in the context that it was given. The fact is that a number of us were invited to a large high level official meeting, but also required to sign a non-disclosure agreement. The bottom line is that there is no public statement, which satisfies the criteria of @sk999 to get name, rank, number, time, location or a video recording. Although Don has confirmed that the person concerned participated in the Lido presentation and was therefore in a position to know.

    Victor has made it clear that the Lido slide was never used in any of the official reports. In my view, if the Lido slide was genuine, then surely it would have been in all the official reports, public or otherwise.

    I fully accept the reservations expressed by @sk999, DrB and others to only consider verifiable official public statements. Everyone is entitled to form their own view as to the veracity of the Lido slide.

  320. Don Thompson says:

    @Richard G,

    Apologies if I was not clear, the person concerned participated in a later Metro Park Lido Hotel briefing on 29th April 2014. On that occasion the depiction of the aircraft track had evolved to exploit a path illustrated on Google Earth.

    @All

    If one decides to discard the information about the track presented in the ‘Beijing Lido’ image, the case for accepting any radar data becomes similarly tenuous regardless where the information was published.

    No ‘raw’, source, data has been published by any Malaysia agency for any radar track. By ‘raw’ and ‘source’ I would accept logs of ASTERIX data exchanged between the remote radar heads and the central ATM system (or central C² system in the RMAF environment).

    The Australian agencies stated that Malaysia provided data describing the aircraft track at 10sec intervals. That description is at odds with the detail provided in the March 2015 Factual Information report that depicts four sparse track fragments. Only two of the track fragments are attributed to a reporting source, that is the DCA (Terminal Area Control) radar at Kota Bharu airport, while the later two are not specifically attributed to any reporting source. One must ask how were the ‘gaps’ filled to provide the data to the Australian agencies?

    The Factual Information report, in section 1.1/figure 1.1B, attributes only six radar observations to military sources:

    5/ P3415 : Appeared at 1747:02 UTC [0147:02 MYT]

    6/ P3415 : Coasted at 1748:29 UTC [0148:29 MYT] Dropped at 1748:39 UTC [0148:39 MYT]

    7/ P3426 : Appeared at 1751:45 UTC [0151:45 MYT]

    8/ P3426 : Coasted at 1752:25 UTC [0152:25 MYT] Dropped at 1752:35 UTC [0152:35 MYT] P3426 last seen on radar display approximately 6Nm South of Penang

    9/ The primary target (military radar) appeared to track west-northwest direction joining RNAV Route N571 at waypoint VAMPI then to 10Nm north MEKAR

    10/ The primary target ended at 10Nm after MEKAR at 1822.12 UTC [0222.12 MYT]

    Reports 5, 6, 7 and 8 appear to be consistent with the two track fragments near Penang, depicted in the FI Fig 1.1F, and with the performance of the ATCR-33 PSR employed for ATM (terminal area control) purposes, operated by military staff and located on Butterworth airfield*.

    So as to pursue an investigation of the cellphone LBS registration, MCMM acquired information from the following sources (per MCMM contribution to RMP):

    a) ATC-Aircraft Radiotelephony Transcript
    b) Eyewitness report obtained from newspaper reports
    c) Celcom’s Detection of Mobile Phone at BBFARLIM2 Base Station
    d) Military Radar Plot and Aircraft’s speed obtained from PDRM

    The aircraft track depicted by MCMM using the above information appears to be a superset of that illustrated in the FI, fig 1.1F, it certainly extends further west with specific notes for position, speed, and altitude. Only d) could possibly provide position, speed, and altitude information.

    The track and position information described in the ‘Beijing Lido’ image, Factual Information, the MCMM contribution to the RMP reports folio, and the publications from the Australian agencies, expose many inconsistencies if one expects the accuracy of reporting available from secondary surveillance technologies. Subsequent to 17:21UTC, all position reports were derived from PSR assets.

    Allowing for the inaccuracies in the PSR reporting, consistency is possible across the various sources.

    My wider observation is that detailed publication of information gleaned from military assets is rarely exposed, in detail, by air accident investigations. Passing references, but nothing in detail. Cases in point: US long range radar that enabled recovery of cargo door lost from N4713U; FAA-USAF ASRS network contributed to final position location and end of flight profile of SU-GAP off Nantucket; HAF and RAF assets contributed to final position location and end of flight profile of SU-GCC over the Mediterranean Sea.

    To be clear: I am not condoning the lack of detailed information concerning this element of 9M-MRO’s loss. The choice is evaluate what is available, then use it or not.

    \[*\] The Butterworth Airfield ATCR-33 radar, used for joint air traffic mananagement purposes covering Penang International and Butterworth, should not be confused with the Western Hill RAT-31DL radar that forms part of the Air Force air defence surveillance network.

  321. Victor Iannello says:

    @Don Thompson: There is no doubt precedent in not releasing military radar data in accident investigations. However, I doubt there was ever a case in which military data was presented by military brass in an open forum with reporters present and later that data was omitted from future reports (both secret and public) and the existence of the data was (privately) denied by officials.

  322. TBill says:

    @David @DrB
    Former NTSB Greg Feith, I think (from listening to his interviews) would say we have to consider what happened to MH370 may have been Human factor involved. You cannot just rule it out for weak reasons.

    He would probably say, accidental hypoxia is counter-indicated by the following:
    (1) Crew would don O2 masks
    (2) ACARS/Satcom/XPonder would continue reporting
    (3) Apparent expert flight path
    (4) Maneuvers out to 18:40 and maybe close to 19:40
    (5) Turn on of SDU at 18:25
    (6) Any fire or other would serious accident would bring the aircraft down
    (7) Coincidence of comms loss at FIR boundary

    Yes we could develop a remarkably rare scenario where what we saw was not intentional. That does not suffice to exclude the possibility that human factor was involved.

    If a comment like DrB wants to make, that intentional event is ruled out because DrB does not feel what we saw is intentional, was peer-reviewed by Greg Feith, I do not think the statement would stand up to peer review.

  323. Ge Rijn says:

    @TBill

    To your list I could add:

    -the turn at Penang
    -and after this the plane avoiding Indonesian- and Thai FIR bouderies.
    Flying save within Malaysian FIR territory along ~N571. Not raising suspision with Indonesian- or Thai ATC or military.

    It seems to me the turn at Penang could be well intended to stay within the Malaysian FIR bounderies.
    Those are rather narrow between Indonesian- and Thai FIR.

    The Indian/Chennai FIR boundery would not have been a problem, for MH370 could not have been detected by India there, let alone be intercepted by them.

    Just for the graphic:

    http://www.airtrafficmanagement.net/2014/03/malaysia-seeks-expert-advice-on-mh370-flight-path/

  324. Richard Godfrey says:

    @Don Thompson

    Many thanks for the clarification, there were indeed a number of meetings at the Beijing Lido Hotel.

    For the sake of clarification, the slide of the purported MH370 Military Radar Trace presented by the Malaysian delegation on 21st March 2014, that I believe is faked is as follows:

    https://www.dropbox.com/s/1ehhy0wl9rblsro/Reuters%20Radar%20Trace%20Beijing%2021032014.jpg?dl=0

    The radar trace begins some time after Penang when the aircraft depicted was en route towards Pulau Pulak:

    https://www.dropbox.com/s/cqy5815z6hfvr85/Reuters%20Radar%20Trace%20Beijing%20Full%2021032014.jpg?dl=0

    This presentation raised 15 questions as summarised in a post by Victor at Duncan’s website dated 24th September 2015:

    http://www.duncansteel.com/archives/1987

    In my view, you are quite correct to question all presentations by Malaysia and all radar graphics submitted by Malaysia without source data.

    Whether the source, who stated that the Beijing image was not from MH370, was present on 21st March 2014, I cannot tell. There is a person in the following picture who looks like him standing off to the right with his head bowed, but the resolution is too poor to be sure:

    https://www.dropbox.com/s/sz3kag3g29xolfg/Beijing%20radar%20presentation%2021032014.jpg?dl=0

  325. TBill says:

    @Ge Rijn @Victor
    One interesting aspect of Deplorable Edwards argument I find, is he shows radar rings from Port Blair but none from Indonesia. So anything too much north of N571 would appear to be in Port Blair radar range (I do not know if they have primary radar). Up til now my focus has been on Sabang radar.

  326. Ge Rijn says:

    Another thing that still bothers me is the co-pilot’s cell-phone connection at 17:52:27 (or 17:52:57, the MRP reports show two different times).

    This matches with @Don Thompson’s point-8 radar capture at Butterworth.
    The RMP report states at that time the plane was flying between 35 and 45Kft altitude at a speed of ~500 knots based on primary radar data (?).

    The report(s) also states that in the control flight they made to test the ability of cell-phones and towers to connect at Penang, BBFARLIM2 made no connection at all and none of the tested cell-phones made a connection to any Celcom-tower above 8000ft.

    Now that the radar-data are under a loop again I feel this is something to consider also again:

    http://jeffwise.net/2016/11/11/long-rumored-police-report-of-cell-tower-connection-leaks-at-last/comment-page-4/

  327. DennisW says:

    @sk999

    There is more to route reconstruction than just sending a plane South. The phase of the satellite in its orbit is important. Does the specific piece of equipment matter? You are not going to get the SATCOM on 9M-MRO back. You can never reproduce the wind pattern of 7 Mar 2014. Magnetic precession has marched forward inexorably. Do you need to fly an actual plane, or are simulations good enough?

    Yes, I tend to get spun up. I was more concerned with trying to chracterize the SATCOM relative to reboot behavior (settling time/behavior and post reboot offset) than route replication. I know both of these behaviors would depend on the particular SATCOM used, but just getting some typical data would be helpful. Likewise with testing the path algorithm. We can never reproduce the conditions of the diverted flight. I would just like to see how the DSTG algorithm performs on more complex flight paths than routine scheduled flights.

  328. Ge Rijn says:

    @TBill

    They have primary radar at Port Blair with a range of 280km if I have to believe the following link:

    https://www.pressreader.com/india/the-times-of-india-new-delhi-edition/20140316/281900181151680

  329. Ge Rijn says:

    @TBill

    To add.. ~280km would have been too short to detect MH370 at Car Nicobar I think.

  330. Ge Rijn says:

    @TBill

    Then your point 5: Turn on the SDU at 18:25.
    You could also say; Turn on the left IDG again.
    Anyway this could not have been done by a pilot suffering from hypoxia for an hour or so. Far too complicated to perform. And then also make a controlled FMT somewhere after this time.
    The hypoxia-scenario just makes no sence.
    All recorded actions clearly point to consious planning.
    Not by someone who was delirious by lack of oxygen.

  331. airlandseaman says:

    Hmmm…Looking back at my thoughts as of 2 weeks after MH370 went missing…

    airlandseaman
    Posted March 24, 2014 at 8:52 PM
    I’m still concerned about the overall solution. If they have independent spot beam data, I’m ready to agree…it went south for sure. However, if they descended to 12000 feet and slowed down to 250 kts or so, and stayed there, rather than climbing back to 37000 feet (why?), the same handshake data can be fit to a shorter path ending up further north. According to the Delta 777 manual, the fuel burn per hr is about 90% of what it would be at 37000 feet and 475 kts. So the fuel time would be about the same or a bit more, but the path would have to cut through the circles with a more easterly heading (for the same radial velocity at a slower speed). It is only by assuming they climbed back to 35-37000 feet that the data fits a more southerly path that went farther.

    Now, 4 years later to the day, with each passing that MH370 is not located, the alternative 7th arc “low and slow” theory is looking more and more likely. The 7th arc is more certain than ever, but where on the 7th arc? If not at FL350/480kts ending at S37.7, how low and how slow was MH370 going? And how far NE on the 7th arc? Let’s hope Chari and his drift analysis is right, and OI finds it in the next 5 weeks.

  332. Ge Rijn says:

    @airlandseaman

    Thinking about your comment my other scenario of one engine INOP (long time ago) comes to mind again. Flying lower and slower afterwards.
    But I saw this scenario only combined with an intervention of the Malaysian military airforce which then did intercepted the plane at ~18:22 and took out an engine with a rocket. Maybe what Kate Tee saw afterwards; a plane with a burning engine heading south.
    I do not endorse this scenario anymore for it’s too complicated without any clear indication or proof.

    The Broken Ridge trenches between ~96 and ~97E and ~32.2 and ~33S have the best opportunities imo. According Chari’s drift-analysis indeed.

    We’ll see. All our efforts will have result anyway.

  333. Victor Iannello says:

    @ailandseaman: Further north on the arc doesn’t necessarily mean the speed after 19:41 was slow. The “delay before cruise south” scenario is also possible.

  334. airlandseaman says:

    Victor: I agree with your comment, but if the plane took a low and therefore, by necessity, slower path (for example, a decompression scenario), then it would have to end up further NE. Other more complicated and faster paths could also end up further NE, but a “low and slow” path must end up further NE.

  335. Rob says:

    @ALSM

    “Let’s hope Chari and his drift analysis is right, and OI finds it in the next 5 weeks.”?

    Hope is all you’ve got left, but hope is is not going to make it any more likely the plane is further north.

    “If not at FL350/480kts ending at S37.7, how low and how slow was MH370 going? And how far NE on the 7th arc?”

    Are you sure he didn’t hitch a ride on a passing ore carrier?

    Do you have any rational, logical, confirmation bias-free reason for assuming the aircraft couldn’t have been flown to S38 and then pilot-glided further south by a pilot who intended to lose the plane in the SIO, apart from your refusal to accept that the rogue pilot theory?

    Email from Ross follows:

    “Hi Robin.

    Do you have a view on this 7/3/18 paper from Dr Bobby Ullich? I find it incomprehensible and my BS detector suspects it places too much certainty on the satellite data analysis.”

    My reply:

    “Hi Ross

    Your intuition is spot on, it’s total b……s. A textbook example of Confirmation Theory at work and quite simply, very embarrassing. Dr Bobby should be ashamed of himself for printing such contrived nonsense. His previous work on developing an accurate fuel model for the aircraft, taking into account such factors as upper air temperatures on that night, the state of wear of the engines etc was brilliant. Only he could have done it.

    I can’t remember if he’s also a member of the IG, but he’s certainly tarred with the same brush. Behind his meticulous, scientific approach (he used to be a professional radio astronomer) there has always been a thinly, to me, disguised mission to prove the crew innocent of any wrong doing.

    I remember about 18 months back, he asserted that his preliminary calcs of the fuel burn rate on the night in question told him that MH370 could never have reached as far as S38, and that the ATSB needed to look further north along the arc. Just before the ATSB called off the search in January 2017, he had managed to persuade them that they should examine S35 instead. He had convinced himself that the BFO data pointed to an unpiloted path, with the autopilot guiding the plane in Constant Magnetic Heading Mode. I dont expect you to understand what that is, don’t worry. Then he decided that Constant Magnetic Track Mode was a better fit, always working out the flight path with extraordinary precision. OI have now searched these areas, guess what, no plane. This latest path takes him further north still, with the same incredible precision. The plane won’t be there either.

    But the irony is, when he finally completed his comprehensive fuel model, it duly showed that MH370 was indeed capable of reaching S38! It was this that helped me identify the path that the Captain actually flew, the path toward waypoint S41 E88. A very important result, thanks partly to DrB. I subsequently pointed this out, and he just brushed it off with “well, it’s been proven that the plane couldn’t have flown a great circle path ending at S38, because the ATSB have searched the area, and the plane isn’t there”

    End of email

    Confirmation bias is an insidious phenomenon.

    ALSM, time for a reality check. You have invested so much intellectual capital in the unpiloted descent theory, your refusal to accept you could be wrong after all will stymie any chace of a successful search. You are in an unenviable position.

    The pilot’ glided the plane further south. The final BFOs are either being misinterpreted, or they happened to catch the pilot exchanging potential energy for kinetic energy, before establishing himself on a glide at a lower level, to avoid the increasing southern Jetstream headwinds he was encountering at 38,000ft at S38? Either way, he ended up south of the ATSB search area.

    I am doing this at the risk of being banned, because you need to wake up. You are sleepwalking to failure.

    I am exploring other avenues.

  336. Victor Iannello says:

    @Rob said: Do you [ALSM] have any rational, logical, confirmation bias-free reason for assuming the aircraft couldn’t have been flown to S38 and then pilot-glided further south by a pilot who intended to lose the plane in the SIO, apart from your refusal to accept that the rogue pilot theory?

    Drift models say it isn’t possible. Of course, you deny the validity of those models.

    But I’ll say that I actually agree with a couple of your points.

    Just out of curiosity…did you get the permission from Ross Coulthart before you posted the contents of an email exchange with him?

  337. airlandseaman says:

    Rob: Re your statement:

    “ALSM, time for a reality check. You have invested so much intellectual capital in the unpiloted descent theory, your refusal to accept you could be wrong after all will stymie any chace [sic] of a successful search. You are in an unenviable position.”

    You continue to make statements about events and people (including me) that are pure BS. You are either totally incompetent, or deliberately making up lies about me. I have NEVER claimed with any certainty that the plane was not piloted during the final descent. Got that? NEVER! You just can’t seem to accept the fact that the BFO data is indicative of a rapid descent whether piloted or not.

    That said, if I consider all of the available information (not just the BFO data in isolation), I do believe it is more likely than not that the descent was unpiloted. But I have never claimed that to be the only possible scenario, as use accuse me of. Try to get that through your thick head.

  338. Don Thompson says:

    @Victor,

    One further observation and comment, before I drop the subject. Without a satisfactory explanation from Malaysia, no firm conclusion is possible and I am only to aware that Malaysia has failed to answer the question you posed many, many months ago.

    The Factual Information report, of March 2015, in section 1.1/figure 1.1B (as I quoted above) describes:

    9/ The primary target (military radar) appeared to track west-northwest direction joining RNAV Route N571 at waypoint VAMPI then to 10Nm north MEKAR

    The observation: a direct track to MEKAR from Penang, or Palau Perak where a visual sighting was alleged, as later depicted for Str of Malacca transits does not intercept VAMPI and then follow N571.

    The comment: Malaysia’s early progress to properly establish a credible investigation team remain one of the many unknowns. The Malaysian response was a mess. Deputations from the air force, MAS, and DCA made their way to the Metropark Lido Hotel in Beijing. When the snappily titled “Malaysian ICAO Annex 13 Safety Investigation Team for MH370” did eventually form, it screwed up on this issue (among others).

  339. Victor Iannello says:

    @Don Thompson: Yes, I am aware of the FI wording, which I think just further clouds things. That’s why I say the validity is uncertain, and I would not use the Lido radar image to prove or disprove any theory. As we’ve seen, others feel differently.

  340. Barry Carlson says:

    @Don Thompson,

    … and it appears that the RMAF have a vested interest in what happens if and when the aircraft is located. The “Malaysian ICAO Annex 13 Safety Investigation Team for MH370” look like they will be consigned to Davy Jones’ Locker.

  341. Don Thompson says:

    @Rob,

    Bullshit detectors

    Intuition

    Hubristic bollocks, to reflect a term you introduced to this discourse.

    Good luck on your walk down lonely avenue. Oh wait, not lonely, you’ll have Ge Rijn for company. ROTFLMAO.

  342. Don Thompson says:

    @Barry C,

    The rumour mill has it that there is potential conflict between DCA and the RMAF. I have no idea for the veracity of those rumours.

    Peter Lloyd of ABC, who spent time in KL during the period around the 4th anniversary, reported that bi-lateral agreement is being sought, by Malaysia DCA, for Australia to take the lead in recovery, debris examination and identification of remains should the debris site be located.

  343. Niels says:

    The failure to find wreckage around S35 doesn’t make the CSIRO drift studies less important. In fact, if you remove the “boundary condition” of the assumed efficacy of the aerial search, it looks to me that CSIRO’s drift models indicate a possible crash area north of S36 (non-detection of high windage items on WA coast), south of S30.5 (timing of flaperon find), and possibly north of S33 (non-detection of low windage items on WA coast). An area between S30.5 and S33 is well compatible with the Inmarsat data and with the UWA (prof. Pattiaratchi) prediction. All in all I remain hopeful something will be found in coming months in the area north of S33.

  344. Barry Carlson says:

    @Don Thompson,

    If the Australian government agree to taking the recovery / examination lead, I would expect the ATSB to ensure that the agreement protocols allow them to publicly report their own conclusions if they materially differ from those given by the Malaysian DCA.

    Hopefully this does happen, and the RMAF are excluded from any interference – particularly if something happened in which they were or can be implicated.

  345. airlandseaman says:

    Niels: Good summary.

  346. Brian Anderson says:

    @Rob,

    “It was this that helped me identify the path that the Captain actually flew, the path toward waypoint S41 E88”

    You keep repeating the same story, with absolute certainty in your own mind. Isn’t this the same confirmation bias that you continually accuse others of having? It gets tiresome.

  347. Andrew says:

    @Rob

    RE: “On the B777, do you know which bus/buses power the outflow valves and aircon pack inlet valves?”

    Each outflow valve (forward and aft) has two motors, left and right. The left motors are powered by the 28V DC Capt Flt Inst bus, the right motors are powered by the 28V DC F/O Flt Inst bus.

    The two air conditioning packs (left & right) each have two pack flow control/shutoff valves, upper and lower. The valves are pneumatically actuated and electrically controlled by the left and right cabin temperature controllers (CTC). Backup control is provided by the left and right air supply cabin pressure controllers (ASCPC). Power sources as follows:

    L PACK:
    Left CTC – 28V DC L Main bus, 115V AC L XFR bus (Normal control)
    Left ASCPC – 28V DC SEC 2 Battery bus (Backup control)

    R PACK:
    Right CTC – 28V DC R Main bus, 115V AC R XFR bus (Normal control)
    Right ASCPC – 28V DC F/O Flt Inst bus (Backup control)

  348. DrB says:

    @Don Thompson,

    Thanks for explaining the ACARS downlink. In order to re-establish an “active session” after the SDU is logged on, so that downlink reports would automatically be sent, what else needs to be done? Check the SATCOM box?

  349. DrB says:

    @VictorI,

    That was a good observation you made regarding the lack of information in the RMP report on the passenger’s phones. Perhaps all those numbers were not thoroughly checked for cell tower connections, and this would be useful to know. I think if any call or text message had been received from the aircraft in flight, we would have heard about it long ago.

  350. DrB says:

    @David,

    Thanks for addressing the possibility that, in certain failure modes, the cabin altitude could reach a stable point sufficiently below the flight level so as to induce hypoxia.

  351. DrB says:

    @TBill,

    You said: “If a comment like DrB wants to make, that intentional event is ruled out because DrB does not feel what we saw is intentional, was peer-reviewed by Greg Feith, I do not think the statement would stand up to peer review.

    I didn’t realize someone else could do my “wanting” and “feeling” for me.

    I guess you did not read my post yesterday, in which I said: “I have continued to point out the several inconsistencies in both hijacking and accident scenarios, and I don’t see a clear preponderance of evidence on this issue. I am on the fence, as they say, and I think it is a mistake to rule out an accidental cause or a hijacking cause without strong evidence. At this point no one knows what happened . . . .”

    I continue to point out the arguments for and against the two main theories. Most people on this blog have decided long ago that the pilot did it, and they aren’t interested in hearing arguments against it. I don’t claim to know, but I’ll be less surprised if he did it than you will be if he didn’t. As a famous quote misattributed to Mark Twain says: “It ain’t what you don’t know that gets you into trouble. It’s what you know for sure that just ain’t so.”

  352. Ge Rijn says:

    @Niels

    Thanks for that summery.
    On the CSIRO drift studies still being important I made a same kind of comment some days ago. West of ~35S (from 7th arc) the ‘small’ band of current makes a sharp bend to the north and then ~due east along Broken Ridge between ~30S and ~33S before going north-east and turning to the west again.

  353. Ge Rijn says:

    Referrence to previous comment:
    CRIRO Drift Report III

    https://www.atsb.gov.au/publications/investigation_reports/2014/aair/ae-2014-054/

    See the graphic on page 7.

  354. DrB says:

    @TBill,

    You said: “He would probably say, accidental hypoxia is counter-indicated by the following:
    (1) Crew would don O2 masks”

    Tell that to Payne Stewart. Oh, sorry, you can’t. He’s already dead.

  355. Ge Rijn says:

    And to mention once angain; this CSIRO drift study would fit quite well with the position of the ‘Blue Panel’ and associated spotted debris field of 13 items.

  356. DennisW says:

    @DrB

    The lack of COMS makes it difficult to embrace an accident scenario.

    The simulator data makes it impossible.

    Looking at statistics only 1 in 5 hull losses are due to an issue with the aircraft.

    So, in summary the accident has a low probability if we had no information. It has a negligible probability when you consider what we do know.

    Of course I reminded of the true Mark Twain quote – “Never argue with stupid people, they will drag you down to their level, and then beat you with experience”.

  357. TBill says:

    @DrB
    I miss Payne Stewart. You path work is very important to me. You asked for comment. So I gave some, because sometimes your text makes some claims, so since you asked, I tried. You said “That set of circumstances does not appear to me consistent with a hijacking by the Pilot.”

  358. ErikN says:

    Conversations reaching a fever pitch. Hopeful all can dial it back somewhat to the realm of constructive. Saddened by breaches of decorum in posting private conversations as well as by seeming disrespect for the accidentally deceased before their time. May a spirit of solidarity toward the purposes of the blog and answers for NOK and the aviation world reign.

  359. David says:

    @TBill. Little credence has been given to hypoxia being the cause of the crash because it has been assumed that hypoxic pilots would become unconscious before completing the navigation and switching apparent.

    To iterate, if instead of assuming a rising cabin altitude, which would lead to a brief period between hypoxia onset and unconsciousness its onset was in a stable cabin altitude, there might well be an extension of consciousness.

    At 4:48 pm on the 22nd @Victor responded a quote from Dr B about extended hypoxia, the quote being:
    “If this is possible, then I would say the evidence we have is more consistent with an accident scenario than a hijacking scenario.”

    His response:
    “I agree. However, I seriously doubt it’s possible. I’d be interested in any data.”

    To my knowlege there are no data but there is the P3 anecdote and at least there are grounds for supposing a stable mildly hypoxic cabin altitude is possible, a beginning.

    That does not mean that even were there hypoxia which extended to the final turn, there would not be conflicting evidence and opinions still but that applies to most if not all theories right now.

  360. Victor Iannello says:

    @David: As I said in my next paragraph, Of course, there is also the extraordinary coincidence of the simulator data, which has captured snapshots of a flight that proceeds from KLIA, up the Malacca Strait, past the Andamans, and turns to the south, and ends in fuel exhaustion in the deep SIO.

    That’s hard to dismiss.

  361. David says:

    @Victor. Fair enough.

  362. sk999 says:

    All,

    Now that we know that the Lido image was explicitly NOT MH370 (according to a confirmed report from an anonymous “Malaysian official”) and, in fact, was FAKED (according to Richard Godfrey), I thought it worth revisiting the very beginning of the route. In particular, why does it start at some seemingly random point WNW of Penang?

    In a report from Nov 20, 2016, I previously supplied a digitized version of the Lido route as best one can, given the fuzzy, ambiguous nature of the points. Recently, the Geoscience website, “The data behind the search for MH370” has an image with another rendition of the radar route that is the highest resolution yet (scroll down partway to find it.)

    https://geoscience-au.maps.arcgis.com/apps/Cascade/index.html?appid=038a72439bfa4d28b3dde81cc6ff3214

    The first several points of the Lido route occur before 18:01:49, and thus overlap track BE 144, which covers the turnback across Peninsular Malaysia. I have updated the old report with a new comparison. Everything (ATSB route, Geoscience route, Lido route) line up during this time. (Remember, this is before the straight-line portion of the route starts).

    Approximate timing information can be obtained by matching longitude against that prediced by the flight path reconstructed from Fig 4.2 of Bayesian Methods. The first Lido point is at longitude 99.293. What is the corresponding time? 18:00:00. I cannot vouch that this time is accurate to the nearest second. However, it suggests that whoever pulled the data used to make the Lido image from the RAT-31DL log selected a starting time of 02:00:00 local time.

    Remeber, this is all fake data, NOT MH370.

    https://docs.google.com/document/d/14hleZyx1pUPL44yaeHKt6jnSQ3DbgRq2zibbKkFLq2c/edit?pref=2&pli=1#

    (Scroll down to Nov 20, 2016 report to get the updated version.)

  363. Don Thompson says:

    @sk999

    Lido, the gift that keeps on giving, ground hog day.

    And in another forum the USS Pinckney question reappears.

    Ho-hum.

    @Erik suggested “Hopeful all can dial it back somewhat to the realm of constructive.

    ☆☆☆☆☆

  364. Victor Iannello says:

    @sk999 said: Remeber [sic], this is all fake data, NOT MH370.

    That’s your statement, not mine. I assign an uncertainty to it that you might not.

    Just out of curiosity, what do you think the path was between 18:22:12 and 18:41?

  365. TBill says:

    @David
    Hypoxia is an important topic re: MH370. I do not mind discussing at all. I feel for the future Cabin Temp, Pressure, Outflow valve settings, etc should probably be reported by ACARS (if out of spec). The interesting thing about Helios, which was a partial depressure case, was nobody to this day seems to know what the cabin temp was, apparently it is not measured in the recorded data. I am also interested in the depressure-thru-hole calcs.

    Hypoxia possibly comes into play after 19:40 on MH370, as well as sooner.

    We should be free to discuss any technical aspects. If we can find any logic or fact that improves our understanding one way or the other, that is why we are here. Victor I take as the mediator at least for me to say if the new idea is strong enough to disrupt the old ideas.

  366. Victor Iannello says:

    @TBill said:Victor I take as the mediator at least for me to say if the new idea is strong enough to disrupt the old ideas.

    That’s not my role, although I do intervene if somebody is being repetitive, ingenuous, or disruptive, which does occur at times.

  367. sk999 says:

    Don Thompson: “My wider observation is that detailed publication of information gleaned from military assets is rarely exposed, in detail, by air accident investigations.”

    A counter-example is offered by the NTSB “Airplane Performance Study” in support of the TWA 800 investigation:

    http://www.twa800.com/ntsb/8-15-00/docket/Ex_13A.pdf

    Attachment IV is particularly illustrative – primary (and secondary) radar charts for all objects in the vicinity of the aircraft.

    Many of the same radar sites observed Egyptair 990. However, when the Egyptians wanted detailed performance information regarding these radars, the answers were not as positive:

    “18 requests were made by the Egyptian Team for specific radar data and information. To date, eight of the requests have been refused with the explanation that the data is classified and is not available to the FAA or to the NTSB investigation.”

    http://www.ntsb.gov/investigations/AccidentReports/Reports/AAB0201.pdf

  368. Perfect Storm says:

    sk999 says: “My comment r.e. complex scenarios with magical outcomes was meant to be general, not limited to MH370. Jean-Luc’s Captio scenario is one such scenario. JW’s Northern route is a second. Locally where I live, the police and prosecutors invented a complex scenario to get someone convicted of murder; it took over 3 years to show that the scenario was BS and get the guy released (his alibi was just as he said all along), and the mess is still not cleaned up.”

    @sk999:
    since you happen to be here, could you give me links to newspaper articles (or anything really) about this case. I would be interested in reading about it. Many thanks.

  369. Don Thompson says:

    @sk999

    Thank you for those references. I will read in detail later but on first glance I don’t regard those reports as “counter-examples’.

    The radar sites for which one might expect reservation of performance data would include the joint USAF-FAA long range primary ARSR’s at Nth Truro, Riverhead, and Trevose together with the infrastructure supporting those (per the response to the Egyptian team enquiry, exhibit 3).

  370. DennisW says:

    @all

    The flight path from IGARI to the 18:25 range ring can be inferred from other data, and I have never made any assumptions relative to it based on the radar data. If we believe the cell phone registration in Penang, that gives a rough data point for the IGARI to Penang flight path i.e. the cell phone range is on the order of 1/3 of a degree in latitude and Mr. Doppler requires the longitude to be very nearly the same as the cell tower (cell tower position nearly orthoganal to flight path). The flight path from Penang to the 18:25 ring had to be almost straight West based on distance and aircraft maximum speed. So the validity of the radar data has never really been an issue for me. The flight path from 18:25 to 19:41 is a much bigger question, and the radar data was never considered to be helpful in resolving it.

    Basically, the LIDO slide should be ignored. It contains no additional information even if it is valid.

  371. Victor Iannello says:

    @All: Another possibility that I previously considered is that the Lido radar data is correct, except the radar timestamps are not synchronized with the Inmarsat timestamps. Using the Inmarsat timestamps as a reference, if the radar data is 1m53s late, that is, the last radar point occurred around 18:24:05 on N571, then the BTO and BFO data is matched quite well without invoking lateral offsets, or turns, or changes in speed. That seems like a big discrepancy, but maybe it occurred.

  372. Ge Rijn says:

    @VictorI

    Interesting. Then the last radar point would ~aline with the onset of the re-log on. That would be an odd coincidence imo.

  373. Lex Luthor says:

    Interesting. Then the last radar point would ~aline with the onset of the re-log on. That would be an odd coincidence imo.

    Even more interesting: How can someone posting 5 times as much as everyone else here have missed this realization discussed for years on both blogs basically every day?

  374. David says:

    @TBill. A couple of points on Helios. First I have it in mind that the outflow valve(s) was fully open so most likely hypoxia occurred during a rising cabin pressure, that hastening unconsciousness. Second, a cabin crew member survived the cold.

    If in MH370 the outflow valves control failed or reverted to manual during climbing, an hypoxic cabin altitude might have been reached at close to maximum climb. Aside from the effect of that on reaching unconsciousness there would be no adiabatic expansion, as with Helios.

    Still I will refresh my memory on Helios thanks, particularly as to confusion with the warning system.

    I for one do want to give the impression that hypoxia-as-cause is likely but I think what Dr B has raised warrants a fresh look.

  375. DennisW says:

    @David

    A commericial pilot is not going to become a hypoxia victim. Give these people some credit. They are professional and well-trained. Hypoxia is a non-starter as a cause of this “accident”.

  376. Mick Gilbert says:

    @DennisW

    Re ‘A commericial pilot is not going to become a hypoxia victim.

    History would most assuredly suggest otherwise, Dennis. Either Don or Mike provided a list of hypoxic crew events involving commercial flights some time back. They’re rare but they happen.

    If you want to get a feel for how hypoxia impacts decision making have a read of the Kalitta Air 66 ATC transcript(https://fearoflanding.com/accidents/accident-reports/hypoxia-on-kalitta-66/). The stand out has to be the hypoxic Captain, who was sitting beside an unconscious and convulsing First Officer, reporting ‘Unable to control altitude. Unable to control airspeed. Unable to control heading. Kalitta six six. Other than that, everything A-OK.
    It sounds a bit like some contributors; ‘Unable to account for data, unable to account for evidence, other than that everything A-OK.’

  377. DennisW says:

    @Mick

    Rather than a list of how hypoxia affects people. How about a link to the actual crew events attributed to hypoxia.

  378. DennisW says:

    @Mick

    Basically, you will come up very empty. Think before you post.

  379. Mick Gilbert says:

    @DennisW

    Basically, Dennis, you are showing that appalling lack of courtesy that bobs up from time to time (splitting wood?), coupled this time with an uncharacteristic lack of scholarship. Did you even bother looking at the Kalitta incident?

    Mike or Don’s previous list wasn’t of symptoms but of hypoxic events involving commercial flights. Since you seem disinclined to look for yourself here’s a quick summary list (if reading it presents as a problem perhaps you can try Google Text to Speech):

    1. Helios Airways Flight 522. B733, en-route, northwest of Athens Greece, 2005: 6 crew and 115 passengers perished due to lack of pressurisation. With the crew incapacitated by hypoxia, the aircraft flew on under flight management computer and autopilot control until it ran out of fuel and crashed.

    2. Malmö Aviation RJ1H, en-route, South West of Stockholm Sweden, 2007: Flight crew failed to notice the aircraft was not pressurised after take off until the cabin crew advised them of automatic passenger mask deployment. Incident compounded by partial failures in passenger oxygen systems, portable oxygen equipment and pressurisation warnings.

    3. ‎Kalitta flight KFS-66, the pilots of a Kalitta Learjet flying into Ypsilanti, Michigan found themselves suffering from hypoxia. Reference already provided.

    4. ‎ Sunjet Aviation Learjet 35, registration N47BA, 1999. The Payne Stewart crash.

    5. ‎Beechcraft 200 Super King Air, 2000. Departed Perth for a flight to the mining town of Leonora, Western Australia. After hypoxic flight across the country the aircraft crashed 1,500 nm away near Burketown, Queensland, Australia resulting in the deaths of all 8 occupants.

    And a military one for good measure:

    6. F/A-18 Hornet, RAAF, 1991. Aircraft went missing while returning to RAAF Base Tindal, in the Northern Territory, crashed 600 nm away on Cape York.

    What were you saying about thinking before posting?

  380. DennisW says:

    @Mick

    I asked for links not your description.

    Your descritpions are hardly representative of commercial airline flights. You are grasphing at straws basically.

    Do you feel good about your response? Probably not.

  381. Victor Iannello says:

    @Mick Gilbert: Do you know of any case where an emergency was declared, there was no communication, hypoxia occurred, there was a failure to descend, and turns occurred over a period of about 1h20m? That would be the fair comparison, it if exists.

  382. Mick Gilbert says:

    @DennisW

    I’m not your research assistant, Dennis. I provided a link to the Kalitta incident and its apparent you didn’t read it.

    You made the statement, ‘A commericial pilot is not going to become a hypoxia victim.‘. That is patently and demonstrably incorrect.

    And I’m feeling pretty good about my response, thanks Dennis.

  383. Mick Gilbert says:

    @Victor

    Seriously, Victor?? To the extent that there was a perfect prior exemplar for MH370 don’t you think that we might have heard of it by now? I’m not here to prove that hypoxia was the cause, I was rebutting Dennis’s statement that ‘A commericial pilot is not going to become a hypoxia victim.

    However, to the extent that you want an example of a case where an inflight problem was observed, there were subsequently no communications, hypoxia occurred, there was a failure to descend, and turns occurred about 1h 20m later, then it’s Helios 552.

  384. DennisW says:

    @Mick

    The incident you referred to is not reprenstative of a commercial airline incident. Nor none of the others. I have to now put you in the whacko category.

  385. Victor Iannello says:

    @Mick Gilbert: I’m not trying to back you into a corner. I’m trying to find fair comparisons. I also asked @DrB for cases in which there was prolonged hypoxia. He didn’t seem to be offended by my request.

    Please let me know if I have this right. In Helios 552, there was a low pressure alarm misinterpreted at 12,000 ft, and a communication with ground crew to troubleshoot the alarm, with no turnaround. For MH370, communications were normal at FL350, yet there was a subsequent turnaround and no communication as to why. So if there was a mechanical failure, the crew knew there was a significant problem, and decided to turn back towards Malaysia. Yet there was no communication and no descent. That strikes me as very different than Helios 552.

  386. Andrew says:

    @DennisW

    RE: “The incident you referred to is not reprenstative of a commercial airline incident. Nor none of the others.”

    Helios Airways Flight 522 was a scheduled passenger flight from Larnaca to Athens. Similarly, Malmo Aviation Flight TF009 was a scheduled passenger flight from Stockholm to Gothenburg. Why are they not representative of commercial airline incidents?

    I’m curious to understand the reasoning behind your statement, not trying to infer these incidents are similar to MH370 in any way.

  387. DennisW says:

    @Andrew

    I am surprised by your support of a hypoxia related cause.

  388. Andrew says:

    @DennisW

    Huh? Please read the last sentence of my post. I am only trying to understand why you think none of of the incidents Mick quoted are representative of commercial airline incidents, when two of them were clearly scheduled airline flights.

    At this point I do not support a ‘hypoxia related cause’. Can I make that any more clear?

  389. David says:

    @Victor.
    A premise for looking into the hour-and-beyond hypoxia-while-active possibility is that the cabin altitude did not rise appreciably after hypoxia onset.

    Should there be information to the effect that;
    . such events have occurred,
    . there was no hypoxic activity extending beyond an hour say, and,
    . there was no communication or descent but there was navigation or switchery;

    then depending on how numerous and representative they were, that would constitute evidence that hypoxic activity could not extend that far. Even better were there were medical research which demonstrated that.

    @DennisW’s “A commercial pilot is not going to become a hypoxia victim.”

    You have had the helios example before you as has been pointed out several times.
    I hope tomorrow is a better day.

    “Hypoxia is a non-starter as a cause of this “accident””.
    You are a top down thinker who makes up his mind. Good for you.
    Others, like Lord Keynes, look into changes to the basis for their beliefs with a fresh mind. Good for them too.

  390. David says:

    DennisW. Second last word should have read ‘us’

  391. DennisW says:

    @Andrew

    The incidents Mick cited were not reprentative of mainstream carriers. They were not even invesigated by mainstream people. You can find examples of anything if you look hard enough. Is that what you want to hang your hat on?

  392. Andrew says:

    @DennisW

    Perhaps you should have said that in the first place, instead of making the massive generalisation that “a commercial pilot is not going to become a hypoxia victim”, which is utter nonsense.

  393. DennisW says:

    @Andrew

    Have you read the detailed reports on the hypoxia incidents? I have. You want to rank yourself in their shoes?

  394. DrB says:

    @DennisW,

    You said: “The lack of COMS makes it difficult to embrace an accident scenario.”

    Are you aware that, on this aircraft, all audio goes through a single Audio Management Unit in the left AIMS Cabinet?

    That means a singe-point failure can prevent all voice radio calls (seems like a poor design choice to me). I don’t know if you can still transmit a carrier by keying the mike. I suppose one could try sending Morse with the mike key, but who knows if anybody could decode it in real time.

    I thnk the AMU operation depends on the ARINC 629 bus in that cabinet, so if that bus goes down, you don’t have voice communications capability. I don’t know whether this would prevent an incoming sat call from being answered, such as occurred at 18:40. Perhaps Don or Andrew could answer that question.

  395. DennisW says:

    @Andrew

    At this point I do not support a ‘hypoxia related cause’. Can I make that any more clear?

    How else could I get you state that? It is more insulting to ZS to accuse him of that (hypoxia related incident) than it is to accuse him of being a mass murderer. Professional pilots do not let something like that happen. That is my only point.

  396. DrB says:

    @Andrew,

    Two questions:

    1. Did 9M-MRO have a manual deadbolt lock on the cabin door (that could not be defeated by removing power from the electric lock)? It appears that some Delta B777s have this feature.

    2. Is it possible to defeat the electric lock from the EEC or the galley?

    Obviously the overall question is can one pilot really keep the cockpit door locked if a second pilot is trying to unlock it?

    If you can’t post publicly on this subject, you can email me instead.

  397. Andrew says:

    @DennisW

    RE: “Have you read the detailed reports on the hypoxia incidents? I have. You want to rank yourself in their shoes?”

    Of course I have read them. I’d like to believe that my training and experience would help prevent such events, but I am not so arrogant as to believe that it couldn’t happen to me. If there’s one thing that 35 years of flying has taught me, it’s that hubris is a dangerous thing in aviation.

    RE: “It is more insulting to ZS to accuse him of that (hypoxia related incident) than it is to accuse him of being a mass murderer. Professional pilots do not let something like that happen.”

    I do not agree. I’m NOT suggesting it happened with MH370, but what if some combination of events causes the cabin to depressurise (or not pressurise at all) and the cabin altitude warning system then fails, as happened in the Malmo Aviation incident? The onset of hypoxia is insidious and while professional pilots learn about the symptoms during their initial training, few have ever experienced it. I did experience it several times during chamber runs in my military flying days. Even so, I’m not sure I would recognise it in time if I didn’t know it was coming.

  398. Ge Rijn says:

    @Lex Luthor

    18:24 is not the same as 18:22.
    And don’t worry. I’ll post a lot less in the coming time.
    I think I tried everything I wanted to get my points across.
    Imo the next ~two weeks are crucial when OI completes the search along the trenches of Broken Ridge.

  399. Andrew says:

    @DrB

    I’ve sent a reply to your gmail address.

  400. Mick Gilbert says:

    @Victor Iannello

    Victor, I understand that you wouldn’t have been trying to back me into a corner. I was simply astounded that you would have asked me for a directly comparable prior exemplar of hypoxic flight that matches what we understand about MH370. There are no directly comparable prior exemplars for either an accidental or deliberate cause; if there were then there’s a better than even money chance that we would have solved it by now. As I have previously stated, ‘Whatever the cause, MH370 represents a first of type (or sub-type) event.‘.

    Your understanding of Helios 552 broadly accords with mine; manifestly it is not the same as MH370 and, perhaps more to the point, I never said that it was.

  401. Richard Godfrey says:

    SC has arrived back in the search area and stopped, around 12 NM north of where it left off before the recent trip to Fremantle.

  402. Don Thompson says:

    @DrB, @DennisW

    The AMU is a single LRU but, internally, it comprises three separate circuit cards, one for the each crew station (two pilots plus observer). The chassis, connector board & rear connectors are common. Power for each station card is separately sourced.

    The single Left Systems Bus enables fault reporting the CMCF, DME tone signalling to the AMU, and SELCAL set/reset. Note that only a single Systems Bus connection is provided, suggesting that the interface is not critical to the AMU operation.

    The AMU is not in the L AIMS cabinet, it is located in the opposite equipment rack.

  403. HB says:

    @Don, sk999, Victor et al
    Re Radar secrecy, Lido, indonesia radar etc.
    Maybe going back to the fundamental assumptions and regurgitating past questions but I still have three points of interests on these:

    1) As a matter of interest, what sort of outputs can be expected from these mil radars, screen shots? what else? are all outputs relevant for secrecy? Can an individual plane target output be isolated from the rest of the data and declassified?

    2) Is there really a case to declare secrecy on these given the location is public domain and probably the type/range/capability also?
    Refering to the article below the Indonesian Air Force spokesman Air Commodore Hadi of Indonesia was proud to indicate the coverage of his own radars. How big a secret it can be?

    3) Also, re Indonesia radar, before the Inmarsat data was endorsed, the Jakarta Post (http://www.thejakartapost.com/news/2014/03/14/ri-radar-did-not-detect-mh370-malacca-strait-air-force.html) quoted Indonesian Air Force spokesman Air Commodore Hadi Tjahjanto saying
    “Indonesian Air Force’€™s radar unit in Lhokseumawe, Aceh, did not detect the missing MH370 in the area where the Malaysian military suggested as being the plane’€™s last detected position around Penang waters.
    ‘€œOur radar information has been shared with our Malaysian counterparts,’€ he said.

    It seems from the above that either the plane avoided coverage, or the radar was not capable to detect the target. However, given that the information was “shared” may imply that the data exists and that the data recording was not turned off nor malfunctioning since the information was “shared”.

    Interestingly, the same article mentions below: “The Indonesian Air Force’€™s Boeing B737-2×9 Surveiller was still searching the plane regardless of the radar’€™s lack of detection, Hadi said. ‘€œWe just finished our first search today. ”

    (a) Based on that, if we assume that both Malaysia and Indonesia positions are factual, should the Indonesia’s radar lack of detection be considered a contraint for candidate flight paths? Why was that constraint never considered as one possibility in the search strategy? Was this discrepancy ever resolved?

    (b) My second question is whether a candidate flight path exists that matches both Inmarsat data and the constraint of no-detection for the Indonesia radars (still the official position from Indonesia) regardless of Lido image validity?

  404. Victor Iannello says:

    @HB: The lack of Indonesian military radar data is unexplained. The simplest explanation is it is turned off at midnight, regardless of ambiguous statements made by Indonesian officials. You say that given that the information was “shared” may imply that the data exists and that the data recording was not turned off nor malfunctioning since the information was “shared”. We don’t know what was shared, i.e., from what radar stations and for what period of coverage.

    As Dennis said, even if we ignore the Lido data, the paths would be roughly the same, as I showed in a previous post. The value of the Lido data is to give us an indication of how the plane was flown before it left radar coverage.

  405. Victor Iannello says:

    @Mick Gilbert: There’s no reason to be astounded. I just wanted to be sure that I (and others) weren’t missing anything when hypoxia is touted as an explanation. (I am not nearly as familiar with past aviation accidents as others here seem to be.) If hypoxia did occur and that led to the disappearance, it occurred in a way that was much different than anything else we’ve ever seen.

  406. paul smithson says:

    Indonesian Air Force’€™s radar unit in Lhokseumawe, Aceh, did not detect the missing MH370 in the area where the Malaysian military suggested as being the plane’€™s last detected position around Penang waters.

    That is not an assertion that they had no data or detection. Only that they don’t have something that matches MY’s claims.

  407. HB says:

    @Victor
    Is there any possible candidate flight path taking this constraint on as opposed to assuming it was off?

    There were many official statements in the Indonesia press confirming the non capture position and also commenting on the accuracy of their radar data. Indonesia officials tone across all the statements portrays a high level of confidence that no foreign airplane was captured. One of the statements specifically confirmed that. We dont know what data was shared but we dont know it was turned off either. Why not considering the two possibilities?

  408. Victor Iannello says:

    @HB: The possibility that Indonesian radar was turned on, properly functioning, and missed the aircraft would mean that either the satellite data is completely wrong, or the plane was flying at a very low altitude at an airspeed much faster than Vmo and burning a lot of fuel. It would also mean that the Malaysian military data was fabricated, because if Indonesian hadn’t seen the low flying aircraft, neither would Malaysia.

    Basically, if you want to reconstruct flight paths based on non-detection of functioning Indonesian radar, you have to ask yourself which data sets you would be willing to throw out? Satellite data? Malaysian radar? Aircraft performance? Which subset of the data is most reliable and consistent?

  409. Victor Iannello says:

    For those of you that believe the satellite data is a hoax, here’s a story that appeared in the Malaysia Gazette. I reference it only so you are aware of what some are claiming.

    KUALA LUMPUR – The MH370 mystery plane lost on flights from the Kuala Lumpur Interstate to Beijing on 8 Mar four years ago was blown in North Sumatra.

    The indictment was made by the International Competition Officer for MH370 Search, accompanied by 14 countries after receiving current data and witnesses who have given the facts believed to be used.

    Because of this, Datuk Dr. George Jacob representing the free agency asked 14 countries that lost their citizens to weigh to help launch the congregation at the stated new location.

    He who is now in North Sumatra said, the officers had met with witnesses who accused of seeing the incident.

    In fact, witnesses regarding giving the oath affidavit had seen the plane in the morning on the day of its loss.

    According to him, as many witnesses claimed to have appeared a plane with no sound of engines with thick black smoke puffed and flew in a state of staggering left and right before crashing in the south of the Hindi Ocean at about 7 am local time.

    Based on the witness’s stories, the plane was charged with drowning as well as falling, George told MalaysiaGazette.

    The plane was charged about 60 meters from the sea during his plunge.

    If they arrive and are in the area for two days, witnesses can still see black smoke in waters believed to be the last location of MH370.

    However, witnesses do not appear to have any human bodies or debris that arise.

    However, they have recoded the sequence of events and readings of the global messaging system (GPS), conjured up the shadow of the fuselage plunging into the ocean.

    George said, holding out witnesses using only fishing boats without high-tech sonar or technical equipment while in it, they were forced to take a few days to declare the loss of MH370 aircraft.

    Witnesses have allegedly proclaimed the case to the ruling party and Malaysia, including the international media that pledged to assist, but so that today there is no communication or further reply.

    George said, do not want to bear the shame, eventually the witnesses took silence so that they were impressed by the free body.

    During the incident, the MH370 aircraft carried 239 passengers including a child ship.

    Operations seeking answers to the mystery of missing aircraft are the largest and most expensive in the history of the aviation industry.

    The mission, chaired by Australia, has explored the Indian Ocean over 46,000 square miles of boardingers in excess of US $ 160 million, turning ships and aircraft.

    During 1,046 days of operation the terminations ended with no results and only recently the operation sought the MH370 to be run by a seabed seafloor, Ocean Infinity Limited with a ‘first, pay for’ method. – MalaysiaGazette

  410. Andrew says:

    @DrB

    RE: “I thnk the AMU operation depends on the ARINC 629 bus in that cabinet, so if that bus goes down, you don’t have voice communications capability. I don’t know whether this would prevent an incoming sat call from being answered, such as occurred at 18:40. Perhaps Don or Andrew could answer that question.”

    As Don noted, the AMU is not located within the L AIMS cabinet; it’s a separate ‘box’. The AMU interfaces with the L & R AIMS cabinets for DME pairing, fault reporting and SELCAL functions, as Don also mentioned, but that’s all. That data is transmitted via the L SYSTEMS ARINC 629 bus. The failure of that bus would affect the functions mentioned, but would not affect the ability to communicate via VHF, HF or SATCOM.

  411. DennisW says:

    @all

    I apologize for my hypoxia stance yesterday. I should have simply let the notion lie. Like some people have a hot button with Rob’s assertions, I have a hot button with hypoxia scenarios. Two obviously deliberate events separated by 1.5 hours, with no communications of any kind, along with a hypoxia “rehearsal” on the simulator are too much for me. I any case, I will not comment further on hypoxia posts.

  412. Niu Yunu says:

    RE: Updated animation by Richard Cole
    https://twitter.com/richard_e_cole/status/977965179479756800

    @Richard Cole:
    I love your animations!

    Does anyone of you know, why the ship sometimes approaches the AUVs to collect them – instead of waiting at the edge of the search area until the AUVs come back on their own, like most of the time ?

    This happens for instance at 0:29 and 1:02.

    So instead of this (* = AUV)

    *
    *
    *
    SHIP > > > > > > > > *

    the ship approaches the AUV:


    * * * *SHIP
    SHIP
    SHIP
    SHIP

    I hope my doodle makes any sense.

    I’m asking because I guess it’s not very fuel-efficient to criss-cross the ocean in this matter vs. resting on the sidelines and letting all the AUVs come back themselves.

    But maybe it’s more time-efficient this way and this trumps any fuel considerations ?

  413. Niu Yunu says:

    ok the doodle doesn’t display correctly. trying again:

    Instead of this (* = AUV)

              *
                      *
                                *
    SHIP > > > > > > > > *
    —— search area edge ——

    the ship approaches the AUV:

            *          *          *        *SHIP 
                                SHIP
                  SHIP
    SHIP
    —— search area edge ——

  414. Victor Iannello says:

    I have been privately asked about a potential search for MH370 north of Sumatra. If there are parties that believe that would be fruitful, I encourage them to organize a search of those locations. The technical requirements, the logistical support, and the cost would be a small fraction of past and ongoing efforts in the SIO.

  415. Victor Iannello says:

    With elections around the corner, Malaysian politicians want to make sure that only accurate information reaches their citizens, so they have introduced legislation that would impose a 10-year sentence on anybody creating or disseminating what authorities classify as fake news pertaining to Malaysia. This extends to fake news created or promoted by foreigners.

    https://sg.news.yahoo.com/malaysia-proposes-10-years-jail-fake-news-064024771.html

  416. Ge Rijn says:

    @VictorI

    This turns things upside down to me. If you support this (which is ok on itself) you seem to hold the possibility open of Inmarsat-data being spoofed and drift-analysis being based on planted debris in a way.

    On the other hand the story reflects the black smoke trail Kate Tee observed.

    I suggest maybe this starts to be an option if the current search has failed. But if it does all options are open again.

  417. Victor Iannello says:

    @Ge Rijn: I think it has a very low probability of succeeding, but who am I to try to stop it, and why would I? I’d also support Jeff Wise excavating Baikonur. In fact, I’d strongly encourage it.

  418. DennisW says:

    @Victor

    Continuing the Mark Twain philosophy…

    “If you don’t read the newspaper you are uninformed. If you do read the newspaper you are misinformed.”

  419. Richard Godfrey says:

    @Richard Cole

    Many thanks for the second animation of the Ocean Infinity search.

    It portrays very well, what an excellent job OI are doing.

  420. Victor Iannello says:

    @DennisW: Luckily, Malaysia will sort that out for us.

  421. Sabine Lechtenfeld says:

    Dennis, Mark Twain is one of my all time favorites 🙂

    Victor, I think that all searches for the plane which are financially and logistically doable, should be encouraged. Nothing wrong with that. It would eliminate one scenario after the other. I would also be very pleased if the 7th arc near Christmas Island and the Javan coast would be searched eventually if the current search comes up empty. This area is at least compatible with the sat data.

    But an excavation in Baikonur would be the icing on the cake. There might be very interesting things under that soil – even if there is no plane.

  422. HB says:

    @Victor
    Ok i see the importance on knowing what that information that was passed to Malaysia.

  423. Nederland says:

    @HB

    Indonesia has not ruled out they might have detected MH370 but not in their airspace.

    It is entirely possible to reconstruct a flighpath that goes around Indonesian airspace in Sumatra.

    https://www.docdroid.net/GvlrLaV/mh370-waypoint-30.pdf#page=6

  424. Ge Rijn says:

    @VictorI

    I truly admire your diplomatic moderator skills (not being cynical at all).

  425. DennisW says:

    @Nederland

    I like it. 30S has a lot going for it IMO.

  426. Nederland says:

    @DennisW

    My money is on 31S (or close to).

  427. DennisW says:

    @Nederand

    Yes, I saw that. I don’t paint with that fine a brush. 🙂

  428. DennisW says:

    @Victor

    Relative to additional searching, I still do not have a strong opinion of how that would work relative to payment by Malaysia or even if Malaysia has an obligation to pay before they can lay claim to the wreckage. Last time I looked at it I concluded that they, Malaysia, did have such an obligation, but the details were murky.

  429. Victor Iannello says:

    @DennisW: The contractual details are unknown, and might have even changed since the contract was signed. Realistically, OI will continue to search until they’ve had enough, which might be as soon as April. If not found by the end of the OI search, the entire proposed list of endpoints around the world will be viewed as equally valid, which means there will be no concerted effort to search anywhere with a reasonable chance of success. Our last chance to find MH370 is for OI to find the plane soon, which is still possible.

  430. Michael John says:

    The key point for me with search in the NIO off the coast of Northern Sumatra is an aside to the SIO. What I find intriguing is the area around the estimated FMT point was never investigated. ISAT is of course the key indicator of where 9M-MRO went but even after 4 years the information is still being debated. When you look at the flight

  431. Victor Iannello says:

    @Michael John: I encourage you to get together with Dato Jacob George and devise a plan to search north of Sumatra. I think your chances of success are not very good, but my opinion is irrelevant. So don’t waste time and energy debating that here. Go ahead and do it. You have my full support.

  432. Michael John says:

    Path we have Mh370 last heading West into the Indian Ocean. I have looked at the potential of the NIO as an end point & ISAT data aside it seems plausible.

    It’s a shame that Inmarsat still after 4. years have seemingly refused requests to make the entire concept transparent & open to independent scrutiny. Experts follow the guidelines they have provided. But I would like independent experts to validate not only the result those guidelines lead to but also the guidelines themselves. ISAT once said the whole concept was a shot in the dark.

  433. Michael John says:

    Of course Victor but which area is the right place to start?

    Every building block needs a solid foundation. We don’t have that. I certainly do not want to go down the Peter McMahon route. The SIO has the foundation block in the form of the ISAT Data. I can’t speak for anyone else but I would like to see the options for the ISAT Data concept fully exhausted. This blog is about the Isat Data concept so discussions like this 1 are to most commentators here most likely to be an unwelcome distraction.

  434. Peter Norton says:

    Michael John says:
    It’s a shame that Inmarsat still after 4. years have seemingly refused requests to make the entire concept transparent & open to independent scrutiny. Experts follow the guidelines they have provided. But I would like independent experts to validate not only the result those guidelines lead to but also the guidelines themselves. ISAT once said the whole concept was a shot in the dark.

    Haven’t you said this already on 19 January and received a lengthy reply from Don Thompson ?

  435. Victor Iannello says:

    @Michael John: Yes, discussion about crash sites north of Sumatra will not be viewed favorably here, as the group is primarily data-driven. On that basis, I support your efforts to find MH370 north of Sumatra, but I don’t support that discussion takes place here, as this is not the appropriate forum, and it will be a distraction. Dato Jacob George says there are witnesses with GPS coordinates of the crash site, so teaming up with him is the obvious choice, and that will provide you a place to start looking. Good luck!

  436. HB says:

    @Nederland,
    You may be correct, many statements referred to the Indonesian airspace as an additional qualifier. This may add some constraints but not as bad as saying no detection at all.

  437. TBill says:

    @Victor
    …thought we had until June for OI search, but OK.

  438. Mick Gilbert says:

    @Don Thompson
    @DrB

    Don, just a quick clarification on power to the Captain’s, FO’s and Observer’s circuit cards in the AMU. You stated, ‘Power for each station card is separately sourced.‘. Is my understanding of the set-up correct in that two of the three cards are in fact powered from the same source (specifically that the Capt’s and Obs’s cards are powered by the 28V DC Captains Flight Instrument Bus)?

  439. Ge Rijn says:

    @VictorI

    I suggested it in genearl before but I request this now directly to you:
    Are you willing to conduct a same kind of reverse drift study as you did on the debris spotted west of the 7th arc at ~31S but now taking the ‘Blue Panel’ as the subject?
    (Based on the CSIRO drift studies as mentioned in the CSIRO Drift Report III)

    I feel a bit like a lost man calling in the dessert (and I know in a way I am).
    But anyway I ask for I think it can be helpfull and important.

  440. Ge Rijn says:

    ..correction; west of the 7th arc at ~29S..

  441. Rob says:

    @Andrew

    BTW, thanks very much for the pressurisation system info the other day. It was useful.

  442. Don Thompson says:

    @Mick Gilbert, DrB

    Thank you for the clarification, the Capt Flt Inst DC supplies the Capt & Obs station cards, while the F/O station card is supplied from the F/O Flt Inst Dc Bus and Capt Flt Inst DC Bus (diode protected from making a tie). Each supply is provided with a discrete circuit breaker. I should have made that clear.

  443. David says:

    Operational Search Update No 9
    http://mh370.gov.my/en/mh370-underwater-search-2018.

    “A Category 5 Severe Tropical Cyclone ‘Marcus’ had transited through the search area
    over the past week, and it’s anticipated to impact the coming week’s search
    operations. Seabed Constructor is expected to remain on weather standby on arrival until weather and sea conditions improve.”

  444. Michael John says:

    @PeterNorton I obviously missed the response by Don so thank you for linking me to it. Thank you Don for taking the time to write such a lengthy but highly informative comment & I apologize for not seeing it sooner.

    @Victor Just FYI Jacob has given a voice to local Indonesian fisherman who are adamant they saw Mh370 come down in the Malacca Strait. Personally for a multitude of reasons I am sceptical of both the claim & even more so the location. I favour a spot further West around the place where the FMT was estimated to have occurred.

  445. Sabine Lechtenfeld says:

    @Michael John, I wish you good luck.
    But how do you explain the lack of oil slicks, debris sightings and the existence of the satellite data which seem to exclude all scenarios where the plane crashed early without flying for hours into the SIO?
    I don’t mean to be snarky. I’m genuinely curious.

    One of the strangest aspects of the MH370 mystery is, that these early-crash scenarios would be perfectly sensible – if there weren’t the sat data (and the sim data). If we wouldn’t have these data sets, everybody who would’ve claimed that the plane continued to fly for hours into a southern direction until it finally crashed into the SIO, would’ve been totally ridiculed.

  446. Victor Iannello says:

    @Michael John: A search to the north of Sumatra will be based on this genre of evidence. I encourage to you join their team of international experts and not use this space to voice your skepticism. As I said, discussions of this type are not appropriate here. No more.

  447. Victor Iannello says:

    @Ge Rijn: Are you willing to conduct a same kind of reverse drift study as you did on the debris spotted west of the 7th arc at ~31S but now taking the ‘Blue Panel’ as the subject?

    Not really. I would have to resurrect what I did, and it would take some time. But you can follow the same methodology as I did, using the results of CSIRO, which they helpfully made available as KMZ files.

  448. TBill says:

    @Victor @all
    I have some generic satellite phone/PC connectivity questions:

    >>If you owned a Iridium sat phone on MH370, does the SDU unit need to be ON to use it, or is the Iridium phone independent network to a different satellite?

    >>Similar question, how could a PC laptop connect to internet on MH370? Via SDU? or via 3rd party sat connect independent of SDU?

  449. Ge Rijn says:

    @VictorI

    I looked at it (KMZ-files etc.) but I lack the skills to pruduce an animation like you made. I hoped it would be easy for you to repeat.

    All I can deduce from the KMZ-CSIRO graphics is non-flap debris would take roughly 20-25 days to go from 7th arc ~96E/~32.5S(degrees) till ~97.8E/~32.5S.
    And flaperon(high windage) debris would take roughly only 10-15 days to cover the same distance.

    The current in the KMZ-animations here (between ~33S/~30S) shows also the ‘small’ band of current running almost due east along Broken Ridge.

    A rough reverse-drift-analysis made this way already shows the ‘Blue Panel’ could well fit an origine close to the 7th arc.

    I still think it’s worth to take a more precice look.

  450. Ge Rijn says:

    @VictorI

    I have to add that I took the (drift) situation on 8-3-2014 at ~35S (the CSIRO/Griffin satelite image referrence). Followed the drifters along their path from there till ~96E(7th arc) and then started counting the days till ~97.8E.
    So, just extra-polating the situation on that date from ~35S to ~32.5S.
    You follow my logic?

  451. Don Thompson says:

    @tbill

    Inmarsat ≠ Iridium. I’m saddened that, in March 2018, this might not be apparent to you.

    how could a PC laptop connect to internet on MH370?” It couldn’t.

  452. Victor Iannello says:

    @Ge Rijn asked: You follow my logic?

    Sort of. But if OI plans to have your endpoint scanned, what difference does it make? In light of the time ticking away, I am more concerned about justification for endpoints further to the north.

  453. Richard Cole says:

    @Niu Yunu

    Those two occasions you list when the ship passes over the AUV’s estimated position have been interpreted as checks that the AUV’s internal navigation system is still correctly aligned.

  454. Ge Rijn says:

    @VictorI

    Till now it does not seem OI is going to scan my stretched end-point around 32.5S(degrees)/~97E. I hope they will and take the ‘Blue Panel’ and associated debris field as serious as your ~29S debris field.
    In essence there is no difference between both, to take one more serious than the other imo.
    It would be worth a reverse drift study as you performed on the ~29S debris. You reject it. That’s your decision.
    But let’s see what evolves.

  455. Victor Iannello says:

    @Ge Rijn: I don’t know what you are talking about. Constructor is progressively scanning along the 7th arc. It will reach 32.5S latitude well before 29S, if it ever gets there. And I didn’t reject your endpoint. I choose to not study the drift pattern of the blue panel because it will take some time to do, and irrespective of what I find, the associated endpoint on the 7th arc will be searched. Is that so hard to understand?

  456. Ge Rijn says:

    @VictorI

    I understand, but with no further indicating better information the area will probably not be searched till ~97E. And that’s what worries me.

  457. TBill says:

    @Don T
    I know Iridium sat system is different from Inmarsat. But I am not experienced on where an Iridium phone can work, reception wise.

  458. Niu Yunu says:

    Richard Cole says:
    @Niu Yunu: Those two occasions you list when the ship passes over the AUV’s estimated position have been interpreted as checks that the AUV’s internal navigation system is still correctly aligned.

    @Richard Cole: That makes sense. Thank you for helping make sense of it. You are all incredibly helpful.

  459. Peter Norton says:

    > Michael John says:
    > @PeterNorton I obviously missed the response by Don so thank you for linking me to it.
    > Thank you Don for taking the time to write such a lengthy but highly informative comment
    > & I apologize for not seeing it sooner.

    Sure. You are very welcome.

  460. DrB says:

    @Andrew,

    I appreciate the information you provided on the AMU. Now that I understand it correctly, Boeing’s design seems quite reasonable to me. My conclusion is that the simplest explanation for no outgoing radio calls is an unwilling hijacker. For all radios to be rendered nonfunctional requires multiple near-simultaneous failures. I suppose this is possible due to the close proximities of the LRUs in the MEC, but it would require a very improbable event. A third interpretation would be that some or all of the radios did function but were not utilized by hypoxic pilots. That’s starting to become the fallback position to rationalize seemingly inconsistent happenings for the accident scenario, although it still might have happened with the right-sized hole in the hull. This is somewhat similar to the way “negotiations” are used to explain the loiter in the hijacking scenario. Unfortunately, there is no evidence for either one having occurred.

  461. DennisW says:

    @DrB

    Yes, i agree there is no evidence for anything. However, people do things for a reason. The aircraft failure scenarios are simply not tenable in light of the simulator data. The hypoxia scenarios are likewise improbable given what we can infer from the flight path prior to the FMT. I am simply going with the most probable events.

  462. HB says:

    @DrB, I would be interested to know what possible initiating event you have in mind for a hypoxia scenario that can justify the sequence of events (disregarding the Satcom data for simplification). For me, the biggest hurdle to that scenario is the timing of the loss of transponder communication.

  463. Ge Rijn says:

    SC is now near 33S and seems to be searching outside the designated area again:

    https://twitter.com/search?q=%23mh370

  464. Ge Rijn says:

    n.b. twitter from Kevin Rupp.

  465. Richard Godfrey says:

    @Ge Rijn

    You stated “SC is now near 33S and seems to be searching outside the designated area again:”

    Fake News!

    Ocean Infinity have launched an AUV at -34.5185 94.3860 to cover the gap announced in yesterday’s Malaysian report: MH370 OPERATIONAL SEARCH UPDATE #9.

    The launch was at 26/03/2018 20:07:09 UTC and reached operating depth at -34.6072 94.2763 at 26/03/2018 23:00:58.

  466. Ge Rijn says:

    @Richard Godfrey

    Really fake news? That’s sad if people going to such lengths to produce the graphics in the twitter I linked.

    Hopefully Kevin Rupp replies. I just linked his twitter post here with my comment.

  467. Richard Godfrey says:

    @Ge Rijn

    Your link is generic and does not show a graphic as you claim!

  468. Ge Rijn says:

    @Richard Godfrey

    I checked and you are right. The twitter from Kevin Rupp has disappeared today from Mike Chillit’s twitter.

    Here you have the new (direct) link:

    https://twitter.com/labratsr

  469. Ge Rijn says:

    @Richard Godfrey

    In his previous graphic (16 hours ago) you see SC was close to 33S.
    This was the one I was referring to.

  470. Richard Godfrey says:

    @Ge Rijn

    But 33°S is where SC is supposed to be!

    The deviation to 34.5°S was to fill a gap missed in the previous tour.

    To be clear the Fake News is your statement “SC seems to be searching outside the designated area again:”

  471. Ge Rijn says:

    @Richard Godfrey

    The graphic clearly shows SC and its tracks outside the designated area near 33S. Is this graphic fake news?
    And if not what do you assume SC was doing there? Sightseeing?

  472. Victor Iannello says:

    @Ge Rijn: What is the designated area? Have you plotted the position using the indicated coordinates to determine if Constructor is beyond what is expected?

  473. Peter Norton says:

    > DennisW says:
    > Re: Search strategy
    > I can’t make sense of scanning full width (+/- 25nm) as the search is extended North.
    > Why not scan a selected arc length (say to 26S) along the arc, and sequentially expand the width?
    > If you postulate a bell shaped probability of terminus distance from the arc, that would result in
    > scanning the highest probability area remaining as a function of time.

    re: sequentially expanding the width
    I made the same suggestion back in 2014.
    I guess this would optimize the order (highest probability areas closest to the arc being searched first).

    I guess the reason for not doing so is the ship’s much longer path ?
    Maybe others with insight into the search strategy can explain.

    It would also assume that all segments of the arc are considered equally promising, which clearly never was the case.

    Absent a very high degree of certainty of finding MH370 in a particular segment (which necessarily corresponds to a very low degree of certainty of NOT finding a MH370 in this particular segment, which I think never was the case), I think it would have been more judicious to cover a longer portion of the 7th arc at the expense of more limited width and – if even needed by then – to sequentially expand the width afterwards (not 1:1 however, but more like at a ratio of 1:3 or 1:4 inside/outside, i.e. add 1 mile inside the arc, but 3 miles outside).

  474. DrB says:

    @All,

    I had another thought on the subject of hypoxia. DennisW, you don’t have to read this since you have already closed your mind. If hypoxia occurred due to undetected cabin depressurization, the question is why wouldn’t the alarms alert the flight crew to the problem? Failure of the alarms themselves adds another layer of complexity to the equipment damage required from an accident event. But what if the hypoxia was not due to cabin depressurization, but to a high CO level? In that case, there would be no loss of cabin pressure and therefore no alarms would go off. As far as I know, the B777 does not have CO-level monitors and alarms.

    From Wikipedia: “CO is a colorless and odorless gas which is initially non-irritating. It is produced during incomplete burning of organic matter. Carbon monoxide is a product of combustion of organic matter under conditions of restricted oxygen supply, which prevents complete oxidation to carbon dioxide (CO2). Headache is the most common symptom of acute carbon monoxide poisoning; it is often described as dull, frontal, and continuous. Increasing exposure produces cardiac abnormalities including fast heart rate, low blood pressure, and cardiac arrhythmia; central nervous system symptoms include delirium, hallucinations, dizziness, unsteady gait, confusion, seizures, central nervous system depression, unconsciousness, respiratory arrest, and death.”

    At a concentration of 200 ppm (0.02%): Slight headache within two to three hours; loss of judgment.
    At a concentration of 800 ppm (0.08%): Dizziness, nausea, and convulsions within 45 min; insensible within 2 hours.

    I wonder if it is possible for a small, smoldering electrical or lithium battery fire to produce dangerous CO levels inside the aircraft? If the levels were high enough, CO poisoning could induce confusion, hallucinations, delusion, and eventually incapacity. The symptoms would be difficult to diagnose, because there would be no alarms due to depressurization and because there are apparently no CO detectors in the aircraft. In addition, the symptoms would progress rather uniformly among the passengers and crew, making it even more difficult for the crew to diagnose and rectify.

    So, thinking out loud, a new accident scenario is this:

    1. Small electrical or battery fire starts.
    2. CO created by smoldering fire.
    3. After some time, fire or smoke noticed by flight crew.
    4. Left AC Bus shut off with intention of denying power to fire location (L AIMS Cabinet?; what is power source for window heaters?).
    5. Turn around and return to mainland.
    6. Accumulated CO level now high enough to induce confusion.
    7. No landing.
    8. Irrational route after Penang.
    9. CO exposure produces incapacity of all on board.

    I agree with HB’s point that the Transponders shutting down are a key element to consider. We know it can be done simply in the flight deck with a switch. For an accident scenario, how could they (possibly both?) go dark at the same time? I don’t have a ready answer. First, I need to find out what power sources they used. If the Left Transponder was on the Left AC bus, then it would go dark at the same time as the SDU. However, if the R Transponder is on the R AC Bus, then it should still be operating. I don’t know how the system chooses which Transponder to reply when being interrogated by radar. Would this switch automatically to the remaining Transponder if the one in use lost power?

  475. DrB says:

    @All,

    In order to compare hypothetical scenarios for the events leading to the loss of MH370, I have put together a list of the observational facts, concentrating on the period from 17:07 to 18:40. In particular, I wanted to create a list which is annotated showing which methods of failures or switch settings could lead to the loss of various functions.

    A straw man list is HERE .

    I hope you will suggest corrections, modifications, and additions. Once we are sure we have got the power sources and switchology correct for this aircraft, then we can see how well the hijacking and the accident hypotheses stack up to the observational data.

  476. TBill says:

    @DrB
    Here is a case of a Delta Flight with CO issue…apparently they never figured out what was causing CO poisoning (cargo vs. aircraft operation). No fatalities, just sickness.

    https://www.ajc.com/news/local/delta-sick-passengers-had-elevated-carbon-monoxide-levels/kkRgu2Rc5xpkelu0TKxfgI/

  477. Don Thompson says:

    Dr B,

    The ATC Transponders are supplied by via the Transfer Buses. Isolating the L or R AC Bus does not deny suplly to the corresponding Transfer Bus.

    The CVR is unaffected by isolation of L AC Bus.

    FMS reports were ‘compiled’ at 5 min intervals but only submitted for downlink at 30 min intervals: that is evident during the Log of the MH37q service, and the Log for the accident flight. You appear to be confusing the ACMS ‘engine’ reports for take-off and climb, with the FMS progress reports.

    Please delete that ‘straw man’ reference, it contains too many factual errors to be seriously considered.

  478. Ge Rijn says:

    @VictorI

    I meam the tertiary area. The graphics show SC tracks east of it till near 33.11`S/96E.

  479. Richard Godfrey says:

    @Ge Rijn:

    Here is the plot of Seabed Constructor, please show me where SC is not keeping to the designated area.

    https://www.dropbox.com/s/s8elc9bqb6b7rva/SC%20Track%2028032018.pdf?dl=0

  480. Sabine Lechtenfeld says:

    @Dr.B,
    Zaharie Shah had – clearvoyant that he was – apparently forseen weeks before it actually happened, that such a bizarre accident might eventually happen on one of his flights. Therefore he plotted helpfully a fmt near the Andamans, followed by a fuel-exhaustive path into the SIO with his simulator in order to point future investigators into the right direction. He then realized how extremely unlikely such an event is and wiped the sim data.
    Sorry Dr.B, but Dennis W. isn’t the only one who has closed his mind to these kinds of contrived accident/disaster scenarios. They were highly unlikely before the sim data were confirmed and analysed. Since we know them now, there are IMO only two possibilities: The sim data are legit and Zaharie ran these simulations. In this case everything points at him as the one who abducted the plane with criminal intent. The other possibility is that the sim data were planted in order to frame Zaharie. In this case all scenarios become really complicated and we have to ask ourselves who might’ve tampered with Zaharie’s computer, what were the motives, what exactly needed to be covered up, and are there are other data we can’t trust. In both cases we are looking at a crime. The existence of the sim data eliminate all innocent explanations of the MH370 conundrum.

  481. Richard Godfrey says:

    @DrB “Now that I understand it correctly, Boeing’s design seems quite reasonable to me.”

    I am so glad that the years of Boeing’s experience in aviation, meets with Bobby’s approval.

    Do I now feel more secure flying in a Boeing aircraft?

    Nope, I always felt secure in a Boeing aircraft.

  482. DennisW says:

    @DrB

    From Wikipedia: “CO is a colorless and odorless gas which is initially non-irritating. It is produced during incomplete burning of organic matter. Carbon monoxide is a product of combustion of organic matter under conditions of restricted oxygen supply, which prevents complete oxidation to carbon dioxide (CO2).

    Perhaps the mangosteens caught on fire.

  483. DrB says:

    DennisW,

    LOL. I had the same thought but was unwilling to say it because some people would take it seriously. On the other hand, . . . .

  484. DennisW says:

    @DrB

    I briefly looked at combustion at altitude. That would be another issue or perhaps it would enhance the CO production. I did not find anything definitive with a brief look.

  485. Perfect Storm says:

    DennisW: “I briefly looked at combustion at altitude.”

    Because mangosteens combust differently at altitude?
    =D

  486. DennisW says:

    @Perfect Storm

    I think so. CO results from incomplete burning which seems like it would be the case (more incomplete) in a reduced oxygen environment. I really don’t know. Just speculating.

  487. DrB says:

    @Sabine Lechtenfeld,

    You said: “Sorry Dr.B, but Dennis W. isn’t the only one who has closed his mind to these kinds of contrived accident/disaster scenarios.”

    OK, I’ll add you to that list of people with closed minds.

    Let me ask you a question: Do you think Z would fly the last 5 hours in a Constant Magnetic Track mode instead of using some waypoint? (None of which quite match the recovered file fragments, but that is not the point here.)

    Your answer to this question must be “YES” since you believe the SIO Route is based on simulator waypoints. Now I propose that the test be this: If the aircraft is found near S31.57, then I contend the only way to end up there is to fly the CMT route, since none of the various Antarctic or nearby waypoints I have seen so far predict a 7th Arc crossing near this location.

    If the aircraft is found near S31.57, then it was not flown by waypoint, and therefore it was not pre-planned by Z. If it is found near one of the few (are we down to 1 yet?) previously suggested and as yet unsearched 7th Arc crossings, then I will concede it was likely pre-planned and flown by Z.

  488. Sabine Lechtenfeld says:

    @Dennis, yes I was always suspicious of those darn mangosteens. There were loads of them on the plane. Spontaneous human combustion is another possibility.

  489. DrB says:

    @Don Thompson,

    You said: “Please delete that ‘straw man’ reference, it contains too many factual errors to be seriously considered.”

    I agree that is the wrong moniker to use, but probably for different reasons.

    My intention was to put up a draft list of observational constraints based on my limited knowledge and the published statements of others, one that could be improved over time by corrections and additions by others such as yourself having deeper knowledge and understanding of certain elements.

    Don, your contributions are much appreciated by me. I would simply suggest that if you want to assist in improving the general understanding of what occurred, you can contribute to that knowledge base by either correcting my errors or creating your own list. I only did this to collect the bits and pieces that are floating about but have never, to the best of my knowledge, been concisely expressed in a collected form. If you think that is not a worthwhile endeavor for you, that’s fine.

  490. DrB says:

    @Sabine Lechtenfeld:

    It’s very rare, but it happens

    https://en.wikipedia.org/wiki/Spontaneous_human_combustion

  491. DennisW says:

    @all

    I just read an interesting short essay by Taleb addressing the notion of ergodicity. It is a fun read, especially that part uses the Russian roulette example.

    https://medium.com/incerto/the-logic-of-risk-taking-107bf41029d3

    The paper he references by Peters and Gell-Mann is an essential (but difficult) read. It truly will change economics, insurance, and many other things.

    Sorry for drifting a bit off topic.

  492. David says:

    @Sabine Lechtenfeld. You and DennisW’s are mindset by the simulations. Granted there is a case they are related. However there are at least two serious ‘loose-ends’.
    What was their purpose if originating with Z?
    Further, why would he shift fuel about?
    While they remain unanswered I question their being holy writ.

  493. Victor Iannello says:

    @DrB: Please allow me provide answers to the questions you posed to Sabine. Moderator’s prerogative.

    Do you think Z would fly the last 5 hours in a Constant Magnetic Track mode instead of using some waypoint?

    Waypoints would be more typical for long-distance navigation, but we can’t be sure how MH370 plane was flown.

    (None of which quite match the recovered file fragments, but that is not the point here.)

    We don’t what autopilot mode, if any, was intended to be simulated between points 10N and 45S1.

    Your answer to this question must be “YES” since you believe the SIO Route is based on simulator waypoints.

    My answer is “maybe”. The sim data shows that a flight session was created starting at KLIA, proceeding up the Malacca Strait over the Andamans, then turned to the south, and ended with fuel exhaustion in the SIO. Some of us tried to use the coordinates at each sim point to infer the flight path of MH370. The actual relationship, if it exists, between the coordinates of the sim points and the flight path of MH370 is open to conjecture. I don’t know anybody that claims to know that relationship with any certainty.

    If the aircraft is found near S31.57, then I contend the only way to end up there is to fly the CMT route, since none of the various Antarctic or nearby waypoints I have seen so far predict a 7th Arc crossing near this location.

    Considering that custom waypoints can easily be created using integer latitude and longitude coordinates, or even down the tenth of the minute with a little more effort, even if the debris is found at S31.57, it doesn’t prove that the plane did not fly in LNAV mode. We’ll only know that when we find and recover the flight data recorder and analyze the data. The Antarctica waypoints were only possibilities for flight paths. They in no way define the entire space of all possible waypoints for LNAV routes. That’s why I prefer to refer to a band of possible of endpoints, with “warm spots” associated with certain waypoints. Because of the uncertainty associated with each warm spot, I have advocated for a progressive search up the arc.

    If it is found near one of the few (are we down to 1 yet?) previously suggested and as yet unsearched 7th Arc crossings, then I will concede it was likely pre-planned and flown by Z.

    By my count, we still have three LNAV warm spots to check. Only one has been checked so far. That said, the locus of possible LNAV paths is a continuous band.

    I do in general agree with Sabine that the existence of the sim data means that either:

    1) The captain was somehow involved in the disappearance.
    2) The captain was framed.
    3) It is an extraordinary coincidence that a flight session was created ending with fuel exhaustion in the SIO just weeks before the disappearance of MH370.

    I take it you are in camp (3).

  494. Victor Iannello says:

    David asked: What was their purpose if originating with Z?
    Further, why would he shift fuel about?

    I have some ideas about this that could explain a lot. Pure conjecture, and worthy of a full post, so I won’t go into it now, if ever. (The subject of the simulator seems to bring out the worst in people.) But I contend that we don’t need to know the purpose of creating the simulation to recognize that it was created, and it’s either incriminating, falsified, or just a big coincidence.

  495. Andrew says:

    @DrB

    Regarding the CO theory, the air conditioning system is designed to keep recirculated smoke and fumes out of the flight deck. According to the FCOM:

    “The flight deck receives 100% fresh conditioned air from the left pack. The flight deck is maintained at a slightly higher pressure than the passenger cabin to prevent smoke and objectionable odors from entering the flight deck.”

    The flight deck does not receive recirculated air unless the left pack is inoperative (or shut down).

  496. Mick Gilbert says:

    @DrB

    Thank you for looking to pull together some of the objective facts and likely inferences; I think that’s essential in any analysis. A few comments:

    Re: ‘2. The L & R ATC/Mode S Transponders ceased to function circa 17:21:13.

    We don’t know that both the left and right transponders ceased to function around that time only that the active transponder ceased functioning. While there are two transponders there is no automatic fail-over between them; if the active transponder fails the other one needs to be selected manually by resetting the XPDR Switch on the Transponder Panel on the Aft Aisle Stand Panel.

    I don’t think that we’ve been able to determine what the MAS SOP was so there’s been a bit of discussion as to which transponder would normally be selected as active; it was probably the Left.

    Re: ‘Depowering the Left AC Bus requires 3 steps: (1) Isolate the Left Bus Tie Breaker, (2) switch off the Left Backup Generator, and (3) open the circuit breaker for the Left Integrated Drive Generator.

    The procedure that you have described depowers both the Left AC Bus and the Left Transfer Bus. Based on the loss of the SATCOM we can only reasonably infer that the Left AC Bus was depowered. There are two ways that the Left AC Bus can be depowered; either manually (using the two step process, (1) select the Left Bus Tie Breaker to ISLN, and (2) select the Left Generator to OFF) or automatically (if a significant fault is detected the Left Generator Control Unit can force the Left Generator Circuit Breaker (GCB) to open and the Left Bus Tie Breaker is held open).

    Re: ‘The loss of power to the Left AC Bus was at a different time than the loss of transponders …’

    We don’t know that; it may have been at different times or it may have been concurrent.

    Re: ‘Loss of the Left AC Bus also turns off the TCAS, the IFE, the CVR, the cabin lighting, and the cabin satellite phones.

    As Don has mentioned you’re a bit off piste on some of those components. Some IFE components are powered off the 115V AC R SEC 1 Bus while others are powered by the 115V AC L SEC 1 Bus. Similarly, the IFE in some seat rows is powered by the 115V AC R AC SEC 2 Bus and others by the 115V AC L SEC 2 Bus and I think that the same applies to the cabin satphones. Depowering the Left AC Bus has no impact on those components powered off the Right AC Bus. That said, the loss of some IFE components would probably render the whole IFE inoperative and the fact that the SATCOM (SDU) would have been without power essentially renders the cabin satphones useless as well.

    The CVR (somewhat surprisingly) has only has one power source, 115V AC from the Left Transfer Bus; depowering the Left AC Bus doesn’t affect it.

    Most of the cabin lighting is powered by the 115V AC Ground Service bus, which is powered by the Right AC bus in flight; depowering the Left AC Bus has at best a very limited impact on cabin lighting.

  497. Sabine Lechtenfeld says:

    @Dr.B,
    I don’t think I will remain the last person with a closed mind you can add to your list 🙂
    But let me say this: as far as the sim data are concerned it’s irrelevant if MH370 was flown into the SIO by way points or a CMT. The route of the sim data isn’t the route which MH370 has actually taken. It’s quite obvious that the simulated route hasn’t been executed one on one. For example the two end points of the simulated route are beyond the 7th arc and were not within MH370’s fuel range. This discrepancy between the simulated route and the actual path of MH370 needs to be explained. Maybe, ZS orininally targeted another flight. A flight to Jeddah is a distinctive possibility. There are others. However, the true significance of the sim data is, that Zaharie had apparently contemplated the highly unusual move of a fmt near the Andamans followed by a fuel-exhaustive flight into the SIO, weeks before something very similar actually happened with MH370. It’s beyond me how you can shrug this off as another meaningless coincidence in a row of other mind boggling coincidences which you need to employ in order to make your theory work!
    As to your suggested crash location and the question if the plane followed a CMT: we can dicuss it’s significance when the plane will be found near your predicted location. To discuss this now is a tad premature. But I wholeheartedly wish you good luck 🙂

  498. Victor Iannello says:

    @DrB said: Depowering the Left AC Bus requires 3 steps: (1) Isolate the Left Bus Tie Breaker, (2) switch off the Left Backup Generator, and (3) open the circuit breaker for the Left Integrated Drive Generator.

    At one point, Barry Martin incorrectly assumed a transfer bus could power a main bus, and therefore it was necessary to isolate the backup generators to power down the SATCOM. (That’s one of the few mistakes I’ve ever seen Barry make.) That statement gets repeated a lot, but not on this blog, because several of us are diligent about correcting it.

  499. Andrew says:

    @Mick Gilbert

    RE: “I don’t think that we’ve been able to determine what the MAS SOP was so there’s been a bit of discussion as to which transponder would normally be selected as active; it was probably the Left.”

    MAS policy is to select the left system on outbound flights from KUL and the right system inbound.
    (OPS-A 8.3.0.0.11.1)

  500. Victor Iannello says:

    @Andrew said Regarding the CO theory, the air conditioning system is designed to keep recirculated smoke and fumes out of the flight deck.

    That makes sense. That also means smoke would be more likely detected in the cabin (unless it originates in the cockpit).

  501. Sabine Lechtenfeld says:

    @Victor, thanks for answering Dr.B’s questions far more exactly than I did. I only read your response after I posted mine.

  502. ventus45 says:

    @Richard Godfrey
    You said at March 28, 2018 at 11:24 am

    Ocean Infinity have launched an AUV at -34.5185 94.3860 to cover the gap announced in yesterday’s Malaysian report: MH370 OPERATIONAL SEARCH UPDATE #9.

    The launch was at 26/03/2018 20:07:09 UTC and reached operating depth at -34.6072 94.2763 at 26/03/2018 23:00:58.

    Seabed Costructor then proceeded NE (average 040) in a straight line at an average 13.2 knots for nearly 6 hours until 27/03/2018 05:26:35 covering approximately 90 nautical mikles.

    Do you think Seabed Costructor was simply repositioning, or do you think they are using their MBES at those speeds to do a bathy check on the planned future paths of AUV’s (in this instance specifically, and / or other “high speed transits”). In other words, iis the MBES capable of generating “good data” at 13 knots, or must it be done at slower speeds.

  503. Mick Gilbert says:

    @Andrew
    @DrB

    Ah, there you go – it should have been the left transponder (and presumably the left VHF) that were selected as active. Thank you for that reference.

  504. Sabine Lechtenfeld says:

    @Dr.B, I’m sure you got this – but just in case: suggesting spontaneous human combustion was a joke 😉 ! And it’s actually hotly contested if its possible. Most likely it isn’t. Especially the “spontaneous” part is hotly contested ever since Charles Dickens used the idea for killing off his villain in the otherwise great novel “Bleak House”.

  505. Andrew says:

    @Mick Gilbert
    @DrB

    RE: “… it should have been the left transponder (and presumably the left VHF) that were selected as active.”

    The policy applies to the transponder and radar systems. The left VHF is normally used for ATC comms, because it is powered by the Capt Flt Inst bus and therefore remains available any time the battery is selected on. That’s important during an evacuation, when all other power sources are shut down.

  506. Mick Gilbert says:

    @Andrew

    Thank you for that clarification.

  507. DennisW says:

    @Victor

    I do in general agree with Sabine that the existence of the sim data means that:

    1) The captain was somehow involved in the disappearance.
    2) The captain was framed.
    3) It is an extraordinary coincidence that a flight session was created ending with fuel exhaustion in the SIO just weeks before the disappearance of MH370.

    I take it you are in camp (3).

    Of course DrB is in camp (3) as are several other posters. The simulator data is a huge pointer that DrB and others refuse to address. They simply ignore it. It is truly incomprehensible to me how they don’t at least address it and offer their version of it in their ramblings.

  508. MH says:

    If #1 or #3 why would a flight simulator be required ?
    He would have enough experience to plan out by hand without
    Leaving any digital bread crumbs …

  509. Victor Iannello says:

    @MH: Unless of course breadcrumbs were part of the plan, which I believe is a possibility. As I said, just because we can’t be sure about the reason the simulator data were created, it doesn’t mean we should deny that it would be an extraordinary coincidence if they were not related to the disappearance.

  510. MH says:

    If using the flight simulator to plan then he would need the plan to be printed for hard copy reference.
    Any print spoiler records from the flight simulator app?
    It be more useful to plan on a laptop and do a replay to confirm this plan during the divination.

  511. HB says:

    @DrB, Camp position opinions do not matter at this stage and with all credit to you to consider a scenario based approach against criteria to rule out causes and entire scenarios one by one in deductive manner. It may not lead to a location but at least we could learn more on what has happened.
    Re accidental fire, the threshold of impairment following the generation of smoke in a fire scenario is not based on CO concentration. Impairment and Visibility and difficulty of breathing will come first before CO becomes a problem. Secondly the ppm values you quote are for long term exposure, at short term exposure (actute toxicity level) CO concentration will be required to be several thousands of ppm. With or without taking into account the skoke control sytems in place, there is no possibility that the effects of CO will materialise in term of hypoxia within the first 30m+.
    For the transponder issue, for a technical scenario, considering time to fire escalation, only a fault in the DB feeding the transponder or the transponder itself could explain the timing (though this will require failure of the electrical protection), however, the consequence of fire escalation in that case would be structural degradation or smoke impairment with reduced visibility and breathing difficulty whichever comes first. None can justify the lack of distress signal from the crew, there will be more than sufficient time to raise a distress signal. One could imagine a fire in the eebay impairing many equipment, however, considering escalation time, it will take several minutes for fire to escalate and impair other equipment in cascade if it can escalate at all given the fire load is very low in this area; an emergency descent would have been observed. Even that scenario cannot justify the lack of distress signal.
    If your scenario includes, onset of hypoxia within 30 min of transponder signal off, neither CO exposure mechanism nor O2 depletion mechanism is credible.
    For me, a hypoxic scenario can be technically ruled out with certainty.

  512. airlandseaman says:

    The simulator data is definitely hard to dismiss as a total coincidence, but it could be. In any event, I don’t think it provides any useful information about the actual POI. Even if it represented some sort of early practice run, I seriously doubt the actual MH370 flight would have followed that (or any other) specific simulated path. If it is related to MH370, it is more likely that the sim was a “general concept sim”, not a sim of the exact path flown.

  513. MH says:

    Both the flight simulator data and radar data being so incomplete make them not useful in this investigation.

  514. Mick Gilbert says:

    @airlandseaman

    Well said, Mike. We can argue about who most likely did what till the cows come home and it will not advance actually advising or refining the search strategy one iota.

  515. TBill says:

    Re: Simulator studies
    We have the small “problem” that the McMurdo path works very well with Arc rings. BFO match is not as good unless BFO data is off anyways.

  516. Sabine Lechtenfeld says:

    @ALSM, I like your idea of “general concept sim”. Personally I believe that this is indeed one information we can extract with certainty from the sim data if ZS did create these simulations: he had conceived the idea to turn a plane around at a location northwest of Sumatra and then fly it towards the SIO until the fuel ran out.
    It’s hard to say if we can extract further information, or answer for sure the question why ZS felt a need to create this simulation. We need to discuss this further. But as Victor said so well: the sim data are there and therefore we need to acknowledge them. It’s crazy to dismiss them as a mind boggling coincidence.
    Could the sim data contain a hint at the plane’s crash location? It’s obvious that the simulation is different from the routes which MH370 could’ve actually flown. But it’s possible that ZS wanted to find out if he could reach a certain area in the SIO. I think it’s definitely worth looking at the area where the path towards McMurdo crosses the 7th arc. Although McMurdo is according to MAS not included in the plane’s data base, ZS could’ve entered the coordinates.
    What the sim data might point at, is a crash location north of Broken Ridge. But there are other hints into that direction – even if we wouldn’t have the sim data.

  517. Ge Rijn says:

    @Richard Godfrey

    RE: “Here is the plot of Seabed Constructor, please show me where SC is not keeping to the designated area.”

    I was only referring to Kevin Rupp’s graphics on twitter. On those SC ‘SEEMS’ to have been searching outside the designated area.

    Looking at those graphics more closely it shows Kevin Rupp has not put in the designated search areas on the exact locations and with the right (bended) shape along the 7th arc.
    I think this has caused confusion:

    https://twitter.com/labratsr

  518. Richard Godfrey says:

    SC is making progress on the S.E. side of the Tertiary Search Area.

    https://www.dropbox.com/s/eimwh4go9ih0yt3/SC%20Track%2029032018.pdf?dl=0

  519. Richard Godfrey says:

    @ventus45 asked: “Do you think Seabed Costructor was simply repositioning, or do you think they are using their MBES at those speeds to do a bathy check on the planned future paths of AUV’s”.

    I think SC was repositioning to -34.5185 94.3860 at 13.0 knots to launch an AUV to fill in a gap in the data coverage and then repositioned back to -33.6616 95.3809 at 13.5 knots to launch the next AUV in the tertiary search area.

    SC later returned to -34.3381 94.1714 to collect the AUV at 28/03/2018 15:19:13.

    Bathymetric data is available for this area, so I do not think there was any check for planned future paths of AUV’s.

  520. Ge Rijn says:

    @Richard Godfrey

    Lesson learned: I’ll stick to your SC updates and those of @ Richard Cole.
    Kevin Rupp’s graphics show unreliable and therefore confusing.

  521. David says:

    @HB. Your, “If your scenario includes, onset of hypoxia within 30 min of transponder signal off, neither CO exposure mechanism nor O2 depletion mechanism is credible.”

    Onset of hypoxia could have been before, if insidious, from outflow valves being jammed open too wide or electrical shorting leading to that (which has been known). Confusion onset could be as low as 14,000 ft cabin altitude from what I have read (Helios an example).

    While the cabin altitude warning in Helios confused at least the aircraft captain that should not have been the case in MH370.

    However there might have been no warning before 15,000 ft, depending on landing altitude being set incorrectly.

    But that would require the extremely unlikely coincidence IMO of the failure and that setting.

    @Mick. I agree that any link between any of this and the crash site is distant at least. Hypoxia as a means to an end for the pilot after setting the final course would be in a different category, ruling out any long glide but there is no way of establishing that without the wreckage.

    I gather that hypoxia can be detected in remains (Helios again) though doubt the duration could be.

  522. Michael John says:

    Aslm is right to throw caution into the wind over the Sim Data. If Shah & I stress on the IF part, was party to any dastardly deed then he may have realised that the Sim Data may be used as part of an overall plan to try to narrow down the POI therefore I think there is a chance that the Sim Data could be a Red Herring.

  523. ventus45 says:

    @Richard Godfrey

    Thanks for that explanation.

  524. ventus45 says:

    I find it interesting, almost amusing, that the subject of the SIM has re-entered the mix, yet again, so for the sake of discussion, I will re-propose my “via-Medan” route once again.

    In my opinion, the Sim Route in the Malaysian Police Report, can, at best, only be regarded as an early “mud map” planning exercise. It needed to be refined, to be useful.

    That said, it is interesting to note that if the north west point is eliminated, and replaced from Vampi with something close to what the IG thinks happened, ie, the flight cut much closer around Ache, you get something like this:-

    Supposed Modified Sim Route to S1/S2.

    __WAYPOINTS___ __Leg__ __Total__
    _WMKK to GUNIP 167nm 167nm
    Gunip to Vampi 154nm 321nm
    Vampi to Mekar 68nm 389nm
    Mekar to Sanob 49nm 438nm
    Sanob to Nopek 75nm 513nm
    Nopek to Bedax 83nm 596nm
    Bedax to Uprob 268nm 864nm (close abeam Uprob)
    Uprob to S1/S2 2794nm 3658nm

    Now, when you use the Flight Path allegedy flow by MH370 to Vampi, then onwards, you get this:-

    Presumed Flight Path to S1/S2

    __WAYPOINTS___ __Leg__ __Total__
    _WMKK to Igari 275nm 275nm
    Igari to KP967 224nm 499nm
    KP967 to Vampi 171nm 670nm
    Vampi to Mekar 68nm 738nm
    Mekar to Sanob 49nm 787nm
    Sanob to Nopek 75nm 862nm
    Nopek to Bedax 83nm 945nm
    Bedax to Uprob 268nm 1213nm (close abeam Uprob)
    Uprob to S1/S2 2794nm 4007nm

    Note, that in the Sim Route, Mekar is 389nm from Takeoff.
    In the MH370 flight, Mekar is 738nm from takeoff – an addition 349nm track miles, or looked at another way, an additional 43.625 minutes flying time (at 480 knots), to get to the same point, with much less fuel on MH370 than on the sim planning flight, and with “loss of the element of surprise” to boot.

    In short, going up the Malacca Strait with MH370’s fuel was simply insane, if the intention was to get anywhere near S1/S2 latitudes (regardless of longitude).

    Now, obviously, it did not go to S1/S2, but all the recently suggested paths to northern intersections on the 7th Arc, are in that same general direction.

    On the other hand, going via Medan is the perfect answer for the “reduced fuel of MK370”.

    My Proposed Actual MH370 Route to S1/S2.

    __WAYPOINTS___ __Leg__ __Total__
    _WMKK to Igari 275nm 275nm
    Igari to KP967 224nm 499nm
    KP967 to Gunip 61nm 560nm
    Gunip to MDN 81nm 641nm
    MDN to Merim 112nm 753nm
    Merim to Uprob 174nm 927nm
    Uprob to S1/S2 2794nm 3658nm

    Now the two points I want to make are:-
    (1) FUEL was the critical issue here – he had to choose the shortest route.
    (2) UPROB is the “pivot point” in the planning “mud map”, that enables using MH370.

    The reasoning is obvious.

    Note that “the total distance of the early part of both the modified sim flight, and My Proposed Actual flight to Uprob via Medan, are virtually the same (864nm and 927nm) – only 64nm difference, which is only 8 minutes at cruise speed of 480 knots (8 nautical miles per minute), AND, he disappears DIRECTLY into Indonesian Airspace at GOTLA, where the RMAF could not chase, even if they were up to the job.

    Again, obviously he did not go to S1/S2 from Uprob either. The BTO’s clearly show he had to have gone towards 38S on the arc, as all the initial analysis suggested, but I contend, he went further to the south west and turned south east at 23:40 at top of descent to meet the terminator, not far further south west than the Fugro search reached.

    Not only that, but my route, as previously outlined ages ago, meets the BTO requirements, and I think is an even more plausible fit to the now more relaxed interpretation of the BFO error bars, than when I first proposed it.

    I know DennisW and others will chime in, and say – “no way – the drift analysis clearly proves ………..” , to which I will reply – “yes – proves what – exactly ?”

    I wish OI well in their northern quest, and I would be quite happy to have them find it, and prove me wrong. But I don’t think they will find it up north, because I do not beleive it is there.

  525. Don Thompson says:

    @DrB

    I have emailed suggested revisions to your ‘strawman’ Rev.3

  526. Richard Godfrey says:

    @Ge Rijn

    No worries! I think Richard Cole is the gold standard in tracking the underwater search activities.

  527. DennisW says:

    @ventus@all

    Yes, I will chime in.

    It makes absolutely no sense to throw data or analytics away be it the simulator data or the drift analysis based on the notion that it does not fit your scenario. If you want to ignore data, then provide a solid reason for doing so.

    The simulator data rules out scenarios based on aircraft failure.

    The drift analytics rule out terminal locations beyond 35S.

    Even JW recognizes that you cannot rule things out without a reason. Hence we have not finding the plane in the original search area proves the ISAT is wrong, and the barnacle characteristics prove the debris was planted. I don’t think JW’s reasons are correct, but I respect the fact that he has the courage to put them out there, and recognizes that it is necessary to do so.

  528. HB says:

    @David, i was discussing @DrB scenario. You raised another one. Your scenario include failure of both pilots to recognise the onset and/or not responding to alarm on time but then somehow an active task has been executed to justify the first left turn and this happened in the time frame after the last voice comm which is evidence that both pilots did not realise hypoxia at that point in time.
    Hypoxia affects the cognitive function and memory. So if there is a task followed up, more likely the task will be erroneous or imcomplete but hypoxia is not a known cause for generating random/new tasks when not required (ie such as changing waypoints, turning off equipement, cycling power off and on, etc.). The turn back would be more likely due to realisation of hypoxia after the last voice call. It will also be unlikely to affect both pilots at the same time, there will be a time offset. Even if TOC can be as little as 30 sec, it will take more than that to depressurise the cabin. I will leave it to the hypoxia specialists but based on the timing issue that is evidence that both pilots did not realise hypoxia at the last voice message and other tasks which were initiated (turn back, several turns, etc.), and the failure of the crew to recognise/respond to alarm, I don’t think it is even remotely possible to have occurred here. Adding to that, the SDU reboot will be a major hurdle for this scenario. Interesting scenario to develop though.
    As an interesting point, would this raise an audible alarm that could be heard from the last radio call? and is the first left turn radius evidence that the autopilot was disengaged or achievable with autopilot?

    btw you may be interested in this study mentioning MH370 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5472446/).

  529. Mick Gilbert says:

    @ventus45

    The first problem with your proposed path from the sim data is that there is no evidence at all that the simulation flight ever reached VAMPI leave alone any waypoint west of there.

    @DennisW

    Re: ‘The simulator data rules out scenarios based on aircraft failure.

    That is easily one of the most ridiculous statements you have made and is probably a podium finish for most ridiculous statement to grace these pages.

  530. Ge Rijn says:

    @Andrew

    A question you certainly can answer but I cann’t find a clear answer on the net:
    What format is used to enter long/lat waypoints into the FMC.
    Is this degrees/decimals/decimals (like 35.50.82S) or degrees/minutes/seconds (like 35.30′.58″)?
    And is it required to also enter the last two decimals or seconds?
    Just thinking about something that makes me curious.

  531. DennisW says:

    @Mick

    Where do you stand on:

    1> ISAT data validity?

    2> Debris planting?

  532. Victor Iannello says:

    @Sabine said: Although McMurdo is according to MAS not included in the plane’s data base, ZS could’ve entered the coordinates.

    It appears the plane in the simulation was moved to 45S,104E to simulate fuel exhaustion. The great circle path that aligns with these coordinates and matches the satellite data crosses the 7th arc at 28.3S latitude. It would require drift/retrace error of the FFB of +6.2 Hz, but the BFO error is only 2.3 Hz if this drift occurred. I hope that Constructor is able to get at least this far north along the arc in its progressive search pattern. Assuming the debris is not found before then, it is an important point to check, regardless of people here who believe the simulator data has no value. We can’t be certain that’s true.

  533. Victor Iannello says:

    @Ge Rijn: The format of the coordinates for custom waypoints has been discussed before. Either integer latitude, longitude, such as S45E104, or in the form of deg, minutes, tenth of minutes, such as S4500.0E10400.0.

  534. ventus45 says:

    @DennisW

    I made my position on the flaperon quite clear, long ago, 13th January 2016 in fact.

    To be blunt, I think it (and the Pemba flap) were “planted”, and here is why, and how.

    The essential points, both then and now are:-

    The French established that:-
    (a) the flaperon was part of the initial “ship set” installed at build by Boeing.
    (b) the data plate was missing and it looked to have been geliberately “removed”.
    (c) there was evidence of an apparent attempt at some “repair” AND also an apparent attempt to comply with implementing an AD, that had been, apparently, botched.

    The Malaysians, when they went to France to see it, emerged after only an hour or so – to proclaim that they recognised MAS Maintenance Markings on it.

    I don’t think either maintenance work or work to impliment AD’s to flight control surfaces are done “on wing” Dennis, in fact, I know for an absolute fact, from practical experience with glider inspections, maintenance and repair in the 1970’s, that they most definately are not. They are removed, and they go to “the shop” to be worked on. Ailerons and elevators being critical, balance and rigging wise. Even painting them required “doing it all again”. In the meantime, the aircraft is “AOG”, or is returned to service by fitting a spare. Note that last bit.

    But to move on from the obvious.

    Notwithstanding all that, there are a multitude of unresolved issues regarding:-
    (a) trailing edge damage,
    (b) leading edge is undamaged,
    (c) failure modes of the mounts and PCU attachments are unresolved,
    (d) and of course, there is the case of the barnacles.

    But setting all that aside for the moment.

    Now consider the Pemba Flap segment.

    Malaysia, after having made a big song and dance about the French refusing to give them the flaperon mind, were apparently not interested in the flap initially – strange don’t you think ? Why ?

    Enter the ATSB. The ATSB eventually gets the flap sent to Australia, themselves.

    Again, no data plate at all, trailing egde damage, no leading edge damage, flap track attachment mount snapped through the middle and a steel rod broken in the middle, but no reported evidence of any damage let alone stress cracking at the point of mounting to the composit structure. The forces required to break those metal parts are huge, yet the composite structure that reacted those loads is undamaged ? Not credible. Then of course, that flap mounting was removed, and has never been seen again. Why ?

    Anyway, the ATSB then decides (after consulting the French) that the damage to both the flap and flaperon is consistent with them both being in the up position at the time of damage, which occurred simultaneously, ie, they damaged each other.

    Now, there is no way that I can see that any of that is consistent with either a high speed dive flutter separation, a high speed dive water entry, or a ditching.

    So what does that leave us ?

    There is a very simple explanation.

    In 2012, 9M-MRO was involved in a ground acident – a taxi accident.
    It stuffed it’s right wing tip through the lower tail and rudder of an A340 in China.
    The A340 was stationary. 9M-MRO was the aircraft in motion, ie, with the “energy”.
    We know the aircraft was repaired and eventually returned to service, but how long it took, who did it, and what components were replaced, I can not find out, and I have looked. Perhaps others can put it on their “to-do-lists”, someone might find some useful information.

    But in the absence of hard evidence, a little though exercise says that a heavy B777 taxing at 10 knots or so, clouting an A340, puts a hell of a lot of energy through the right wing, with a very long leverage arm, which flexes the wing spanwise, both laterally (to the rear) and in torsion about the wing box elastic axis. Trailing edge bits, hanging off two mount points each, get compressed together laterally. Well low and behold, they are in the up position too.

    Now it gets speculative.

    Being damaged, they are removed, and are replaced with spares, as well as the outer 5 feet or so apparently, of the right wing tip. All good, signed off, and the aircraft retuns to service.

    So, what happens to the removed components ?
    Chuck them in the nearest dumster ?
    No. Too valuable, and damage is apparently minimal.
    Obviously, they are shipped back to MAS for repair.

    Some time later, repair is attempted, and ……….. botched ?

    As a result, the item(s) are then condemned, so the data plates are removed and destroyed as part of the standard protocols in place, (safety remember – fake unauthorised parts – unorthorised recycling of damaged and condemned parts – etc – a BIG issue in the aviation industry) and the big bits go to the disposal guys. All this takes time, sometimes people are slack, and it takes a while. It didn’t get done.

    They languish under a tarp somewhere, out the back of a hangar, or more likely, at some spare parts storage facility. Every major airline has them, and they are most often not on the airport. Airport space is limited and expensive. You keep your big spares and repairables in an industrial warehouse area, usually not too far away.

    Come 2014, 9M-MRO goes missing.

    By 2015 people were openly asking – “where is the debris”.

    Now it gets really speculative.

    Someone needed an answer, for, “where is the debris”.
    Suddenly, those two major components, have a “resurrection”, of sorts.

    As I see it, desperate attemts to prove that the first and most major component (the flaperon) can hit a specific target (a small island), in the middle of a huge ocean, at pretty much precisely the right time frame, to supposedly prove, that it came from a defined range on the northen part of the arc, on 8th March 2014, leaves me cold. The fact that it was found on that very small island of Reunion, is way too “cute”, way beyond credible. Thus, I do not consider the drift studies to be worth the paper they are printed on. Sorry to be so brutally blunt about it – but there it is.

    And then, there is the Pemba flap, the flaperon’s partner, allegedly “found” by fishermen, in a hidden little bay, under a cliff, only visible, let alone accessible, from and by, a small boat ?
    I wonder who tipped them off to go look in that little bay.

  535. DennisW says:

    @Ventus

    Wow. You have obviously put a great deal of thought (and effort) into those two pieces of debris. I am very impressed, and I have no rebuttal to provide.

  536. Victor Iannello says:

    @ventus45: Besides throwing out the drift analyses and the BFO data at the log-on at 18:25, in your scenario, if I understand things correctly, the Malaysians are co-conspirators in planting the flaperon. And why would they do this?

  537. Rob says:

    @VictorI

    “It appears the plane in the simulation was moved to 45S,104E to simulate fuel exhaustion. The great circle path that aligns with these coordinates”

    Sorry Victor, but I can’t let what you’ve just said go unchallenged. S45 E104 is obviously way beyond the 7th arc, as everyone can see. Shah happened to be using a B777LR to simulate his flight into the SIO. The aircraft he flew on 7th March 2014 was an ER, as everyone knows. How can you seriously suggest that there’s even the remotest chance he would be dumb enough to use the same waypoint coordinates on the actual flight, even if he was confident about being able to completely wipe his hard drive?

    The only useful fact about S45 E104 is that he was planning to keep to whole numbers, for simplicity’s sake.

    The really useful thing about the simulated flight is that it showed he was testing a plan of waiting until he was a measured distance (say 30nm) from a pre-selected waypoint, he used DOTEN in the SIM but he used IGOGU on the night, and then divert to a manual waypoint calculated to be just beyond fuel exhaustion.

    It turns out that S41 E88 was 206nm beyond fuel exhaustion. Suspend your disbelief about the drift data for a while, and allow yourself to think it through rationally for just once.

    It also turns out I’ve discovered that if he kept on the 15nm offset from N571 until he was 30nm from IGOGU then diverted to S41 E88, his FMT would work out as a 15° bank angle, 13.50nm approx radius turn initiated at 18:35.20 (iniated 2nm,or15 seconds late – pilot reaction time) and covered 2788nm between 18:29 and 00:17:30 which is exactly what the fuel load was capable of if he flew at M0.82 at 40,000ft.

    You must not forget the Sun elevation factor. If he had flown to S45 E104, he would have been in daylight for an extended period, and flying toward OZ! Does that sound rational, in the circumstances?

    S41 E88 would have him in Sun 4° above the horizon territory at anticipated MEFE, a synchronisation that would take careful planning and is only possible for this flight for a few weeks per year. Coincidence? Hardly.

    I have already dealt with the final BFO issue – he went into a dive to avoid the high altitude Jetstream, and continued the glide lower down. DSTG were right on with they analysis. The BTO are reliable, they are simply a timing measure, the BFO are not. The BFOs are out to lunch. I’ve already dealt with the debris issue.

    All you have now to cling to are the drift studies. It’s time to ditch them, before it’s too late.

  538. TBill says:

    @David
    On the cabin pressure alarm, note that the before the alarm goes off, the cabin pressure reading is displayed on the screen. So not only would they be ignoring the alarm, but they would be ignoring the display values.

    As far as unforeseen CO issue, I think that is a “stretch” theory, but it would be an important quality-of-life parameter that should be measured in the cabin and reported by ACARS (if in alarm) along with O2 and CO2 and flammable gases in a confined space situation.

    @Victor @DrB
    I note that the McMurdo path shows on my FS9/FSX map as the planned flight path. So even if Z was not flying to McMurdo, if Z had it entered as the original flight plan (which is the way I have it set) then he would know where to move the aircraft by hand to hit that curve.

    One possibility is Z was actually flying 180S CMH from 1090E or DOTEN. With the Fair Skies setting (25 knot wind), that path goes almost right thru DrB’s 31.57S pin which is close to Z’s BDAY 31.71S 96.1E manual waypoint. Bottom line I have much interest up to DrB’s point. That sim path then intersects almost right at 45S1.

  539. TBill says:

    P.S.-
    As far as NZPG, waypoint 78S67 is a very close approx waypoint that I use all the time in MH370 work.

  540. Victor Iannello says:

    @Rob: I never said that 45S,104E was on the 7th arc. I said the plane would cross the 7th arc at 28.2S latitude.

    The main problem with your approach is you don’t know the difference between assumptions and facts. That’s led you to an endpoint that is not supported by evidence, which you then throw out. As a result, your scenario has little support here.

    Please do not repeat what you already have said numerous times. At this point, people will either agree with you or not. Everybody is fully aware of your scenario.

  541. Victor Iannello says:

    @TBill: The search approach has been to progressively search north along the 7th arc. If OI can get to 27S, I think the majority of scenarios will be covered without needing to choose between them. I hope there is a willingness and ability to get there.

  542. Victor Iannello says:

    @TBill said: As far as unforeseen CO issue, I think that is a “stretch” theory…

    Due to the positive pressure of the flight deck, any poisoning or oxygen depravation caused by gases generated outside of the flight deck does not seem possible, unless those gasses were somehow generated or carried by the air supply to the flight deck.

  543. formula says:

    @Michael John “Aslm is right to throw caution into the wind over the Sim Data. If Shah & I stress on the IF part, was party to any dastardly deed then he may have realised that the Sim Data may be used as part of an overall plan to try to narrow down the POI therefore I think there is a chance that the Sim Data could be a Red Herring. ”

    True enough, perhaps a Red Herring although: –

    1. Is it known what efforts were made to erase the Sim Data that I understand was recovered only or largely through reliance upon the FBI’s forensic skills? Clearly, if there was a good expectation that the erasing techniques employed were final, then a Red Herring would not arise – absent a double bluff!

    2. Without an awareness on the part of Captain Shah of the Inmarsat data retention he would very likely suppose that the SIO destination was and would remain undetected and undetectable (save for debris find tracking*) and therefore the purpose of some Red Herring pointing to that area would not seem opportune.

    * Debris finds were likely anticipated by the perpetrator (given the apparent extensive planning) which might help explain selection of an end point as far east as was consistent with debris tracking westward (to then maximize the passage of time before landfall in the hope the incident would have by then lost attention and any search would have ended meanwhile). Might it be plausible to suppose that had a SIO endpoint not have been considered, then the washed-up debris may well not have been connected with MH370 at all as no-one knowledgeable would have issued alerts and no searchers would have been on the look-out in the probable areas?

    3. There is, clearly, an apparent absence of any other indicators or clues from the perpetrator as to either motive or end point and all appearances suggest a desire to create a mystery. The Sim Data would therefore seem to stand alone (over the past four years) if it is indeed a clue woven with a Red Herring in mind. That of course may be the case but would it have been given any credence at all (amidst all the other erased data on the Captain’s computer) had the Inmarsat data not first unexpectedly shown the way?

  544. TBill says:

    @Victor
    “>I hope there is a willingness and ability to get there (up to 27S).”

    Yes that would be greatly appreciated.

  545. airlandseaman says:

    Victor: Re: “Assuming the debris is not found before then, it is an important point to check, regardless of people here who believe the simulator data has no value. We can’t be certain that’s true.”

    I want to make it clear, I believe the fact that we have this sim data is valuable information that must be considered, regardless of the specific points found in the data. But, as a pilot, I don’t think Z would have picked a precise but arbitrary point in the SIO from a simulation, and then carried out the deed using the same point. Had that point been, say, a WP or an airport, then yes, I would put some weight on that specific path. But an arbitrary even number poit in the SIO strikes me as a concept test, not a plan.

  546. Ge Rijn says:

    @Ventus45

    That’s really a thoughtfull story you put up there. The big problem with it is you have no evidence at all and there is no evidence at all still for all your arguments.

    The barnacles cannot serve as evidence of planting in any way.
    Those barnacles attach, grow and die along the route. Get eaten off (by fish, turtles etc.), or die by subjected to the air too long and replaced by a new generation of barnacles possibly a different species depending on latitude. Once a piece has beached all barnacles disappear within a few weeks/months because they are eaten away or rot away which is essentially the same (RR-piece).
    The Pemba-piece and many other pieces show signs of many previously attached barnacles by the circular cement rings they leave.
    This proves to me those pieces have been in the ocean for a long time before they beached.

    The damage-pattern on the flaperon and the outboard flap section are not indicative of planting at all either.
    Those can well be explained by a ditch-like impact or a in-flight seperation (much less likely imo).
    The Hudson-ditch shows both first outboard flap sections broke off during the ditch.
    So this proves to me those hinges can snap and the carbon-fibre skin and structure can break, even during a perfect ditch at low speed.

    Then there are still about 26 other pieces left to be explained with your assumptions. Think twice. It’s just impossible to conduct a planting operation at that scale unseen and unnoticed by anyone.

    All you offer is a thoughtfull conspiracy scenario that has no bases in any known evidence/data. Just like JW clings to. It leads to nowhere and certainly not to finding the plane.

    With the evidence/data at hand we have a chance to find the plane. And that has proven to be difficult enough without all those distracting conspiracy theories.

    We’ll know pretty soon.
    My only concern is the width OI will take at certain latitudes.
    I truly hope +/- 25Nm will be sufficient but I doubt it.
    We’ll see if they skipp this along the trenches of Broken Ridge till ~97E. They might have not a change to return to this area.

  547. Victor Iannello says:

    @airlandseam said: Had that point been, say, a WP or an airport, then yes, I would put some weight on that specific path. But an arbitrary even number poit in the SIO strikes me as a concept test, not a plan.

    First, let me say that I don’t know how MH370 was flown. We are all guessing.

    An airline pilot that wishes to navigate a long distance would likely choose LNAV, which requires waypoints. There are no airports, airways, or defined waypoints in the SIO. And we believe that the Antarctica waypoints were not loaded in the navigation database of 9M-MRO. So, if there was a desire to fly until fuel exhaustion in that part of the SIO and use LNAV, it’s possible that custom waypoints were used. A custom waypoint that corresponds to the coordinates of a simulated fuel exhaustion in the SIO doesn’t seem like a bad guess for us to consider. But it’s still just a guess.

  548. airlandseaman says:

    Victor: Re “…a simulated fuel exhaustion in the SIO doesn’t seem like a bad guess for us to consider….”, I agree, but I doubt that the point used for the sim would necessarily be the same point used on March 8th, if that happened at all.

  549. Ge Rijn says:

    There are two points in the ZS simulator that show fuel exhaust.
    One at high altitude and one shortly after that shows 4000ft altitude.
    4000ft altitude could have been the limit to recover from high altitude.
    Maybe this is what he practised or wanted to be visualised on his simulator (which he deleted).
    ChinaAir 006 recovered from it’s near mach dive at ~5000ft.

  550. HB says:

    @Victor, i would like to add that even in that case, with or without the smoke control system, it will take more than one hour to reach harmful levels of CO. I believe there should also be a design criteria for survivability time due to smoke impairement (tbc). It should be noted that, in an accidental fire, CO will not separate from the smoke which is detectable. This is a totally different issue from CO intoxication in domestic fires where CO remains undetected. The main hurdle is the timing of CO build up not matching the event sequence.

  551. Victor Iannello says:

    @HB: OK, thank you. If true, that’s another strike against theories revolving around incapacitation due to abnormal atmosphere in the flight deck.

  552. Ge Rijn says:

    When the captain said, sounding completly relaxed ‘Goodnight Malaysia MH370’ there where no alarms ringing , no signs of any distress. No signs of intoxication or hypoxia.
    One minute later all communications were switched off one way or the other and the plane turned in a controlled way towards Penang then turning into the Malacca Straight and made a controlled FMT towards the SIO later.
    With specific end-waypoints I’m sure.

  553. Victor Iannello says:

    @Ge Rijn: The plane was moved to position 45S1 at 37,600 ft with no fuel and then the simulation ran for about 28 seconds, reaching position 45S2. During that time interval, the plane climbed (due to the MSFS anomaly I’ve described many times) and using pilot input, returned to the starting altitude of 37,600 ft with wings nearly level, and the simulation was paused. Then, while the simulation was paused, the altitude was manually changed to 4,000 ft. This means, the simulation was not run between 37,600 ft and 4,000 ft. The sim data points do not show a high speed descent and recovery. We can only guess what he was doing.

  554. lkr says:

    @David “Confusion onset could be as low as 14,000 ft cabin altitude from what I have read (Helios an example)”

    That seems awfully low for a healthy adult. I have been to 14+K quite a few times in the mountains, without particular discomfort or confusion. The carloads of tourists to be seen at parking lots atop Mt. Evans and Pikes Peak don’t seem to be in a haze either.

    Individuals vary, and smoking, etc, would lower effective ceiling. On the other hand, aircrew spend hours a day at effective 8K pressurization, equivalent to the low-oxygen tents used by endurance athletes to increase hematocrit. Overall, 16K or 17K is a better representation of altitude where confusion might affect all of the aircrew.

  555. Sabine Lechtenfeld says:

    @ALSM, as I said before, I really like your idea to view the sim data as a “concept test”. But that raises the question what kind of concept might’ve been tested. Obviously a fuel-exhaustive run into the SIO. But that’s not much of a concept because it’s self-evident that a plane with not enough fuel onboard will eventually crash if it is flown long enough into the middle of nowhere. But many of us think that ZS didn’t just want to crash the plane into the SIO; he also wanted to make it as difficult as possible to find and salvage the plane. That might even have been the driving idea behind flying the plane into the SIO in the first place. And in this case I would expect that ZS spared at least a few cursery thoughts on the most suitable areas which were still within his reach. I think he would’ve looked out for an area with a sea floor as deep and rugged as possible and with a low vessel traffic density. If a ships crew would’ve witnessed his crash, all carefully laid out plans would’ve been busted. I think he would’ve tried to minimize this risk. The sim data could’ve been a first test run how he could reach certain suitable areas. The end points of the sim data would’ve indeed been satisfactory. But for example the area immediately north of Broken Ridge on the 7th arc doesn’t seem to be quite optimal. The sea floor is a lot less deep than south of Broken Ridge, and there’s quite a bit of vessel traffic going on between Perth and Cape Town, as ZS could’ve easily checked by looking at one of those world wide vessel traffic density maps . Don Thompson posted a link some time ago. Why would Zaharie settle for this area if he had more suitable areas even further north still within his fuel range?
    This might sound a bit simplistic and strictly utilitarian – but who says that ZS had very complicated criteria for choosing a suitable crash area?

  556. Mick Gilbert says:

    @DennisW

    Apologies for the delay in responding.

    Where do I stand on ISAT data validity? I think that the Inmarsat data is valid. I don’t think that it is has spoofed or doctored. I am not an expert on satellite communications but a number of contributors to this forum are and I rely on their expertise in this regard; nothing that they have said has ever given me cause to doubt the data.

    On debris planting? Well, I’ve tried it a few times and never been able to grow anything, I might need to try a different fertiliser. Seriously though, I think that notion that someone has toured islands in the SIO planting debris is just a nonsense.

  557. DennisW says:

    @Mick

    Thx.

  558. Mick Gilbert says:

    @Victor Iannello

    We’ve had this discussion before but for the sake of a few contributors who think that the sim data points were deleted can we revisit the fact that the blocks of flight sim data recovered from the Shadow Volume reflect overwritten data not deleted data.

  559. Mick Gilbert says:

    @Rob

    Re: ‘The only useful fact about S45 E104 is that he was planning to keep to whole numbers, for simplicity’s sake.

    The problem with that interpretation is that the sim data doesn’t show whole numbers. The values for the first SIO data point (45S1) were:

    [SimVars.0]
    Latitude=S045 05′ 06.8357″
    Longitude=E104 08′ 43.9576″

    And the 45S2 values were:

    [SimVars.0]
    Latitude=S045 07′ 39.5993″
    Longitude=E104 08’ 26.9998″

  560. DrB says:

    @Sabine Lechtenfeld,

    Yes, I got the joke regarding human spontaneous combustion. I thought it was quite funny, so I played along.

    You said: “However, the true significance of the sim data is, that Zaharie had apparently contemplated the highly unusual move of a fmt near the Andamans followed by a fuel-exhaustive flight into the SIO, weeks before something very similar actually happened with MH370. It’s beyond me how you can shrug this off as another meaningless coincidence in a row of other mind boggling coincidences which you need to employ in order to make your theory work”

    I don’t think it is a “meaningless coincidence”, and I never said so. It is another piece of evidence favoring one theory. It might be hugely indicative, or it might not be. We’ll see.

  561. DrB says:

    @VictorI,

    You said: “We don’t what autopilot mode, if any, was intended to be simulated between points 10N and 45S1.”

    Of course we do. The answer to that question is obvious. Knowing only the start and end points, you only have sufficient information to navigate via a great circle, as is done for waypoints, unless you first find a trial and error solution by integrating constant heading or constant track methods. The FMC does not do that, nor does the PC simulator, as far as I know. For instance, neither one will tell you what constant magnetic track is needed to pass through an arbitrary end point, but both will compute the track angle needed along the route to follow a geodesic passing through the end point. In simple terms, LNAV is the only way for Z to reach a specific end point.

    You also said: “The Antarctica waypoints were only possibilities for flight paths. They in no way define the entire space of all possible waypoints for LNAV routes.”

    No one said that they did. The Antarctic landing fields were chosen to create a route passing close to the sim file waypoints. Now you are backing away from even that. Not all LNAV routes are “consistent” with the sim file, just those that pass very close to those recovered waypoints.

  562. DrB says:

    @Andrew,

    Thanks for the comments on the flight deck air conditioning. Any idea on how long it takes to replace the air volume? Can fumes from the MEC enter the flight deck?

  563. DennisW says:

    @all

    An updated Weibull distribution has been computed for the debris finds in the categories of confirmed, almost certain, and highly likely. A dated list of these finds is linked below. A 19th “imaginary” find near today’s date is included to set a reasonable Weibull slope.

    https://photos.app.goo.gl/UF1wpcywZvimeiqs2

    Weibull thinks that ~75% of the debris that will ever be found has been found, some 19 pieces including the “imaginary” piece. So total debris finds are predicted to be ~25 pieces in the categories above. Using Henriks “fractional washed ashore” data results in an original debris population of ~400 pieces.

    https://photos.app.goo.gl/HCFDAMR6jaksBFPJ2

  564. DrB says:

    @HB,

    I agree that a significant difficulty for the CO impairment theory is the high concentrations needed and the relatively short time (~ 70 minutes) available from take-off to reaching Penang (where a landing was not made). That’s why I listed a couple of exposure times. The heart of the matter is whether high CO levels could be generated in the flight deck soon after take-off without accompanying smoke sufficient to set off the smoke alarms. I don’t know if or how that could happen.

    My notion was that something may have occurred that unknowingly impaired the pilots by 17:52, and that precipitating event might have begun much earlier than the diversion at 17:21 (perhaps even at take-off). In this case the diversion would be a response to becoming aware of a different manifestation of the precipitating event, and it was not necessarily influenced by mental incapacity.

    If human remains are recovered from the crash site, presumably they will be tested in Australia for residual evidence of the cause of death.

  565. Sabine Lechtenfeld says:

    @Dr.B, thanks for your answer. I’m glad that you got the joke, and I didn’t mean to be snarky at all 🙂
    I started to research spontaneus human combustion after I read Charles Dickens’ “Bleak House” because I didn’t understand what he was talking about. And then I discovered that there have been indeed a few very puzzling cases of bodies which had been severely reduced through burning while the surroundings were curiously unharmed and there was no obvious point of origin. Arson experts have conducted intricate experiments with pig carcasses in order to clear this up. In the end statistics have been most helpful. Most cases featured overweight unhealthy smokers who had fallen asleep, and the ignition most likely didn’t happen spontaneous but because of an out-of-control cigarette butt, and the body fat fed the slowly smoldering fire without igniting the surroundings. It’s a bit mysterious, though, why the victims didn’t wake up when they started to burn. Maybe, carbonmonoxide took them out early. It’s certainly a very spooky scenario.

  566. DennisW says:

    @Sabine

    Smoking and alcohol use/abuse are very highly correlated.

  567. Victor Iannello says:

    Mick Gilbert said: We’ve had this discussion before but for the sake of a few contributors who think that the sim data points were deleted can we revisit the fact that the blocks of flight sim data recovered from the Shadow Volume reflect overwritten data not deleted data.

    Sorry. The flight files were likely deleted then wiped.

  568. Andrew says:

    @GeRijn

    RE: “What format is used to enter long/lat waypoints into the FMC.
    Is this degrees/decimals/decimals (like 35.50.82S) or degrees/minutes/seconds (like 35.30′.58″)?
    And is it required to also enter the last two decimals or seconds?”

    As Victor noted, lat/long waypoints are entered as degrees/minutes/decimals, to 1/10 of a minute. For example (from the FCOM):

    • N47° W008° is entered as N47W008 and displays as N47W008

    • N47° 15.4′ W008° 3.4′ is entered as N4715.4W00803.4 and displays as N47W008.

  569. HB says:

    @DrB, i understand the reasoning but CO cannot possibly separate from smoke in an accidental fire. It is a different situation that one might occur in a domestic CO intoxication case. One scenario worth exploring further is the one raised by @David about early cabin depresurisation. I think it willl be very unlikely bur more work would be needed to confirm its pissibility.. The last voice call is still evidence that pilots did not realise anything was happening.

  570. Victor Iannello says:

    DrB said: Knowing only the start and end points, you only have sufficient information to navigate via a great circle, as is done for waypoints, unless you first find a trial and error solution by integrating constant heading or constant track methods.

    You are stating facts that don’t address my point. We don’t know what autopilot mode was intended to be simulated. The plane was moved, not flown, for at least part of that leg. It could be he was intending to simulate a constant magnetic heading, for instance. Perhaps not very accurately, but that could have been his intent. We don’t know.

    The Antarctic landing fields were chosen to create a route passing close to the sim file waypoints. Now you are backing away from even that. Not all LNAV routes are “consistent” with the sim file, just those that pass very close to those recovered waypoints.

    No, that’s just not correct. Despite your self-professed expertise in this, I should know why those routes were chosen, because after all, I did the simulations. Richard Godfrey likely feels the same way, but I won’t speak for him.

    There is only ONE Antarctica landing field that was chosen because it matched the simulator data. That was the path to McMurdo Station (NZPG). The others don’t. I have proposed another path that aligns with the final sim points at 45S104E. That’s not even an airport.

    The other paths to airports (South Pole, YWKS) were chosen because they crossed the 7th arc in a feasible range of endpoints based on the satellite data, and the sim data might suggest flying in LNAV mode based on the alignment of two points with McMurdo Station.

    But if you go back and read my post on Great Circle paths, you will see that I explored a range of great circle paths that included paths aligning with airports (South Pole, NZPG, YWKS) and a path aligning with 45S,104E. I have been consistent in recommending progressively scanning the 7th arc and not concentrating on those “warm spots” because it is difficult to assign a high probability to any one of them. That was also my recommendation to Ocean Infinity when queried while they were planning the search.

    Now, if you want to talk about people backing away from predictions made with high certainty, your track record has not been very good. Do you really want to go there?

  571. Andrew says:

    @DrB

    RE: “Any idea on how long it takes to replace the air volume?”

    I don’t know the figures, but it doesn’t take very long for the air in the cockpit to be replaced by ‘fresh’ air. The air flow is quite high to ensure sufficient ventilation and equipment cooling.

    “Can fumes from the MEC enter the flight deck?”

    Theoretically, no. Section 11 (Smoke Penetration Tests) of FAA Advisory Circular 25-9A (Smoke Detection, Penetration, Evacuation Tests and Related Flight Manual Emergency Procedures) states the following:

    “a.(2)…any penetration of smoke into occupied compartments from cargo, storage, or baggage compartments, equipment bays, equipment cooling systems, or other non-continuously occupied areas (e.g., galleys, lavatories, or crew rest areas) during the tests is unacceptable because the toxicity of the smoke is unpredictable and the smoke exposure might continue or increase to a hazardous level before a landing can be made. The smoke concentrations and exposure time in an actual fire or smoke situation might be well beyond those demonstrated during the limited duration of the smoke penetration tests. Genera11y, any smoke penetration during the tests demonstrates that the smoke containment means or control methods are unacceptable.”

    FAA AC 25-9A SMOKE DETECTION, PENETRATION, EVACUATION TESTS AND RELATED FLIGHT MANUAL EMERGENCY PROCEDURES

    Other reading:

    FAA AC 25.795-3 FLIGHT DECK PROTECTION (SMOKE AND FUMES)

    FAA AC 25.795-4 PASSENGER CABIN SMOKE PROTECTION

  572. Mick Gilbert says:

    @Victor Iannello

    Re: ‘The flight files were likely deleted then wiped.

    Victor, that is at odds with the position that you have previously articulated here and here where you made it quite clear that you believed that the the blocks of data in the Shadow Volume reflect overwritten data.

    What’s changed your mind? And what do you mean by ‘wiped‘?

  573. DennisW says:

    @Mick

    All you need to know is that the sim data found on ZS’s hard drive proves that the flight was not the result of an aircraft failure.

  574. Victor Iannello says:

    @Mick Gilbert said: Victor, that is at odds with the position that you have previously articulated here and here where you made it quite clear that you believed that the the blocks of data in the Shadow Volume reflect overwritten data…What’s changed your mind? And what do you mean by ‘wiped‘?

    No, that statement is NOT at odds with the position I have previously articulated, and I have NOT changed my mind. Wiping a file would overwrite the data, but save the overwritten data in the Shadow Volume. Tons of information about this are available if you do an internet search.

  575. Ge Rijn says:

    @DennisW

    Thanks for your up to date Weibull distribution analysis.
    I’m not suprised it predicts my earlier estimates of a few hundred initial floating pieces but your analysis gives scientific weight to a relatively small initial debris field.

    Am I right in concluding your analysis gives ~400 pieces as a maximum based on including the imaginary piece nr.19?
    What is left when you take that imaginary piece out?

  576. Mick Gilbert says:

    @Victor Iannello

    Victor, we previously had this exchange:

    ‘Me> Do you know if these data sets were deleted or just overwritten?

    You> The blocks of data in the Shadow Volume reflect overwritten data.’

    You’ll note that the choice was binary – deleted or overwritten. You appear to have selected overwritten.

    Now you are saying ‘The flight files were likely deleted …’

    What am I missing?

  577. HB says:

    Appologise for spelling issues, having a hard time with the auto functions of my new phone. I meant *possibility*. It was not a lapsus.

  578. Andrew says:

    @DrB

    I should qualify my previous comment where I implied that smoke/fumes cannot enter the flight deck from the MEC. According to the regulations, any smoke that enters the flight deck during certification testing is unacceptable; however, that testing obviously cannot replicate every scenario that might occur in the real world. Nevertheless, the system is designed to prevent smoke entering the flight deck.

  579. Andrew says:

    @all

    The following article might be of interest in light of the recent discussion about smoke & fire scenarios:

    https://airlinesafety.blog/2012/02/02/inflight-fire/

    Note: I am NOT suggesting MH370 encountered a fire.

  580. Mick Gilbert says:

    @DennisW

    For an ostensibly well educated fellow you say some patently stupid things at times, Dennis. All the sim data can do is lend weight to the deliberate malicious action hypothesis. Unless you have stumbled across some form of temporal quantum entanglement linking a sim to the myriad components on an airliner the sim data cannot disprove an alternate hypothesis. You must know that. You could make more factual and less inflammatory statements in that regard and yet you seem to want to make a virtue of spruiking nonsense.

    My alternative response is simply to say ‘Yes Dennis, here’s a cracker.’

  581. Victor Iannello says:

    @Mick Gilbert: Deleting AND wiping would overwrite data. (You conveniently cut off my sentence before “then wiped”, changing the meaning.) We know the data was overwritten. I suspect the files were deleted AND wiped. Just deleting the files without intentionally or unintentionally overwriting would not write the data to the Shadow Volume.

  582. TBill says:

    @Andrew
    I recall the South African Airways Flight 295 fire (Combi style with PAX and cargo in the cabin) had the issue that the smoke did not stay out of the PAX section as it was designed.

  583. DennisW says:

    @Ge Rijn

    Link below to the Weibull plot without the “imaginary” find.

    https://photos.app.goo.gl/dWqDLHb9c5RFhtSZ2

    Weibull predicts that 30% of the debris has been found in this case which translates to a debris field of about 1000 pieces. Take your pick. I was trying to be conservative. I don’t think it is realistic to ignore the fact that nothing has been found in over a year now. Hence my inclusion of an “imaginary” piece.

  584. Mick Gilbert says:

    @Victor Iannello

    That’s about as clear as mud, Victor. Are you saying that ‘deleting AND wiping’ is the same as overwriting?

    Are we agreed on the fact that in order for the data to be written to the Shadow Volume the file must be overwritten?

  585. DennisW says:

    @Mick

    I think we are hung up on what it means to “disprove”. Obviously the sim data does not “disprove” the aircraft failure scenario in the strict sense of what it means to disprove something. I was thinking more in terms of what a “reasonable man” would conclude.

  586. DennisW says:

    @Mick

    The “reasonable man” test is actually a legal term evoked in the Phoenix, AZ area. People were going out and taking mining claims on desirable real estate. The issue was would a reasonable man expect to extract anything of value on such a claim. The claims were purely an attempt to extort money from developers.

    I had several of my mining claims revoked when I lived there.

  587. Mick Gilbert says:

    @DennisW

    Fair enough, Dennis, but I’m just reading what you write using the generally accepted meanings of the words you use. The sim data can not prove the malicious action hypothesis nor can it prove that an alternative hypothesis is not correct.

    I think that a reasonable man would conclude that the sim data lends weight to the deliberate malicious action hypothesis. I have no problem at all with that statement.

  588. Andrew says:

    @TBill

    RE: “I recall the South African Airways Flight 295 fire (Combi style with PAX and cargo in the cabin) had the issue that the smoke did not stay out of the PAX section as it was designed.”

    Yes, there are many examples of smoke or fire where the intensity overwhelmed the system design. That’s why I qualified my previous comments. The smoke/fire protection systems are designed and certified to cope, but only up to a point. Note that in most of the examples quoted in the article, the fire became non-survivable within 20 minutes. If a fire of that intensity broke out on MH370, it is highly unlikely the aircraft would have continued flying for a number of hours after the event.

  589. Mick Gilbert says:

    @DennisW

    Thank you for the ‘reasonable man’ explanation. I’ve got a bit of experience with contract law so I’m familiar with the term and the concept. It’s been around since the 1830s and is probably a pivotal concept for common law. Of course just about everybody thinks that they are a shining example of a ‘reasonable man’ (or woman).

  590. DennisW says:

    @Mick

    🙂

  591. David says:

    @HB. Hypoxia. As has been discussed this has not much to do with finding the wreckage but I continue because right now I can offer nothing about that other than (wait for it) repetition/reinforcement of my view that Z passively awaiting fuel exhaustion is unlikely.

    If you take a look at the Helios example and particularly pages 104 – 108 of:

    http://www.aaiu.ie/sites/default/files/Hellenic%20Republic%20Accident%20Helios%20Airways%20B737-31S%20HCY522%20Grammatiko%20Hellas%202005-085-14.pdf

    you will read examples of the failure of pilots to recognise pressurisation problems. Some of these are attributable to peculiarity of the 737 warning system but there are exceptions as you will see.

    It is not just the 737 though. Notice p108 second last para. NASA reported that over a 10 yr period there were 171 instances of 737 air conditioning and pressurisation problems, 58 critical. In the period, to 2004, the rates for the 727,757 and DC10/MD80 per departure were considerably higher still.

    If hypoxia had not been recognised in MH370 and one pilot had slumped the other, hypoxic, would not necessarily have noticed. Helios almost certainly was such a case and I have read of another like that.

    Yes, I suppose the audio alarm would have been audible in the last radio call, that is if it had not been selected off.

    I think the left turn was much sharper than the autopilot would do.

    Thank you for the reference. There are copious papers on the topic, many related to mountaineering and there are/have been various symposia on it, some periodic.
    I note the sense affected earliest is night vision.

    However I really do not suggest that this scenario should be elevated to ‘likely’ since it remains so conjectural and personally I prefer the scenario of him selecting all AC power off to make quite sure the aircraft would go dark and immediately, flying by hand across the Peninsula at least, even if difficult, before autopilot restoration.

    There is no question though that simulation data need to be taken as seriously, though in my view not to the point of being seen as definitely indicating a pijacking intention by Z. Aside from fuel quantity changes during the simulations they entailed other manual manipulations including, as @Victor mentioned above, fast forwarding from position to position and an inexplicable manual drop of altitude at the end.

    My earlier speculation that Z was testing or examining software still seems possible to me though I certainly agree that the SIO finish would be quite a coincidence if unrelated to MH370. However even the drop of altitude at 45S could be for such as checking magnetic variation/declination corrections do not change with altitude (?!) or just to prove that such an altitude change can be done manually? Just loose speculation of course, better quality speculation being right now in short supply.

  592. David says:

    @TBill. You will see from at least one example from the pages 104 – 108 of the above, it seems commonly there has been little pilot monitoring of cabin altitude, reliance being on check lists, alarms and alerts; much the same as for some other gauges and read outs I imagine.

    I cannot comment on CO toxicity risks except that it is normally associated with piston engine aircraft fumes and unpressurised aircraft I think, and ingress of exhausts on the ground. I have my doubts it is either measured or reported: or for that matter oxygen partial pressure, which is humidity dependent.

  593. David says:

    @ikr. In the Helios case the captain was confused at 14,000 ft, ie the aircraft was at 18,000ft and there was a 4,000 ft cabin altitude lag.

    Subsequently his seat was seen to be vacant by chase F16s. I think he went behind the seat attempting to pull a circuit breaker he had asked the ground engineer about in his confusion; and in fact was irrelevant. It seems quite likely the co-pilot was confused also, if not worse, since he did not intervene.

    Also, please see the pages to which I have referred HB and TBill.

    Finally I refer you to the NTSB report on Cessna 208B, N430A, referred to by the ATSB and Mick Gilbert, wherein the FAA reckons performance starts to drop off at 8,000 ft, “the effect markedly increasing at 12,000 ft. (Halfway down under MEDICAL AND PATHOLOGICAL INFORMATION.)

    https://www.ntsb.gov/_layouts/ntsb.aviation/brief2.aspx?ev_id=20071017X01609&ntsbno=LAX08MA007&akey=1

    As you say individuals vary and maybe cold lapse-rate mountain air you experienced is denser so offers a more oxygen per unit volume/breath than the heated cockpit air. But then again low humidity in the aircraft will increase the proportion of oxygen by decreasing water vapour. Yet another, “on the other hand”. Smiley here.

  594. Nederland says:

    If you hike (or even drive up) a mountain, you are already starting from relatively high altitude and have plenty of time to adjust. If the drop in oxygen is sudden, the effects are far more serious. Still much depends on individual constitution.

  595. Mick Gilbert says:

    @David

    Re: ‘… one pilot had slumped the other, hypoxic, would not necessarily have noticed. Helios almost certainly was such a case and I have read of another like that.

    You are possibly refering to the Kalitta flight KFS66 incident in which the hypoxic but still conscious captain failed to make any reference to his unconscious and convulsing first officer when he reported to ATC,

    Unable to control altitude. Unable to control airspeed. Unable to control heading. Kalitta six six. Other than that, everything A-OK.

    The reason the Captain was experiencing control problems was that his convulsing FO had forced the autopilot to disengage.

  596. Richard Godfrey says:

    It appears that SC has completed the first cycle of launches and collections in the Tertiary Search Area on the S.E. side of the 7th Arc deploying 7 AUVs.

    SC has just started a second cycle of launches in order to complete the Tertiary Search Area possibly by Sunday evening 1st April or early Monday 2nd April.

    Then I expect SC to move to the Diamantina Escarpment at the start of the Broken Ridge area.

    https://www.dropbox.com/s/qqvq1b4vu112wjz/SC%20Track%2030032018.pdf?dl=0

  597. David says:

    @Ventus45.
    FLAPERON:
    – “the data plate was missing and it looked to have been deliberately “removed”.”
    I gather that their being missing is not that uncommon (earlier JW discussion).

    – “there was evidence of an apparent attempt at some “repair” AND also an apparent attempt to comply with implementing an AD, that had been, apparently, botched.”
    Do you know of evidence and details of this?

    – “leading edge is undamaged,” It is in fact crushed at the outboard end, rearwards and upwards.

    FLAP:
    – “The forces required to break those metal parts are huge, yet the composite structure that reacted those loads is undamaged ? Not credible. Then of course, that flap mounting was removed, and has never been seen again. Why ?”
    There was sideways bending I would think, that not being the strong axis and it being comparatively brittle. The Pivot Arm, ductile, broke about its comparatively weak axis, in bending and torsion by the looks. The “flap mounting” remainder was removed presumably to check for any signs of hole elongation and differences between securing points to determine what shear and torsion had been imposed (2nd photo below). It “was never seen again”? It is in the 3rd photo. Has it disappeared since?

    1. Installed: https://www.atsb.gov.au/media/5772933/mh370_inboard-section-of-flap-6.jpg

    2. Holes (blurred) and earthing strap: https://www.atsb.gov.au/media/5772921/mh370_inboard-section-of-flap-24.jpg

    3. “Flap mounting” separated: https://www.atsb.gov.au/media/5772914/mh370_inboard-section-of-flap-20.jpg

    FLAPERON/FLAP COLLISION:
    “Now, there is no way that I can see that any of that is consistent with either a high speed dive flutter separation, a high speed dive water entry, or a ditching.”
    An example. The flaperon departs from its housed position, inboard end separating first, the flaperon rotates around vertical and spanwise axes, its rear outer spar extension then striking the flap inboard end, flexing that upwards and also rearwards with such force that the flap breaks locally at that end and at its “flap mounting” as the flap inboard end rotates to the rear. I have posted a mini paper on this with details of flap internal damage and will again should you like to see it.

    From the tip damage you quote, the horizontal flexing, tip to rear, to more than close the gap between the two would be large. Permanent deformation surely? Also, the impact on the flap inner end from the flaperon was upwards primarily.

    REUNION ISLAND FLAPERON LANDING:
    “The fact that it was found on that very small island of Reunion, is way too “cute”, way beyond credible.”
    Other parts made their way to islands. If planted, the planters anticipated the outcome of the drift studies well, both in geography and timing (‘Roy’ being an exception albeit not on an island).

    PEMBA FLAP FINDING:
    “I wonder who tipped them off to go look in that little bay.”
    Why would they need a tip off if locals? If indeed tipped off why not earlier.

    FLAPERON BARNACLES:
    You did not return to this. JW rightly has questioned how barnacles could grow on the broken trailing edge when the French photo of the flaperon afloat show the trailing edges well above water. Barnacles apparently do not attach to a submerged surface or endure where dry. I have quizzed the barnacle specialist the ATSB used about this without response.
    My supposition is that since the flaperon suffered damage to the honeycomb (or its attaching resin) to its carbon fibre skin and structure, that allowing water ingress with time. After retrieval and before the French floated it the flaperon dried and lightened so floated higher. There are implications if so for the CSIRO testing of the replica, not subject to that damage, as to whether it had too much windage.

  598. David says:

    @Mick Gilbert. “You are possibly referring to the Kalitta flight KFS66 incident..”

    Yes I was. All the same there may be some doubt about that one which apparently is unknown to the FAA. Even so I believe I have come across others and expect I could find at least one if needed.

  599. Mick Gilbert says:

    @David

    There was no NTSB investigation into Kalitta 66 because it didn’t meet the 49 C.F.R. § 830.2 definition of an accident (viz an occurrence associated with the operation of an aircraft in which any person suffers death or serious injury, or in which the aircraft receives substantial damage). I understand that the FAA’s Civil Aerospace Medical Institute in Oklahoma City uses Kalitta 66 as a training case study but I haven’t been able to verify that yet.

    There’s little doubt as to the bona fides of the incident. The National Air Traffic Controllers Association awarded the 2009 Archie League Medal of Safety to the two air traffic controllers involved in that particular incident for their diligence in recognising the symptoms of hypoxia and managing to convince the PIC to take appropriate remedial actions.

  600. Ge Rijn says:

    @David

    The remainder of the pivot arm in your first photo is particularly telling imo. Especially the surface break failure pattern. When enlarged this failure pattern is clearly vissible.

    First (not enlarged) you can clearly see the pivot arm remainder is bend inwards pointing in the longitudal axis towards the inboard side of the flap section.
    This indicates the flap section must have moved almost straight upwards first causing the pivot arm to bend and then break under torsion.

    When enlarged the fracture zone indicates this too, cause the outboard side of the fracture shows the material is stretched under tension before it broke (look at the tappered shape under the thin break line there).
    While the inboard side of the fracture shows two distinctive curls of material indicating this side was under huge compression before breaking.

    So I think the sequence of failure has been:

    -first the outboard flap section moved upwards causing the pivot arm to bend under torsion forces.
    -then the outboard side of the pivot arm got stretched under tension while the inboard side got compressed under the opposite compression forces.
    -then the outboard side failed first almost immediately follewed by the shearing of the inboard side.

    I add just an illustrative video on tension failure where you can see the stretching before the breaking:

    https://www.youtube.com/watch?v=67fSwIjYJ-E

  601. HB says:

    @David, i agree, it is technically very unlikely. The main hurdle is that the timing does not match. The last voice comm is a clear evidence that neither pilot realised anything then roughly a min later there is a turn (ie either they noticed or they initiated a turn back erroneously). Then if you add erroneous cycling of power breakers or turning off equipment plus the relog on one hour later under hypoxic condition, the probability is one in a billion.
    The highjack scenario is from far the most likely. Time will tell.

  602. Victor Iannello says:

    @Mick Gilbert asked: Are we agreed on the fact that in order for the data to be written to the Shadow Volume the file must be overwritten?

    Yes, I’ve said that now several times. That overwriting might have been from a wiping operation. If that explanation remains as “clear as mud” to you, I advise you read how that is done with products that are commercially available, or with the Windows utility Cipher using the wipe (/w) switch, which would be used after the files are deleted.

  603. Mick Gilbert says:

    @Victor Iannello

    Victor, our previous discussion dealt with the relatively simple options of delete or overwrite. You’ve only recently introduced the notion that the data files might have been ‘deleted AND wiped’ so pardon me for being a tad unclear as to what you were talking about.

    I’m curious as to why you think that the data files were deleted and a commercial file scrubber or a security tool such as Cipher was employed? Do you have any evidence that that operation was used?

    Surely the simplest explanation for how the data files ended up in the shadow volume is the user kept saving over the same flight file as the simulation session progressed or that the session was stopped and then returned to using the Previous Flight saved flight file. They’re both very common practices.

  604. Victor Iannello says:

    @Mick Gilbert: Surely the simplest explanation for how the data files ended up in the shadow volume is the user kept saving over the same flight file as the simulation session progressed or that the session was stopped and then returned to using the Previous Flight saved flight file.

    Actually, we know that what occurred is NOT what you described. That’s simply not how Shadow Volumes work. The purpose of the Shadow Volume is to restore the files to the state they were at the time of the creation of the Shadow Volume. In this case, that would be Feb 3, 2014. At the time of the first overwrite on or after Feb 3, 2014, the overwritten data is saved in the Shadow Volume. However, subsequent overwrites would not save additional data in the Shadow Volume. The fact that fragments from multiple flight files are present in the Shadow Volume means that there were separate files that were overwritten, not a single file that was overwritten more than once.

  605. Mick Gilbert says:

    @Victor Iannello

    So, going back to why you now believe that it was likely that the files were deleted and then wiped, do you have any evidence to suggest that the data files were deleted and a commercial file scrubber or a security tool such as Cipher was employed?

  606. Victor Iannello says:

    @Mick Gilbert: There were multiple flight files created on drive MK25, likely on Feb 2 based on timestamp data in the files, and those multiple flight files were overwritten on drive MK25 on or after Feb 3. FS9 was also uninstalled on Feb 20. The MK25 drive was found disconnected from the computer. I look at the evidence and say it is likely the flight files were deliberately deleted and wiped. To me, that’s the simplest explanation. It’s also possible the files were deliberately deleted, the blocks were unallocated in the file system, and then those blocks were overwritten without specific intent in the normal course of drive activity.

  607. DennisW says:

    @Victor/Mick

    The simulator data says a lot of things to me. If ZS was on a suicide mission, why not simply destroy the simulator? The fact that the data was deleted tells me that returning was an expected outcome, and that the simulator path was part of a plan B, not the preferred outcome.

    The flight path on the simulator is consistent with what we know about the actual flight path. There is nothing to suggest any deviation. Why the ATSB did not search the area around 28S on the 7th arc is a mystery to me. It seems like such an obvious place to search. The simulator evidence pointing to 28S is MUCH stronger than analytics performed with the Inmarsat data which “suggest” a path further South. The analysts simply misjudged the accuracy of the BFO data. I regard this as pretty much inexcusable since the expected accuracy was easy to quantify. This failure will evolve into the biggest single reason the aircraft was not found (assuming the OI effort is unsuccessful).

    Also the similarity of the simulator flight path to the actual flight path makes the effort to come up with an aircraft failure scenario look truly silly. Why would any reasonable person not focus on the obvious? I can think of no reason to ignore the simulator data and go chasing made up low probability alternatives.

    If OI abandons the search before looking at the 28S area it will truly be a case of compounded misjudgment.

  608. airlandseaman says:

    Dennis: You make a good argument for paying attention to the fact that the simulator data exists, and may be indicative of planning. But remember, that sim was very different from (the start of) the MH370 flight. It looked more like a diversion from a flight going to Dubai or Europe, not Beijing. Given that, why put weight on S28 specifically? To me, it just adds weight to theory that Z took the plane generally south, ending somewhere on the 7th arc. Victor’s hot spot around S29.7, based in part on debris spotted nearby and CSIRO drift models, seems more likely and generally consistent with the sim data too.

  609. Victor Iannello says:

    @DennisW: I think we might get a fourth, shortened swing. Hopefully that gets us past 28S. Or better yet, the plane is found before then.

  610. DennisW says:

    @ALSM

    Yes, we are on the same page.

  611. airlandseaman says:

    I can confirm Victor’s SC projected SC search plan. It now looks like the search will continue to ~May 21 (after one more trip to Freo). I’m expecting the search width to be reduced to ~40 nm (±20 nm) after BR to allow the search to be extended further NE in the available time. S28 will be a stretch, but possible with good WX and a little luck. The coverage rate should improve a bit in the shallower water above BR.

  612. HB says:

    @DennisW
    1000 does not seem very high. Any confidence level attached to that?

  613. DennisW says:

    @HB

    Very little confidence, HB. Both Poisson and Weibull stop with the last input. There is no way to take them forward in time without inputs i.e. one has to create an imaginary input in the absence of a real one to advance the time. When one creates this input is arbitrary. While Henrik’s stuff is generally very reliable he is using third party data which I have no way of assessing the quality of. Another unknown is the “lifetime” of the floating debris. How much if it get waterlogged and sinks and over what period of time is not taken into account at all.

    I think several hundred pieces of debris is a reasonable lower bound. The upper bound could be in the thousands of pieces, but I would be guessing based on the floating lifetime of the debris.

  614. Richard Godfrey says:

    @airlandseaman

    Victor’s hot spot around 29.7°S is based on the RNZAF debris fields, that are not confirmed from MH370.

    My transoceanic drift analysis, originating most probably from 30°S ±1°, is based on 25 floating debris items either confirmed or likely from MH370.

    Whereas I am encouraged by Victor’s hot spot, in my view the transoceanic drift analysis carries more weight.

    As Victor says, we only want OI to find the plane and I am glad that they are going to run a shortened 4th swing as far north as 28°S.

  615. Don Thompson says:

    Seabed Constructor is now repositioning to launch an AUV for the last full survey block within the Australia agencies’ proposed Primary-Secondary-Tertiary areas.

    Earlier today an AUV was launched running a line down a 1250m seamount and along its flanks before heading into the penultimate P-S-T area.

  616. lkr says:

    @Dennis, HB:

    I think the debris population we are talking about is bounded by the type of debris recovered. Almost entirely composites — from no more than 15% of the plane’s exterior and a few interior panels — and recognizable to a [single] individual who’s developed a search image for such items. If BG had been free to, and inclined to, stroll or ORV the beaches of Madagascar and Mozambique all of the past four years, I have no doubt that the Weybull would fit to a population of such items in the high hundreds at least.

    I think we can imagine what a debris field might look like, a day or two after impact, for varying scenarios. The pieces we’re recovered would not be typical of what was afloat. There certainly would have been seat cushions and luggage, human remains, and perhaps chunks of the wings.. But aside from the composites, most of this sank long before landfall. [Or would have been unrecognizable as related to MH370 — remember BGs effort to associate luggage, sandals, etc..]

    But what can we infer of the number, size distribution, and location on the plane of this initial population of composite fragments? This depends a lot on what happened after impact. Are these fragments sinking before landfall — certainly some do, or we would expect continued landfall [now including searched areas like Australia] from pieces that went into the gyre? That wouldn’t be a random process, as smaller or more damaged pieces would be expected to fall out first. Are they becoming weathered to a point that they aren’t recognizable? I’m sure that’s the case, and especially when experienced searcher isn’t looking.

    I offered this once before, but it’s been a while: We have another starting point for the population of composite debris: From BGs notes, it appears that he recovered one fragment for each 3-5 km of beach searched in ne Madagascar. That would be no more than 2% of the eastern coastline of Madagascar, so a complete BG search would scale to about 500 such pieces.

    Outside BGs work we mostly have the big or “interesting” pieces that might look like something from an airplane to a casual observer. There we see a peripheral population. “Roy” was overlooked [surely many times] on a popular beach. Another piece came to light only because a kid refused to throw his vacation souvenir out. There are similar anecdotes suggesting that there were a myriad [‘a thousand’ in Greek, to be precise] small or dull chunks along the African coast within the first 2 years. Further we had a poster on this board [Paul Smithson, I think] that he found a likely BG-grade piece on a beach stroll in Tanzania, not long after the Pemba piece was recognized. If this scenario is correct, presumably most have by now been buried or shattered by wave action. Or eaten by turtles.

    The point of all this is that, though we may have recovered most of the big pieces of composite [there are only so many flaps and flaperons], we know very little about the size distribution of the initial debris field. But there are many reasons to believe that there were initially a great number of very small fragments.

  617. DennisW says:

    @ikr

    Yes, i agree with your tone and general perceptions. There are many variables including the population demographic and remoteness of most of the shoreline in the region. It is not like the plane debris was washing up in Long Beach or San Diego. I think it is safe to conclude that a large debris field was generated – the plane did not skim along the surface, shed a few parts, and sink largely intact.

  618. Victor Iannello says:

    @DennisW: I’ve many times said I think the three main possibilities for explaining the simulator data are:

    1. The captain was somehow involved in the disappearance.
    2. The simulator data was manipulated to frame the captain.
    3. The simulator data represents an extraordinary coincidence.

    If I had to put probabilities on these three options, I would say:

    1. 75%
    2. 24%
    3. 1%

    I’m surprised more people don’t consider option (2). In fact, there are lots of questions about the simulator data that are not easily answered. Those questions increases the probability of (2) at the expense of (1). However, doubts about the authenticity of the simulator data don’t increase the probability of (3).

  619. DennisW says:

    @Victor

    I agree with your regard for #3. I am less inclined to give as high a rating to #2 than you are. The skills required to pull that off are not trivial. Plus that it would increase the number of people involved and the chance of a leak (whistleblower).

  620. Victor Iannello says:

    @DennisW: Most people would say I pegged (2) too high. It would be hard to disagree, but I still put it significantly higher than (3). I can think of some ways that (2) could be done, and who might do it.

  621. MH says:

    I would switch @VI’s probability items 1 and 2

  622. Peter Norton says:

    > Victor Iannello says:
    > there are lots of questions about the simulator data that are not easily answered.
    > Those questions increases the probability of (2)
    > (The simulator data was manipulated to frame the captain.)

    Susie Crowe summed it up well over there:
    « The facilitation of this [sim] data is weak, flawed by the contradictory early reports of FBI’s analysis finding nothing incriminating, subsequently filtered back through controlled Malaysian channels, silenced for over a year until the only method of fuition was being sneaked out to the public domain. Not exactly a credible process. »

  623. Peter Norton says:

    @ DennisW:
    re: sequentially expanding the search area width

    Maybe wave height during winter is another reason for not adopting this search strategy.

  624. DrB says:

    @VictorI,

    OK, I should have said an “Antarctic landing field” was chosen to create a route passing close to the sim file waypoints, not “Antarctic landing fields”.

  625. DrB says:

    @All,

    A revised version of the list of observational data, with inputs from several contributors to this blog, is HERE .

    There are several significant issues for the accident scenario. First, a Left Transponder stoppage seems to require either loss of Left 115VAC Transfer Bus power or damage to the LRU or antenna transmission line. How could this occur?

    Second, unintentional loss of Left AC Bus would seem to require at least two failures in generators and contactors. In addition, it’s difficult to see how in this case power could be restored to the SDU an hour later. This case seems very unlikely to me.

    Intentional turn-off of the Left AC Bus as a response to an onboard problem could fit the observations. In this scenario the PF shuts down the L AC Bus because there was a problem, such as smoke coming from an item on the L AC Bus. The problem is then isolated using individual circuit breakers, allowing the AC Bus to be powered up again later.

    With the Left Main Bus is isolated, what else needs to happen to then lose power to the Left Transponder?

    Third, it seems very unlikely that all the radios would fail or lose power, since they have multiple power supplies. However, all the radios have audio signals passing through the Audio Management Unit. Physical damage to that unit could prevent voice transmissions from all radios.

  626. Victor Iannello says:

    @Peter Norton: To be clear, I don’t generally agree with her thoughts on this matter.

  627. Victor Iannello says:

    @DrB: For the first item, “No ACARS reports were sent after 17:07,” it is stated that “One method to do this is unchecking the SATCOM box in the Data Communications System ACARS Manager screen presented on the FMC Multi-Function Display.”

    After the SATCOM is re-powered before the 18:25 log-on, there continued to be no ACARS reports. Do we have a plausible reason for why there were no ACARS messages other than the SATCOM link in the ACARS Manager was de-selected? If we don’t have another explanation, then it becomes very likely that the box was indeed unchecked. That would also explain why there was no Flight ID transmitted during the log-on, as this requires (according to Don Thompson) the ACARS system to be operable.

  628. Mick Gilbert says:

    @Victor Iannello
    @DennisW
    @airlandseaman

    Victor, any hypothesis as to how the flight sim data files were transferred to the Shadow Volume needs to account for as much of the evidence we have as possible. As you have noted, there are lots of questions about the simulator data that are not easily answered. Perhaps foremost among those questions in terms of how the data files were transferred to the shadow volume is, why do we only have fragments of the normal .flt files?

    We know that in its normal operation Windows Volume Shadow Copy Service operates at the block level and, in the interest of both task and hardware efficiency, only writes changes to data blocks to the shadow volume. Those changes typically relate to the data blocks being overwritten.

    As you know, the PSS Boeing 777 flight simulator program running on Microsoft Flight Simulator 2004 normally produces .flt files that record something like 1,700 individual parameters divided into some 48 categories. The recovered flight data files contain very significantly less data than a normal .flt file; only something like 185 parameters in 15 or so categories. I think that any hypothesis as to how the shadow volume data was created should be able to explain that anomaly.

    If, as you propose, the data was transferred to the shadow volume by the files being deleted and then wiped, then the wiping process would overwrite every data block associated with the file; the file would be completely overwritten. Would you therefore not expect that the entire .flt file would be transferred largely, if not entirely, intact to the shadow volume?

    My thinking that the transfer process might be related to the relatively common practice of simply re-saving the flight as the simulation session went along under the same file name was largely geared around the fact that that process would generate only partial overwrites; very large portions of the data would remain unchanged and would therefore not be written to the shadow volume. You’ve noted that that is not how the shadow volumes work and I have no reason to doubt that. However, I’d suggest that whatever process generated the shadow volume data files needs to be be closer to one that entails partial overwrites than one that entails complete overwrites.

    Separately, regarding your ‘three main possibilities for explaining the simulator data are:

    1. The captain was somehow involved in the disappearance.
    2. The simulator data was manipulated to frame the captain.
    3. The simulator data represents an extraordinary coincidence.

    This might be a bit pernickety but I’d suggest that you are mixing categories and sub-categories there and that assigning probabilities on a zero sum basis across the three possibilities isn’t necessarily the right way to go. The first step in dealing with simulator data is probably a simple binary :

    1. The simulator data is legitimate, OR
    2. ‎The simulator data has been manipulated.

    Assign probabilities as you see fit.

    IF 1. The simulator data is legitimate THEN you have a second relatively simple binary:

    1.1 The data supports the theory that the Captain was involved in the airplane’s disappearance, OR
    1.2 The simulator data is coincidental.

    Assign probabilities as you see fit.

    Given that a lot of interpretative work has gone into inferring possible flight paths into the SIO from the simulator data might I pose this qualifying binary regarding the penultimate pair of data points:

    1. 10N and 45S1 are connected by a direct route, OR
    2. ‎10N and 45S1 are not connected by a direct route.

    What probabilities would you assign there?

  629. Mick Gilbert says:

    @DrB

    Bobby, have you linked the correct document? That looks like the original.

  630. HB says:

    @Victor, probability at this point is only a function of individual perception. The RMP effectivelly cleared the pilot of likely involvement. #1 and #2 are hence less likely despite perceptions.

    @DennisW, ikr: i am thinking the chisq approach would be a better one for prediction at different confidence levels.
    Yes caution should be used regarding the approximation to a random process. I think it is still relevant and indicative. East Africa coast is not that small. Small debris should have arrived first i agree but not many of these were found after the increased awareness following the reunion discovery. The size of the initial field is not very convincing with the data in hand. I admire BG for his exceptional finding rate and obtaining most of the debris. If the size of the field is big it should warrant a systematic search as a big percentage of the clues will be found there. The search efficiency is a factor also. The official search has been virtually inexistent. I am only aware that the Reunion coast has been systematically searched. This of course raises questions.

  631. DennisW says:

    @Mick

    Given that a lot of interpretative work has gone into inferring possible flight paths into the SIO from the simulator data might I pose this qualifying binary regarding the penultimate pair of data points:

    Actually that is the source of my beef. None of the inferred flight paths into the SIO were based on the simulator data.

  632. TBill says:

    @HB
    “The RMP effectivelly cleared the pilot of likely involvement.”

    Really?
    I’d say the RMP report does not in any way rule out pilot involvement.

    A recent Aussie radio report indicated Malaysia’s draft final report is out for review, and the draft report apparently takes the approach of listing possible causes, including pilot hijacking and cargo management.

  633. DennisW says:

    @HB

    @DennisW, ikr: i am thinking the chisq approach would be a better one for prediction at different confidence levels.

    I have never used chi^2 in a predictive mode. Only in a goodness of fit test mode. Not sure how to use chi^2 in this case, but I make no claims relative to statistical prowess.

  634. MH says:

    Here is further detail on VSS and each needs further investigation to determine what mode it was configured.
    Then this can be used for analyzing what was recovered.

    https://youtu.be/ZCJ4xRjpWDU

  635. HB says:

    @TBill, the emphasis is on ‘likely’ given there was no ‘incriminating’ evidence. Of course it is a possibility, i agree. This was also a preliminary report right? I am not sure about the status of this investigation whether it has progressed or not. There are certainly more work to do such as checking whether other phones registered anywhere, maybe more work on the simulator.
    A basic principle of an investigation is that you could only find what you are looking for and a lot more data is out there waiting to be analysed. The search effort focussed on one clue only and has been strictly minimum to that regards. If one sticks to the competence of the RMP, the RMP did not find incriminating evidence and they seem happy with that #3 case. This of course could be part of a strategy to close the case asap or to cover a #2 case. Victor could be right re #2. A framing / cover up scenario is not excluded. The overall lack of investigative effort and lack of transparency from MYG could turn up to be a clue for #2. Still time for MYG to put things right. A public enquiry on that case would be more than welcomed btw.

  636. Richard Godfrey says:

    @HB
    “The RMP effectivelly cleared the pilot of likely involvement.”

    I find the RMP investigation into Zaharie Shah very superficial, considering he is a main suspect. You appear to agree that there is a lot the RMP could still investigate, in your subsequent post.

    In my view, the weight of circumstantial evidence shows, the most likely cause of the MH370 disaster is a murder/suicide by Zaharie Shah.

  637. Ge Rijn says:

    While waiting today for the end of the search regarding the priority search areas there might be time for some to read the Greek legend of Icarus:

    https://en.wikipedia.org/wiki/Icarus

    Similarities, moral and psychology are interesting I think also regarding MH370.
    Who knows the pilot knew the story.

  638. Richard Godfrey says:

    @Don Thompson has already pointed out yesterday that SC is finishing off the Tertiary Search Area.

    http://mh370.radiantphysics.com/2018/03/18/mh370-search-update-mar-18-2018/#comment-13946

    SC has already moved to the Diamantina Escarpment at the start of the Broken Ridge area.

    Here is an updated graphic as of 30th March 2018 at 07:41 UTC.

    https://www.dropbox.com/s/7dfzkeb65ofzgwf/SC%20Track%2031032018.pdf?dl=0

  639. HB says:

    @Richard, i agree it is superficial. The real question is why a number of analyses of evidence are superficial in a high pofile like this? The RMP is just one of them. It could be the pilot, a third party with or without accomplicies or sponsor from a state. The hypoxia possibilities accidental or not merit some official analyses. All possibilities are on the table, it s beyond my competence to analyse further based on data at hands. It won’t be a first time for any of these. But my gut feeling is that there are some intentions to close the case without knowing more on the possible scenarios and who was involved and i am not comfortable with this. The final report has already been drafted despite all the superficial analyses. It is astonishing. Saying that there will be a discussion on scenarios so maybe we are not up to date with the analyses.

  640. David says:

    @Mick Gilbert. Thanks re Kalitta

    @Ge Rijn. “This indicates the flap section must have moved almost straight upwards first causing the pivot arm to bend and then break under torsion.”
    If you mean it rose at the outboard end, pivoting around the auxiliary support track at the inboard end, that would not bend the pivot arm as far as it exhibits I do not think. I think also that rising at the inboard end, if that is what you mean, would bend the pivot link the other way.

    “While the inboard side of the fracture shows two distinctive curls of material indicating this side was under huge compression before breaking.”
    The curl to me indicates that besides bending there was twist.

    Lets hope the pivot arm and damage to the inside of the inboard flap end have been analysed and the results will be published.

    I go into this more again below for any interested.
    https://www.dropbox.com/s/0iliiyjaneqgqlx/INTERPRETATION%20OF%20MH370%20OUTER%20RIGHT%20FLAP%20INBOARD%20PIVOT%20ARM%20FAILURE.docx?dl=0

  641. Rob says:

    @DrB

    The accident scenario has another problem, which is that the loss of transponder and turn back began just 2 minutes (give or take) after Shah’s routine signoff. A turn back is a major decision. The signoff was routine, which means the crew were unaware of any technical problems at that point. The transponder was lost within 2 minutes and the turn back began almost immediately after. So the crew had 2 minutes or less in which to troubleshoot, assess a life threat situation, discuss and weigh the options and initiate a turn back. Additional obs: No further communications, no premature engine shutdown, no erratic manoeuvring subsequently observed, no loss of speed or altitude, navigation system appeared fully functional.

  642. Victor Iannello says:

    @Mick Gilbert: In regard to why only fragments of the flight files were in the Shadow Volume, I think the more relevant question is why, independent of what fragments may or not have been in the Shadow Volume, were whole files not recovered? Forensic tools should have reconstructed entire files.

    I have a theory. The recovery of the file fragments, based on some statements in the RMP report, seems to have used a key word search. We know that the file fragments that the ATSB were given by the Australian Federal Police had additional information compared to what was included in the RMP report. (The ATSB files included the “top” of the file.) In fact, the ATSB believed they had the entire files until I demonstrated that they did not. (They were missing the “bottom” of the file.) There was confusion because the content of the FSX-PMDG-777 files is different than the content of the FS9-PSS-777 files. It may be that the key word search only recovered file fragments because the wrong flight file template was used for the search. I think it is unlikely that the bottom of the file was somehow not overwritten while the rest was.

    But all of this just changes the assignment of probabilities between (my) options (1) and (2). We need more information to determine that split (which we are unlikely to ever get). In my opinion, option (3) still has a very low probability.

    And I appreciate your offer to renumber my options, but I prefer mine. Your 1a,1b, and 2 correspond closely to my 1,3,2, respectively, but confuses the fact that the extraordinary coincidence of the simulator data and the disappearance is independent of whether or not the files were tampered.

  643. Ge Rijn says:

    @David

    Thanks for your comment. Yes, ofcourse if the flap-section had initialy only moved upwards and this caused the pivot arm to bend and break the bend would be the other way while the carrier is on the underside of the flap section..

    Then -as you say- the pivot arm must have rotated and this caused the twist, bending and eventually snapping of the pivot-link.
    I think it must have been an anti clock-wise rotation while the broken flap section with it’s allready broken carrier was forced backwards.
    I think the curls on the inboard side of the pivot-link remainder indicate this.

    Nice report you put together again.
    By lack of a detailed ATSB-report still, this is not bad at all imo.

  644. Rob says:

    @DennisW
    @Richard Godfrey

    Dennis “Actually that is the source of my beef. None of the inferred flight paths into the SIO were based on the simulator data.”

    The source of your beef! Actually, there is one inferred path that if not actually based on it, then takes it as a model. The one inferred path that you choose to ignore. The path that uses a point 30nm before the next N571 waypoint as a cue to commence FMT then flies a great circle toward a pre-chosen manual waypoint until fuel exhaustion. I won’t elaborate further in deference to Victor, as the path has been eliminated on the grounds that it is unsupported by drift modelling.

    Dennis, you told us all that you were finally comfortable when OI began their search. And you were also as comfortable as anyone could be with S30.

    How comfortable are you now? I would be interested to know.

    Richard “In my view, the weight of circumstantial evidence shows, the most likely cause of the MH370 disaster is a murder/suicide by Zaharie Shah.”

    Richard, welcome (finally) to the club

  645. Victor Iannello says:

    @HB said: The RMP effectivelly cleared the pilot of likely involvement. #1 and #2 are hence less likely despite perceptions.

    On this matter, the RMP report was at best inconclusive. The RMP report has many obvious omissions, including the intense then fractured relationship with Tim Pardi. But also, there is precedent for the RMP to issue a report to whitewash an investigation. For instance, that same RMP said that Prime Minister Najib had no involvement in $700M from the 1MDB fund showing up in his personal bank account.

    Interestingly, when Jho Low’s yacht was recently seized by Indonesian authorities, the RMP was not contacted.

    https://www.usnews.com/news/world/articles/2018-03-08/indonesia-to-hand-over-luxury-yacht-to-us-amid-1mdb-probe

  646. DennisW says:

    @Rob

    Dennis, you told us all that you were finally comfortable when OI began their search. And you were also as comfortable as anyone could be with S30.

    How comfortable are you now? I would be interested to know.

    I still feel good about it.

    When I say no paths were based on the simulator data I am referring to the end points which would imply 7th arc crossings further North than the previous priority search area.

  647. Ge Rijn says:

    I must say, although Kevin Rupp’s positioning of the search areas is confusing if you don’t know, his updates on SC progress are fast:

    https://twitter.com/LabratSR

    SC is continuing searching well into the Diamantina trenches at Broken Ridge.
    Suggesting to me nothing has been found in the final tertiary area also.
    Which would not suprise me as you all know.
    I gave the probability of the whole priority area a near zero in the poll @VictorI held here shortly before the search.
    Still no-one here seems to take the arguments and suggestions I bring up for so long now serious.
    No-one is even willing to conduct a reverse drift-study on the ‘Blue Panel’ and associated debris field.
    I think this is misplaced and could turn out in missed chances.
    Hopefully OI is taking my arguments and suggestions more serious than most here the coming week.

  648. Richard Godfrey says:

    @Rob

    You stated “Richard, welcome (finally) to the club”.

    Little did you know it Rob (obviously) , but I have been saying this for years!

    Not just in Victor’s blog, but also in Duncan’s blog before that.

    I have gained many enemies for this conclusion. Some have tried to discredit me publicly.

  649. Victor Iannello says:

    @Ge Rijn: No-one is even willing to conduct a reverse drift-study on the ‘Blue Panel’ and associated debris field.

    What’s more telling is that YOU are not willing to conduct a drift study of the “blue panel”. Why does anybody here have an obligation to study YOUR theory, especially since that endpoint will be searched?

  650. TBill says:

    @Victor @Ge Rijn
    The search vessel now seems close to Joe Nemo’s observed back scatter point in the bathy data.

  651. Victor Iannello says:

    @TBill: Great! I think that contact was explained as a geological feature, but I know OI will be looking at it carefully in case it is not.

  652. Ge Rijn says:

    @VictorI

    I told you I don’t have the skills and the tools to perform something like that!
    You or @Richard Godfrey or anyone much more accomplished than I am in drift-simulations and computer animation could do this in the shortest of time.
    You are just not interested. That’s the point. So don’t put it on me.
    You reject the ‘Blue Panel’ and debris field and the possibility of a crash point at ~32.26’S till ~97E. You are the scientists here not me.
    Your unwilling views can cost opportunities. So beware.
    Imo the search has just begon.

  653. Victor Iannello says:

    @Ge Rijn: You are being stubborn and foolish, and frankly you are wearing my patience thin. As I have said MANY times, the endpoint associated with the blue panel will be searched. I don’t reject that endpoint. I just don’t see a point in spending time on it considering we’ll know the answer soon enough. That really is not hard to understand.

    You say, You are the scientists here not me. Yet, you express strong, unwavering opinions on EVERYTHING.

    Contributors here have privately asked me to ban you, and I have defended you. When I see you carrying on like you are now, I wonder whether that decision was wise. Think about that before your next comment.

  654. DennisW says:

    @Ge Rijn

    I told you I don’t have the skills and the tools to perform something like that!

    You should have people for that.

  655. Richard Godfrey says:

    @Ge Rijn

    I have spent weeks analysing, what I call the “Flaperon” and what you call the “Blue Panel”.

    I have concluded, it is [ir]relevant to MH370.

    Please do not threaten me, because of my conclusion.

    [Comment edited to reflect subsequent comments by Richard G.]

  656. Ge Rijn says:

    @VictorI

    I know you regard all positions on the 7th arc north of where we are now as possible crash location, but close to the 7th arc.
    My big worry now is OI is not going to search till ~97E along the trenches.
    This is what I try to bring forward all the time.
    I have to do this. Not to be stubborn, but out of all information I gathered in the passed four years. And to have the feeling I did all I can to contribute to a solution.

  657. Victor Iannello says:

    @Ge Rijn: Stop your whining. 97E longitude will be searched. You are close to be being banned because you take up bandwidth without adding anything new. You are not the first here to be quite certain about an endpoint to the exclusion of all others. Last warning.

  658. Ge Rijn says:

    @Richard Godfrey

    Unless you’re not cynical I really appreciate your comment.
    It’s kind of hard to accept it as real after all..

  659. Richard Godfrey says:

    @Ge Rijn

    Apologies, I meant irrelevant!

  660. Kenyon says:

    @DennisW, thank you for taking the time to pull a comprehensive debris list together and for sharing your expanded Weibull analysis.

  661. Ge Rijn says:

    @Richard Godfrey

    Just one word. Sad..

  662. Richard Godfrey says:

    @Ge Rijn

    I was typing irrelevant and the autocorrect changed it without my noticing!

  663. DennisW says:

    @Kenyon

    You are welcome. Like most aspects of this problem there are more degrees of freedom than constraints. Most of the analytics have to be taken as approximations. I am confident that a lower bound of several hundred (400-500) pieces of debris is realistic. That tells me the plane did not glide into the SIO. I feel very good about OI’s prospects if they can search to 28S +/- 20nm.

  664. DrB says:

    @Mick Gilbert,

    Yes, the document I linked is the correct, current one (Rev.4). Google Drive allows one to insert updated versions of a document. Unfortunately, it continues to use the original file name (including any embedded date of the first version).

    @Rob,

    I agree the short time difference between the last radio call and the diversion is highly suspect. However, it is not evidentiary proof.

  665. Victor Iannello says:

    @DrB: In a previous comment, I asked you whether you know of plausible reasons for no ACARS reporting after 18:28. The only reason I know of is the SATCOM link was deselected in the ACARS manager via the MFD, which would also explain the missing Flight ID in the log-on request. Do you (or anybody else) have another explanation?

  666. DrB says:

    @Victor Iannello,

    You said: “. Do we have a plausible reason for why there were no ACARS messages other than the SATCOM link in the ACARS Manager was de-selected?”

    That is a good question. I think the answer is maybe. Possibly there was physical damage to the couplers and/or twisted pairs used in the ARINC 629 Bus connections, but I don’t know if this could prevent ACARS messages from being sent without other discernible symptoms appearing as well.

  667. Victor Iannello says:

    @DrB: You are questioning the integrity of the ARINC 629 bus connected to which unit(s)?

  668. Jerry M says:

    If I could ask a question, there has been a lot of discussion of causal factors like hypoxia, accident, murder/suicide, etc over the years so what I’m wondering could the PIC have committed suicide, taking the entire plane with him soon after disabling the transponder? In other words, could he/she have programmed the autopilot to follow a flight path to the Indian Ocean and then checked himself and the passengers out through depressurization or whatever would kill everyone on board, including him/her? I realize the restart of the SATCOM at around 18:25 is a problem with any early suicide scenario but I keep thinking the following:

    If the PIC lured everyone out the of cockpit, locked the door and turned the plane around, there would be 238 passengers/crew trying to break it down. The door might withstand their efforts but I just can’t imagine how heartless it would be for a pilot to listen to the anguish on the other side for hour(s) and not react. However, if instead he/she committed suicide on behalf of everyone soon after disabling the transponder, then the plane would ghost flight the rest of the way for whatever twisted reason this maneuver was taken.

    Hopefully OI will find the plane and some answers will be found.

  669. Don Thompson says:

    @DrB

    Your original draft of Observational Data included a reference to the Left Systems ARINC 629 Data Bus, and contrived a spurious dependency between its operation and other avionics and electrical systems on the aircraft.

    In private correspondence, I have provided information that contradicts the suggestion of such a dependency.

    The aircraft has four instances of the Systems Bus. None are involved with the ADIRS or the AES (SDU).

  670. David says:

    @Ge Rijn. “Nice report you put together again.
    By lack of a detailed ATSB-report still, this is not bad at all imo.”

    Thanks. Now tidied some, a merit of Dropbox.
    With hands on and a lot more information I hope the Malaysians might be more conclusive. Possibly the ATSB saw it as beyond its remit.

  671. Niels says:

    @VictorI, HB
    It was not only RMP who cleared the pilot. On April 2nd 2014 (US) ABC News reported:
    ““They (FBI analysts) have finished with the simulator. There is nothing suspicious whatsoever about what they found,” a senior U.S. official told ABC News. “There’s nothing at all (criminal) about the pilot. Right now there is zero evidence of a criminal act by the flight crew,” the official said.”

    To my surprise the original article is down, however I have no doubt this was published. There are “secondary” reports citing ABC news, there is an ABC tweet, and I have been sent a screenshot from the article showing the same text.

    See for example:
    http://www.slate.com/blogs/the_slatest/2014/04/02/zaharie_ahmad_shah_flight_simulator_abc_news_reports_fbi_finds_nothing_suspicious.html

    https://mobile.twitter.com/abc/status/451600079997001728

  672. DennisW says:

    @Niels

    Thanks very much for your post. It is indeed remarkable. How can this possibly be reconciled with the data? I am completely taken aback by it actually. There is no doubt that a cover up of massive proportions is going on.

  673. Victor Iannello says:

    @Niels: Yes, we are aware of that report. We don’t know who leaked that information and for what reason, whether it was accurate at the time, and whether the FBI’s views were modified over time based on new evidence it was able to extract. These are all questions that should be asked the relevant authorities. You’ve probably seen that facts from anonymous sources have been notoriously wrong. I have no communication channel with the FBI, so I can’t help. But unless the simulator data evidence was fabricated, I’d say it is very incriminating, as I’ve said above.

  674. DrB says:

    @Victor Iannello,

    Another answer to your question as to why ACARS reports did not resume after the SDU was repowered is Item 1.c. Because of the unsuccessful communications attempts from the ground in the interim, a new “session” must be established according to Don Thompson. Perhaps he will tell you how that could be done. I simply don’t know.

    I also defer to Don T regarding the 629 Bus. I can’t follow his explanations, but I don’t doubt their accuracy.

  675. Victor Iannello says:

    @DrB: At one point, we wondered why no ACARS messages were received after 18:28, as deselecting the SATCOM link in the ACARS Manager only disables transmissions. Don explained that after the ACARS session expired, the AES needed to log onto the ACARS server to re-establish the connection. That did not occur at 18:25, and no ACARS messages were received or transmitted despite the log-on to the Inmarsat network.

    The question is why the AES did not log onto the ACARS server after 18:25. The only plausible explanation I know of is the deselection of the SATCOM link in the ACARS Manager. If true, that would be another serious blow to any scenario involving a mechanical failure.

  676. DrB says:

    @Victor Iannello,
    @Don Thompson,

    Don, would checking the SATCOM box be sufficient to establish a “session” so that ACARS reports would be transmitted after 18:25? If not, what else is required?

  677. Mick Gilbert says:

    @Victor Iannello
    @Don Thompson

    Are you saying that the Flight ID data for an AES log-on is extracted from ACARS?

  678. Niels says:

    @VictorI
    It is not clear to me what was the exact source for the sim data you have used in your analysis. If it was not FBI who apparently recovered the data (call this “first hand”), who was it? Was it second, third, fourth hand?
    I agree with you there is a substantial probability this data is not legitimate (or complete). Doesn’t that mean one should not incriminate the Captain?
    Honestly speaking I experience some of the stronger accusations here on the blog based on such thin “evidence” as quite disturbing.

  679. Victor Iannello says:

    @Niels said: Doesn’t that mean one should not incriminate the Captain? Honestly speaking I experience some of the stronger accusations here on the blog based on such thin “evidence” as quite disturbing.

    Despite any reservations I may have about the authenticity of the simulator data, in my opinion, for many reasons, the captain remains the prime suspect. Is there enough evidence to convict him in a court of law? Not close. But this blog is not a court of law. My whole goal in creating this blog was to generate discussion to solve the mystery.

    Your opinion is yet another demonstration of why it is difficult to discuss criminal matters in an open forum. Do we not suspect somebody until we have enough evidence to convict in a court of law? As far as this blog is concerned, the answer is no [i.e., we can suspect somebody on less evidence]. I think our only obligation is to be honest and sincere in our approach.

  680. TBill says:

    @David
    That is interesting…I was trying to remember who first said the sim data was not suspicious in any way.

    So who first said, maybe there is something there?

  681. Victor Iannello says:

    @Mick Gilbert asked: Are you saying that the Flight ID data for an AES log-on is extracted from ACARS?

    Not exactly.

    The missing Flight ID at 18:25 has bugged some of us for some time. I theorized that when a new route was created, the Flight ID was initialized as blank. But after a trip to the Level D simulator, @Andrew informed us that the Flight ID for a new route was initialized as the Flight ID for the old route, so that theory was proven to be wrong.

    @Don Thompson offered a different explanation that I’ll only summarize here, and Don can further elaborate if he chooses. After initialization by the SDU, the log-on is under the control of the Data Communications Management Function (DCMF). The only observed additional configuration is the Flight ID. The IFE is not controlled by the state of the ACARS Manager link window. However, when the MFD ACARS Manager-SATCOM check box is cleared, DCMF makes no additional configuration to the Log On. Hence, no Flight ID is set.

  682. David says:

    @TBill. Up a bit @Niels mentioned an article, the original of which had been taken down. I provided a copy.
    You may have missed the exchanges before my post.

  683. TBill says:

    @JerryM
    Until about 18:02 just past Penang, the flight path appears hand-flown. After that, automated waypoints could conceivably have been used.

    For example, insert the following flight path into SkyVector:
    VAMPI MEKAR NILAM 0796E 0894E ISBIX 3594S

    In reality I favor active pilot to best fit the data. I do not think we have any prominent proposed paths assuming automated flight until about 18:40.

    Some proponents of mechanical failure have suggested the pilot, before hypoxia, could have had time to activate a MAS company flight path out to the Andamans on the way to Europe or Middle East (with the rationale of finding an airport e.g.; Banda Aceh for emergency landing). But I do not think anyone has flushed how that could explain the curve around Penang and the later FMT turn South. DrB’s prior path had the pilot losing control about 18:40.

  684. Peter Norton says:

    @DrB:
    I applaud you for coming up with the idea to draft a guideline document, for the efforts you put into it and for the move to invite document collaboration. I’m glad to see your brainchild gaining traction.

    If I may comment on some of your points:

    (1b) ACARS reports also stop if the Left AC bus is depowered, because that disables the SDU used to transmit the ACARS data.
    If satellite connection is not available for ACARS transmission, wouldn’t radio (not powered by the left bus) be used as an alternative?

    (1c) An active ‘session’ must be re-established after 18:09.
    Why 18:09 ?

    (3) The SDU lost power
    Technically this is an assumption, neither a fact nor directly observable, correct?
    Could we therefore instead list the observed facts that lead to that assumption ?

    (3) The SDU lost power, beginning at a time after 17:08.
    17:08? Why not 17:07 as mentioned in (1) and (1e) ?

    (4) Loss of the Left AC Bus also turns off the TCAS, the cockpit door lock
    So depowering of the left lets anyone enter from the cabin ?
    How does that square with the pilot suicide theory ?

    (4c) The only means of communicating from the cabin in this case is by mobile phone.
    Well no, since you just established in (4a) that the radios are not lost.

    (6) No radio or satellite phone call transmissions were received from the aircraft after 00:19:30.
    You probably mean 17:19:30 UTC (=01:19:30 MYT) ?

    (7) Satellite phone calls to 9M-MRO from Malaysia Airlines at 18:40 and 23:15
    ATSB says 18:39:52 and 23:13:58
    MDCA says 18:39:53 and 23:14:01

  685. Peter Norton says:

    (4c) The only means of communicating from the cabin in this case is by mobile phone.
    Well no, since you just established in (4a) that the radios are not lost.

    sorry, you can strike that.
    Somehow I read “cockpit” instead of “cabin” …

  686. William Shea says:

    Suicide ?

    No note?

    No Manifesto?

    No video?

    Hijack? With no claim of responsibility ?

  687. DrB says:

    @Peter Norton,

    Thank you for your questions and suggestions. They are helpful in identifying a few issues that perhaps need additional explanation.

    I will answer most here and now and onelast one in the morning. it’s nearly midnight here.

    (1) I am not aware that ACARS reports can be sent via HF/VHF radio. I also don’t see any reason for the pilot to do so, regardless of scenario. If anyone can provide documentation that this is possible and a standard practice, then please post a link.

    (3) The SDU lost power. We know this happened because the SDU logged onto the Inmarsat network again at 18:25. This happened again at 00:19. As far as I know, the only possible cause for this happening is that the SDU rebooted its software when power was restored after an outage.

    (3) The last ACARS position report was for an epoch of 17:07:00, but the last ACARS data were transmitted at 17:08, rounded to the nearest minute. The last fuel readings were for an epoch of 17:06:43. So, to the nearest minute, ACARS functioned until 17:08, and it failed to transmit the next set of position reports for 17:12-17:37, shortly after 17:37.

    (4). There is a manual deadbolt that may be used to lock the cockpit door regardless of the electronic lock.

    (4c) There are no radios in the cabin, as far as I know, just on the flight deck.The point I was making is that no one in the cabin (outside the flight deck, actually) could have communicated with the outside world by any means other than their mobile phones.

    (6) Yes, there is a typo in the time of the last radio call. It should be 17:19:30.

    (7) The times you listed were the start times of the two phone calls. I used the average of the start and end times, rounded to the nearest minute.

  688. DrB says:

    @TBill,

    I believe the last maneuver occurred circa 19:30, and it was setting a CMT using the MCP while on a 180.0 true track. This could have been done by either pilot or even by a flight attendant. I also don’t read into this act any strong evidence for or against hijacking.

  689. Andrew says:

    @Peter Norton
    @DrB

    RE: “If satellite connection is not available for ACARS transmission, wouldn’t radio (not powered by the left bus) be used as an alternative?”

    ACARS can use VHF, if it is enabled on the MFD ACARS Manager page. Some aircraft (not 9M-MRO) can also use HF. The VHF/SATCOM selections are automatically enabled at power-up or during a manual reset of the data communication system. If enabled, ACARS will automatically use VHF or SATCOM (if VHF is unavailable). If VHF and SATCOM are both deselected, ACARS cannot send downlink messages, but can still receive uplink messages. (FCOM Pt 2, p5.40.51)

    RE: “(4) Loss of the Left AC Bus also turns off the TCAS, the cockpit door lock”

    The cockpit door lock is powered by the L DC bus, which is powered by the L XFR bus via the L TRU. The L XFR bus will be powered by the backup generator system if the L Main AC bus is lost, unless the XFR bus is deliberately isolated or there is some fault that prevents the backup system supplying power. In the event that the L XFR bus is unpowered, the DC bus tie relay will close and the L DC bus will be powered by the R DC bus. In short, the loss of the L Main AC bus should not affect power to the cockpit door lock.

  690. DrB says:

    @Peter Norton,

    Your last question regarding the need to re-establish an active session after 18:09 is answered in Don Thompson’s March 23rd post: http://mh370.radiantphysics.com/2018/03/18/mh370-search-update-mar-18-2018/#comment-13630

  691. Andrew says:

    RE: My post above: “If VHF and SATCOM are both deselected, ACARS cannot send downlink messages, but can still receive uplink messages.”

    …but only if an active session exists between the aircraft and ACARS server, as Don Thompson previously mentioned.

  692. Mick Gilbert says:

    @Victor Iannello

    Thank you for that clarification regarding the Flight ID and ACARS. The missing Flight ID has also bugged me and continues to do so; one more item on a longish list of difficult to explain phenomenon under any scenario. Like you, I thought that it may have related to the active route being either dumped or replaced but, as you’ve pointed out, that’s not likely. I’ve also wondered if it might relate to a single or dual FMC failure. The FCOM makes reference to the possibility of a software reset occurring while in single FMC operation (ie after a single FMC failure). In the case of a dual FMC failure the active route and the performance data gets dumped, LNAV and VNAV are unavailable and the autothrottle may be affected.

    I would think it unlikely that deselecting SATCOM in the ACARS Manager would make the Flight ID unavailable to the AES which leaves you with something else having caused the Flight ID to go missing. Absent an answer to the missing Flight ID, I’ve looked at the issue from the other end, so to speak, and wondered if the absent Flight ID would prevent the AES logging onto the ACARS server to re-establish a connection. I haven’t been able to find any documentation on the matter but basic reasoning suggests that the AES log-on to the ACARS server should require the Flight ID, because the ID is essentially the address used to route messages from the aircraft to the various ground parties and vice versa. My thinking is that it’s the missing Flight ID rather than an unchecked SATCOM box that prevented ACARS from reconnecting.

  693. Niels says:

    @VictorI
    I understand your position about the role of this forum and agree on the difficulty to deal here with what looks most like a case with a criminal component. The main point I try to get across is that we still know very little about who did what and why. This means imo we should be really careful in using terms like “suicide” and “mass murder(er)” in the context of a specific person.

  694. Nederland says:

    @Niels

    Your source above only cites an unnamed “US official” who are not even in charge of any investigation. Much information of this kind has been published by news agencies cashing in on the general interest in the flight at the time.

    Khaild, the Malaysian official in charge has said at the time that the probe into the flight simulator might have turned up suspicious material. However, no one noticed that headline at the time.

    http://www.thesundaily.my/news/1004871

    @Mick Gilbert

    AFAIK, you can either delete the flight ID manually or simply swap to a new flight plan, without assigning a new Flight ID. Also, the FI makes it very clear that there was “no Data-2 ACARS traffic” (p. 54). Of course, the Inmarsat data show the same.

  695. Victor Iannello says:

    @Nederland: AFAIK, you can either delete the flight ID manually or simply swap to a new flight plan, without assigning a new Flight ID.

    If you read my comment, you’ll see that switching the active route does not delete the Flight ID.

  696. Victor Iannello says:

    @Mick Gilberet said: I would think it unlikely that deselecting SATCOM in the ACARS Manager would make the Flight ID unavailable to the AES which leaves you with something else having caused the Flight ID to go missing.

    So what part of what Don describes do you believe is incorrect? Or it just your gut feeling?

    In the case of a dual FMC failure the active route and the performance data gets dumped, LNAV and VNAV are unavailable and the autothrottle may be affected.

    Dual FMC failures? Really?

  697. Niels says:

    @Nederland
    Many thanks for the link. I hadn’t seen the article before and it contains some interesting aspects.

    First of all “the clues”: it is not clear these relate to the simulator analysis or something else.
    Quite remarkable that Khalid speaks: “We are sorry but we cannot reveal any details of the case as it may affect prosecution or a trial if there is any in the future.”

    Concerning the simulator analysis:”It may be cleared on one aspect but we have to look into other areas as well. No, it has not been cleared,” he said in a press conference”
    Indeed this would suggest the ABC news item was premature or incomplete. However, I hope we agree that it not necessarily means something suspicious had been found. It could “just” mean the analysis had not been fully completed on all angles.

  698. Mick Gilbert says:

    @Victor Iannello

    I don’t doubt Don’s explanation, what I’m suggesting is an alternative, specifically that the Flight ID had been dumped and that that prevented the AES log-on to the ACARS server. If you have any information suggesting that an ACARS log-on is possible without a Flight ID I’d be grateful for the reference(s).

    With regards to a dual FMC failure, yes, really. Boeing clearly contemplate the possibility as dual FMC failure is referenced at least half a dozen times in the B777 FCOM. The FCOM details its impact on LNAV, VNAV, autothrottle and NAVAID tuning functions together with the required remedial actions. Separately, there is an example of a B777 dual FMC failure detailed in the NASA Aviation Safety Reporting System database under ACN 1334269.

  699. Peter Norton says:

    @Drb:
    Let me say upfront that I don’t intend to criticize but to only help improve your document, the merits of which I have abundantly underscored in my previous posting. With that being said, here are my remarks:

    (1) I am not aware that ACARS reports can be sent via HF/VHF radio.
    They can. (see here and previous discussions)
    I also don’t see any reason for the pilot to do so, regardless of scenario.
    Regardless, that’s a fact and has implications any MH370 theory has to deal with (which, as I understand it, is the whole point of your PDF) namely the question why ACARS messages weren’t transmitted via radio when the satellite connection was lost. Months ago, Andrew and Don Thompson have explained the configurations which lead to ACARS msgs being sent/not sent via radio. This may be another fact accident theories could have difficulty explaining. And it’s an important point for other reasons, I will come back to this another time.

    (3) The SDU lost power. We know this happened because the SDU logged onto the Inmarsat network again at 18:25. This happened again at 00:19. As far as I know, the only possible cause for this happening is that the SDU rebooted its software when power was restored after an outage.
    Sorry for nitpicking, but this is a theory. The SDU lost power (theory) because the SDU logged on at 18:25 (fact), so we assume that the SDU lost power beforehand (theory). The whole point of your PDF (as I understand it) is to enumerate facts, that all theories have to deal with – and not to introduce theories itself, which would defeat the whole purpose. You adhere to that principle with admirable consistency throughout your whole document, e.g. by saying “the transponder ceased to function” (fact) instead of “the transponder was shut off” (theory). So may I suggest not to deviate from that principle here. I agree (which is completely irrelevant) that the theory that the SDU lost power is the most likely. But a number of other theories exist – ATSB final report 3 Oct 2017, Appendix B, p.3:
    « Apart from a period between […] 17:07 [and] 18:25 […] the SATCOM link was available during the flight. This interruption of the SATCOM link occurred after ACARS had stopped transmitting messages and may have occurred for a number of reasons such as cycling of the electrical power, the aircraft’s antenna losing sight with the satellite or the resetting of the aircraft’s Satellite Display Unit (SDU). There is no record in the satellite Earth Station log of the link having been logged-off from the cockpit through the Control Display Unit (CDU); such an activity would have been automatically captured in the Earth Station log. The reason for the loss of the SATCOM link is currently being investigated by the aircraft and equipment manufacturers. »
    Rather than stating an assumption (“SDU lost power”), I would suggest to list the observed facts that lead to that assumption (i.e. the 18:25 logon and the conclusions from the OXCO discussions) and then mention the SDU power loss as one possible explanation. That’s exactly how you handled it in (1) where you say “no ACARS reports were sent” followed by (1b) “ACARS reports stop if the left AC bus is depowered”. I would suggest to adhere to the same principle by stating that no satellite communication took place between 17:07 and 18:25 and subsequently, as a sub-point, that one possible reason for that is SDU power loss.

    (3) The SDU lost power, beginning at a time after 17:08 and probably before 17:37, until 18:23
    Why “probably” ? Is it possible that SDU power loss occured after 17:37 ?

    (3a) This implies the Left AC Bus was depowered, because the SDU circuit breaker is located in the MEC, not in the flight deck.
    I would also reword this since “imply” may be interpreted as being the only possible conclusion, when other theories are available (accident/fire scenarios, JW theory, etc.).

    (3) the last ACARS data were transmitted at 17:08
    Then it’s not correct to say:
    (1) No ACARS reports were sent after 17:07
    (1e) no ACARS data after 17:07 were ever sent

    In case this is a rounding issue, the solution to this contradiction would be to either use timestamps with seconds-precision (my preference) or else to settle on either 17:07 or 17:08. But to say “no ACARS data was ever sent after 17:07” is not correct when elsewhere the document says “the last ACARS data were transmitted at 17:08”.

    (4). There is a manual deadbolt that may be used to lock the cockpit door regardless of the electronic lock.
    This could be added to your document.

    (4c) There are no radios in the cabin, as far as I know, just on the flight deck.
    I have immediately added right after my posting (see above) that it was a reading error

    (6) Yes, there is a typo in the time of the last radio call. It should be 17:19:30.
    just a reminder for correction

    (7) The times you listed were the start times of the two phone calls. I used the average of the start and end times, rounded to the nearest minute.
    That’s confusing. I would use the start times. All other documents (including the 2 quoted in my previous posting) do this.

  700. Victor Iannello says:

    @Mick Gilbert said: I don’t doubt Don’s explanation

    When you said “I would think it unlikely that deselecting SATCOM in the ACARS Manager would make the Flight ID unavailable to the AES which leaves you with something else having caused the Flight ID to go missing”, that certainly sounds like doubt.

    Boeing clearly contemplate the possibility as dual FMC failure is referenced at least half a dozen times in the B777 FCOM.

    Sure, many things are possible. I try not to propose scenarios requiring multiple events each with low probability.

    If you have any information suggesting that an ACARS log-on is possible without a Flight ID I’d be grateful for the reference(s).

    We have evidence that unchecking the SATCOM link in the ACARS Manager will disable ACARS and prevent the Flight ID log-on request. Even if ACARS log-on is not possible without Flight ID (which I seriously doubt because the ICAO Address seems much more relevant in the ACARS server’s identification of an airframe), your explanation for the missing Flight ID is double failure of the FMCs. Unless we have reason to doubt the simple explanation (the SATCOM link was deselected in the ACARS Manager), I don’t see a reason to consider an unlikely scenario (dual failure of FMCs).

  701. TBill says:

    @Victor
    “@Nederland: AFAIK, you can either delete the flight ID manually or simply swap to a new flight plan, without assigning a new Flight ID.

    @Victor: If you read my comment, you’ll see that switching the active route does not delete the Flight ID.”

    So what about intentionally deleting flight ID? PSS777 I can change it to “-” a dash.

    Jean-Luc suggested TCAS *might* have been used after 18:25…does TCAS transmit flight ID to the other aircraft?

  702. Don Thompson says:

    @Mick Gilbert wroteI would think it unlikely that deselecting SATCOM in the ACARS Manager would make the Flight ID unavailable to the AES which leaves you with something else having caused the Flight ID to go missing.

    My understanding of the 777’s AIMS, SDU, and IFE configuration leads me to conclude the opposite: if SATCOM is deselected prior to the SDU powering up, and the ACARS Manager selection remains unchanged, no communication is possible from the FMCF and FDCF functions to the SDU.

    When power is removed from the SDU, non-volatile configuration data are lost. The Flight ID (or per SDM, Alphanumeric Flight Identifier) is not stored by the SDU in non-volatile memory.

    The Flight ID originates in the FMS, it is represented in the ICAO flight plan compliant ‘MAS370’ form (not the IATA/ARINC ‘MH0370’ form). At power restoration prior to the 18:25 Log On the Flight ID was not communicated to the SDU, nor was any ACARS related communication initiated.

    In order to make the 18:25 Log On: the SDU did receive IRS data from AIMS/DCGF over the ‘IRS’ ARINC 429 Bus; the SDU did receive the aircraft 24bit address over an ARINC 429 bus from AIMS/CMCF (the address originates from a pluggable module on the transponders and is distributed via the CMCF).

    Immediately following the GES Log On, the SDU processed the two ISO-8208 (X.25) Satellite Sub Network connections originated by the IFE system. The ISO-8208 SSN correspondence is routed from IFE, via AIMS and its SDU ARINC 429 handler, to and from the SDU. A common ARINC 429 data bus is employed by AIMS for ACARS and 8208-SSN correspondence with the SDU.

    The 18:25 and 00:19 Log Ons were both absent anything concerned with ACARS communication: no Link Test message (DCMF testing end-to-end link connectivity), no Media Advisory message (DCMF initiating active ACARS session), no Flight ID (communicated from FMCF via DCMF to SDU).

    My analysis of the operations involved, the avionics functions responsible, and the datalink traffic recorded in the full Stratos Log leads me to conclude that unchecking SATCOM in the ACARS Manager screen disables initiation of any ACARS related operation to the SDU.

  703. Peter Norton says:

    @DrB:
    (1c) An active “session” must be re-established after 18:09 between the aircraft and the ACARS server on the ground in order for the ACARS reports to be transmitted when power is restored to the SDU, and this was not done.
    I would rephrase this as follows (leaning on Don Thompson’s clearer wording):

    (1c) At 18:09 the active session between the aircraft and the ACARS server on the ground was lost (due to the failed ground-to-air message from MAS ODC) and must be must be re-initiated in order for ACARS reports to be transmitted when power is restored to the SDU. Yet no session was re-initiated when the SATCOM datalink was re-established at 18:25.

  704. Victor Iannello says:

    @TBill: So what about intentionally deleting flight ID? PSS777 I can change it to “-” a dash.?

    Sure, the Flight ID could have been manually deleted. I don’t see an innocent explanation as to why that would occur.

    Jean-Luc suggested TCAS *might* have been used after 18:25…does TCAS transmit flight ID to the other aircraft?

    Then Jean-Luc will have to explain how TCAS was used while the transponder was in standby. As you know, we have no SSR data.

  705. Mick Gilbert says:

    @Victor Iannello

    Re: ‘… your explanation for the missing Flight ID is double failure of the FMCs.

    No, it is not. Please read what I wrote, Victor. I said, ‘I’ve also wondered if it might relate to a single or dual FMC failure.

    Re: ‘I try not to propose scenarios requiring multiple events each with low probability.

    That is of course your prerogative. As I have pointed out on a number of occasions the history of aviation safety is peppered with accidents and incidents comprised of ‘multiple events each with low probability‘. Moreover, do you really see a murder/suicide over 6-odd hours (no previous event went for more than about 100 minutes) with a turnback and transit back across the Malay Peninsula to within visual range of the country’s third busiest airport (no previous event entailed a diversion from the planned route) and all the other many facets of this flight as being a single event? Whichever way you want to look at this it is a series of multiple events of low probability. And like it or lump it, the simple most probable answer is not always correct.

    Any old how, I guess we’ll just have to wait till they find it because just as sure as God made little green apples arguing about it here isn’t going to settle the matter.

  706. Don Thompson says:

    @Mick Gilbert wrote “Flight ID had been dumped and that that prevented the AES log-on to the ACARS server.

    Incorrect.

    The Alphanumeric Flight Identifier, when informed via a Log On, is not passed through to the ACARS server, it’s a useful piece of information for Inmarsat network management, that’s all.

    The implication of any AES Log On is that a logical channel is created over the SATCOM datalink for ACARS traffic. In turn, the GES establishes a virtual circuit through the ground network to the ACARS server. This behaviour is evident in the Stratos Log, described in the additional logs included in othert worksheets. However, absent any Media Advisory message, initiated by the aircraft, no communication session is established.

  707. Mick Gilbert says:

    @Don Thompson

    Thank you for the more detailed explanation, Don.

  708. Victor Iannello says:

    @Don Thompson: Thank you for sticking to the facts, as always.

    So I ask the group again: Is there a plausible explanation for no ACARS reporting after 18:25 other than deselecting the SATCOM link in the ACARS Manager?

  709. TBill says:

    @Victor
    “Then Jean-Luc will have to explain how TCAS was used while the transponder was in standby.”

    I assume Jean-Luc is suggesting the pilot could turn on Transponder for collision avoidance when out of secondary radar range.

  710. Victor Iannello says:

    @Mick Gilbert said: I’ve also wondered if it might relate to a single or dual FMC failure.

    Since your explanation of a missing Flight ID includes the failure of a single FMC, please explain how a failed FMC would remove the Flight ID used by the remaining FMC.

  711. Mick Gilbert says:

    @Don Thompson

    Don, a clarification please; if the Flight ID is not required for an ACARS connection how is an AES linked to a particular flight for ACARS message routing? (ie how do messages sent via ACARS to a particular flight reach the right AES?)

  712. DennisW says:

    @Mick

    Whichever way you want to look at this it is a series of multiple events of low probability.

    No!

    This way of thinking only applies to an aircraft failure scenario. A deliberate hijacking explains the observables without the necessity of even assigning probabilities to events.

    The DrB approach requires coming up with possibilities for a series of low probability events that provide an alternative to a deliberate hijacking whereas the deliberate hijacking stands on its own.

    A number of posters here reject the deliberate hijacking on the grounds of “unfair accusations”. That is so lame. It is a plea to reject the obvious.

  713. Mick Gilbert says:

    @Victor Iannello

    For the avoidance of any and all doubt, let me say this again;

    A single or dual FMC failure IS NOT my explanation for the missing Flight ID. I have wondered (as in speculated curiously) whether it might (as in a tentative suggestion) be an explanation.

    With regards to a single FMC failure, the FCOM notes that, ‘A software reset may occur while in single FMC operation.‘ The software reset causes a similar but not as extensive disruption to a dual FMC failure, viz the active route becomes inactive, the performance data is erased, and LNAV and VNAV (if engaged) modes fail.

  714. Mick Gilbert says:

    @DennisW

    We’ll just have to disagree on that then Dennis.

  715. DennisW says:

    @Mick

    We’ll just have to disagree on that then Dennis.

    Sure. We can disagree on the deliberate hijacking versus some sequence of events related to a failure.

    There is no room for disagreement on a series of low probability events being needed for your preference. It is what it is.

  716. Victor Iannello says:

    @Mick Gilbert: So the theory you are wondering about requires a single FMC failure and a software reset on the remaining FMC or a dual FMC failure. I understand now. I’ll stop inquiring.

  717. Mick Gilbert says:

    @DennisW

    Just to be clear, a deliberate hijacking/murder-suicide can be relatively simple (1. Take control of the airplane as early as possible, 2. Crash airplane immediately) or relatively complex (1. Take control of airplane, 2. Disable communication systems, 3. Divert airplane, 4. Change course of airplane, 5. Re-enable some communication systems, 6. Change course again, 7. Fly for 5 hours, 8. Crash airplane). Would you say that both are equally as likely?

  718. DennisW says:

    @Mick

    No. I regard the 2nd scenario as very unlikely. That is why I do not think the diversion was a murder/suicide from the onset. The diversion was related to something else, and the murder suicide was a result of the “something else” not working out. I have never been an advocate of the murder/suicide scenario.

    That thinking is what motivated my original CI scenario – a way to prolong the options. Unfortunately the timing of my CI scenario was suboptimal. That was back in the day when all the black belt analysts had pins in a map based on minimizing BTO/BFO errors.

  719. Mick Gilbert says:

    @Victor Iannello

    Victor, I don’t know how others have approached this but my approach has been to cast as wide a net as I can with regards to possible explanations for the observed events, research them as best I can (which normally entails a bazillion or so questions to Andrew) and then eliminate those that are simply not possible (if you think some of this stuff is unlikely you should see the list of stuff that has been eliminated). Everything that is left, however likely or improbable, I entertain as a possibility. I do not just settle on a ‘this must be it‘ explanation and exclude all other considerations. If researching aviation accidents and incidents has taught me anything it is ‘never say never’ and that ‘there’s a first time for just about everything’.

  720. Mick Gilbert says:

    @DennisW

    Okey doke, thank you for that explanation.

  721. HB says:

    The missing flight ID is more puzzling than what i thought. If true it would be difficult to tally this event with an accident. One more item in the list of items to be clarified in the final report. I am wondering why the investigation did not draw any conclusion from this clue when they can actually test the systems. This would tell a lot about the quality of the investigation.
    Also as a side note, is there not a log report available from the ACARS server which could be used to check if there was an attempt to log on it? This could be used to cross check information.

  722. DennisW says:

    @Mick

    Everything that is left, however likely or improbable, I entertain as a possibility. I do not just settle on a ‘this must be it‘ explanation and exclude all other considerations.

    Your approach is typical, and is often referred to as “complexity bias”. We all suffer from it, and it is exploited by marketeers.

    https://www.fs.blog/2018/01/complexity-bias/

  723. DrB says:

    @Andrew,
    @Peter Norton,

    The revised version of “Observational Data” incorporating your suggestions, is available HERE .

    Any errors are mine alone. Let me know what I have got wrong or omitted.

  724. Don Thompson says:

    @Mick Gilbert asked “if the Flight ID is not required for an ACARS connection how is an AES linked to a particular flight for ACARS message routing?

    I have purposely stressed that the AES Log On may include an ‘Alphanumeric Flight Identifier‘. This is not equivalent to the ACARS Flight Identifier. The Classic Aero SDM describes that the ‘Alphanumeric Flight Identifier’ may be defined in the Log On for use by the Inmarsat Network Channel Management System, not for any role in the delivery of ACARS messages.

    The ACARS Flight Identifier (ACARS FI field) is not involved with message delivery. I purposely avoid the term ‘routing’, in today’s understanding of networks routing is often a dynamic concept. There is nothing dynamic in the ACARS working!

    The AN, Aircraft Number, field in the ACARS message header is necessary for ground-to-air/air-to-ground message delivery. The AN, using the aircraft registration, is unique to an aircraft, e.g. 9M-MRO.

    The FI field of the ACARS message, Flight Identifier – eg MH0370, is not significant for message delivery but, dependent on message type, it may be significant in processing a message. For example, identifying an OOOI message for a flight segment of which there may be many on any single day.

    SATCOM emulates the VHF ACARS datalink service. Perhaps… consider the GES as similar to the VHF RGS, but rather than broadcasting into the ether as the RGS does, the SATCOM datalink that is established at Log On provides a dedicated logical channel to the AES for ACARS traffic. That SATCOM logical channel is extended from the GES, via a virtual circuit, to the ACARS server. The AES Log On to the GES connects the ‘plumbing’, but the ACARS communication session is only established when the ACARS function on the aircraft initiates correspondence.

  725. Don Thompson says:

    @HB asks “ is there not a log report available from the ACARS server which could be used to check if there was an attempt to log on it?

    A ‘log in’ isn’t really the concept here. Admittance to the ‘network’ is, essentially, permitted with an assumption of the appropriate equipment and some static identification credentials including an aircraft 24bit hex address and its registration¹.

    The Factual Information does include the ACARS Traffic Log which records the traffic exchanged with 9M-MRO by the MAS Operations Dispatch/Control Centre (ODC/OCC²). The Log records messages between 9M-MRO and the SITA message exchanger host QCSXMXS.

    [1] ‘Appropriate equipment’ is no longer a significant barrier, with today’s SDR tech it’s not so difficult to intercept aeronautical communications. However, exploiting an intercept to effect control is an entirely different matter.

    [2] ODC-OCC: select preferred acronym.

  726. Victor Iannello says:

    @DrB: I’d say that if your intention is to provide Observational Data and supporting facts to test various scenarios, it is important to note that nobody has yet proposed a valid reason for no Flight ID at the 18:25 log-on and no ACARS messages after log-on other than deselecting the SATCOM link in the ACARS Manager. (@Mick Gilbert has proposed multiple problems with the FMCs causing a loss of Flight ID, but this would not affect the ACARS log-on.) I may be harping on this point, but I do think it’s important if you are trying to discriminate between scenarios.

  727. Victor Iannello says:

    @HB said: The missing flight ID is more puzzling than what i thought. If true it would be difficult to tally this event with an accident.

    The missing Flight ID together with the absence of an ACARS session with the ACARS server are quite damaging to any explanations involving an accident scenario, in my opinion.

  728. DrB says:

    @Victor Iannello,

    Yes, as I have said several times, that is the purpose of the “Observational Data” list – to allow discrimination of any and all “flight failure” scenarios.

    Accident scenarios now have 4 holes that I can see. There are possibly no explanations for the following observations:

    (1) lack of Flight ID in 18:25 AES log-on,
    (2) lack of post-17:07:29 ACARS data transmissions,
    (3) lack of transponder returns after 17:21:13, and
    (4) lack of post 00:19:30 radio transmissions.

    A possible explanation for #4 is physical damage to the Audio Management Unit, but loss of power to multiple radios does not seem to be a plausible reason.

    For #3 the loss of transponder power does not appear to be a valid reason since there are multiple backups. There is one person on Twitter claiming to be a Boeing pilot who says that loss of certain air data can inhibit the transponder function. I certainly don’t know if this is true or not. Perhaps someone here can investigate this claim to see if it can be settled one way or the other.

    So far, there are no “accident scenario” explanations for items #1 and #2 that have survived scrutiny.

    The current ledger for a malevolent pilot scenario is clean, with all observations being explainable by actions that could have been performed by one pilot within a locked flight deck.

  729. Peter Norton says:

    @DrB:
    I defer to you and others with more knowledge. From my more limited point of view, I think you solved all remaining issues elegantly. I like the result, excellent work Dr.

    one apology:

    (1c) ” […] and must be must be re-initiated […] “
    I am very sorry for this error I overlooked in my text submission. My apologies.

    and one suggestion:

    I would merge (1f) into (1a) and (1b) because they are technically linked together:

    (1a) One method to achieve this, is unchecking both SATCOM and VHF radio boxes in the Data Communications System ACARS Manager screen presented on the FMC Multi-Function Display.”

    (1b) ACARS reports also stop if VHF radio is not selected for ACARS transmission (see 1a) and the Left AC bus is depowered, which disables the SDU (Satellite Data Unit).”

    (1f) could then be removed.
    If the second part of (1f) (“but there are no reports that any such data were received”) should be accomodated, it could be added to (1) like so:
    (1) No ACARS reports were sent received after 17:07:29 (neither via satellite nor radio)“.

    PS: My statements need verification by someone better qualified before they could be used.

  730. Peter Norton says:

    > DrB says:
    > […]
    > (4) lack of post 00:19:30 radio transmissions.

    17:19:30 UTC = 01:19:30 MYT

  731. Victor Iannello says:

    @DrB said: There is one person on Twitter claiming to be a Boeing pilot who says that loss of certain air data can inhibit the transponder function.

    That person (@Paul Onions aka @Pete Noetic) was banned here because I got tired of correcting his long list of technical errors, which he would just repeat. I doubt he is a Boeing pilot, but I can’t be sure. Maybe he considers flying the PMDG 777 equivalent to being a Boeing pilot.

  732. David says:

    @DennisW. You said to Mick Gilbert,”There is no room for disagreement on a series of low probability events being needed for your preference. It is what it is.”

    Mick Gilbert said, “Everything that is left, however likely or improbable, I entertain as a possibility. I do not just settle on a ‘this must be it‘ explanation and exclude all other considerations.”

    You responded, “Your approach is typical, and is often referred to as “complexity bias”. We all suffer from it, and it is exploited by marketeers.”

    Do you think the Comet crashes were caused by pilot hijacking?
    In fact the complex and novel cause would not have been found if such facile solutions as bombs on board had been accepted.

    Your hypothesis might prove right but it is not self-evidently so as yet. Elsewhere you have agreed that you cannot explain non-disclosure of purpose in such a pijacking; and implicitly have agreed its importance.
    Without that please do not disdain other possibilities just because you think they are complex. They might be both complex and right.

    From your reference, “Occam’s razor suggests that the simplest solution or explanation is usually the correct one……”

    and,

    “An important point to note is that Occam’s razor does not state that the simplest hypothesis is the correct one, but states rather that it is the best option before the establishment of empirical evidence.”

    Not very helpful really. With the Comet, fatigue proved to be the simplest solution because it was the only one.

  733. HB says:

    @Victor, DrB: re flight ID
    Regardless of the very low probabilities, what about a total DC power failure for the missing flight ID? This would lead to a reboot of the FMC and other systems, FMC is DC powered? Not sure on the effect on the IDE/satcom and other AC powered electronics. The scenario could involve a DC arcing/fire, crew isolating the fault (not knowing where the fault is, crew isolates the entire system DC powered electronics including batteries), most electronics will be down and AC powered system wont get input from FMC and DC power if required, after fault isolation, power is restablished to the functioning equipment. After power restablishement most likely one or more dc bus would be left isolated. I am not sure whether this could lead to a missing ID and what is the procedure to respond to a DC fault. Normally it should be to isolate the faulty equipment only but it can be tricky if the fault is on the power suppy to the FMC equipment. A DC fault can also discharge the batteries. Not sure it fits all the criteria, Just suggesting.

  734. Mick Gilbert says:

    @DennisW

    Thank you for that article, Dennis, it’s an interesting read (nice that they reference Judgement Under Uncertainty, I’m a fan of Tversky and more particularly Kahneman). I have no preference for more complex explanations I am simply averse to rejecting them outright.

    There’s no shortage of fodder for behavioural psychologists in this exercise, there are a whole raft of biases potentially in play that’s for sure. It is interesting, if not instructive, that just about any time there is a choice between hypotheses involving people (or a person) and ones involving inanimates such as components and systems, anthropocentrism/anthropomorphism tends to see us prefer the former over the latter. We are, by nature, a storytelling animal so we have a preference for a story over a dissertation and we prefer stories about people over objects. That goes so far as when we have to include objects in stories we have a preference for imbuing them with human characteristics. And it also means that we often oversimplify agent-based models, despite their inherent complexity.

    Any old how, we could spend a lot of time debating the virtues of Occam’s Razor over Hickam’s Dictum, particularly as they apply to complex systems, and it would probably not advance the likelihood of finding the debris field one jot.

  735. DennisW says:

    @David

    I have no idea what you are talking about.

  736. DennisW says:

    @Mick

    Any old how, we could spend a lot of time debating the virtues of Occam’s Razor over Hickam’s Dictum, particularly as they apply to complex systems, and it would probably not advance the likelihood of finding the debris field one jot.

    True.

  737. Mick Gilbert says:

    @Don Thompson

    Thank you for the expanded explanation, Don. I don’t doubt your explanation but does it not strike you as odd that a piece of data that ACARS doesn’t appear to be dependent upon is nevertheless inhibited from being passed to the SDU because of a selection in the ACARS Manager?

    Are you aware of any other reasons why ACARS would not connect other than deselecting SATCOM in the ACARS Manager? For example, how are the FMCF and FDCF functions physically connected to the SDU?

  738. David says:

    @DennisW. “I have no idea what you are talking about.”
    Never mind.

  739. Mick Gilbert says:

    @Don Thompson

    Don, sorry to labour the point on this matter but if the Flight ID is inhibited from being passed to the SDU when SATCOM is deselected in the ACARS Manager, that would mean that when only VHF ENABLE is selected in the ACARS Manager (which is a not uncommon setting as I understand it), then, because ostensibly no communication is possible from the FMCF and FDCF functions to the SDU, the Flight ID would be absent from all other (ie non-ACARS) SATCOM signalling and traffic, wouldn’t it? That doesn’t sound right.

  740. Don Thompson says:

    @Mick Gilbert asked “does it not strike you as odd…

    No, it doesn’t strike me as odd. There are many elements that comprise this system, they are broken/decomposed into functions and sub-functions. The core GES Log On (Log On Request thru to Log On Acknowledge, without ‘Alphanumeric Flight Identifier’ provides all the facilities needed to transact packet data exchanges for ACARS or ISO-8208SSN, and voice traffic.

    and “… the Flight ID would be absent from all other (ie non-ACARS) SATCOM signalling and traffic, wouldn’t it?

    No, it wouldn’t. My position is that the function/sub-function that includes processing of the available ICAO Flight ID as the ‘Alphanumeric Flight ID’ for the Log On is inhibited, not that distribution of that word is in any way inhibited.

    A process is necessary to recognise that the FMC sourced Flight ID is non-null, or has changed from one valid value to another, and consequently submit a command block to the SDU that initiates a Log On renewal containing the encoded ‘Alphanumeric Flight ID’ – my position is that process is inhibited when the ACARS Manager check box for SATCOM is cleared. That process is not resident in the SDU.

  741. Andrew says:

    @HB

    RE: “…what about a total DC power failure for the missing flight ID?”

    The FMCF is hosted within both L & R AIMS cabinets, which are powered by several DC sources, including the L & R DC buses, the Capt & FO Flt Instrument buses and ultimately the Hot Battery bus. Operation of the DC system is automatic. The L & R DC buses could be isolated by de-powering both transfer buses, but that would cause the RAT to deploy, ensuring the Flt Instrument buses remained powered. Additionally, the Hot Battery bus can’t be de-powered by the flight crew. The DC system can’t be completely isolated and the FMCF will not therefore ‘reboot’.

  742. Peter Norton says:

    @DrB:
    I also noted that you removed the cockpit door lock from (4) (affected by L AC bus).
    Can we add it to (4a) or (4b) (unaffected) ?

  743. Julia says:

    If I remember correctly, the flies on Zaharie’s flight simulator, which tracked the path towards the Indian Ocean, i.e the path we now know MH370 took, had been deleted by Zaharie. The FBI and possibly Malaysian criminal investigators but more probably the FBI retrieved the files from the hard drive. They took away his flight simulator from his home for investigation.So Zaharie had plotted the route but had deleted the very same file at some point before the ill fated flight he piloted.I am sure the date Zaharie deleted the file is available and was very near to the day/night of the flight.

  744. Victor Iannello says:

    @Julia: A summary of what we know about the simulator data is here.

  745. Mick Gilbert says:

    @Don Thompson

    Don, when as part of normal operations where VHF ENABLE has been checked and SATCOM ENABLE has been left unchecked in the ACARS Manager, would you expect to see the SATCOM log-on complete with no Flight ID sent to the GES?

  746. Julia says:

    Thank you @victor.This is a very detailed and informative document you produced re. the pilot’s simulator
    http://mh370.radiantphysics.com/2017/10/12/simulator-data-from-computer-of-mh370-captain-part-1/

    It leaves the reader almost in no doubt and made me feel very sad reading your findings. It’s important to keep these results and your paper at the forefront of our minds as memories fade fast.

  747. ventus45 says:

    I have mentioned this document and this video before, but no one here seems to have taken any serious interest in their contents, which I find surprising.

    Now that the CSIRO areas have been searched by OI’s AUV’s, with no result, and given the number of data holidays in the prior search areas, (see slides “Data Gap Capture”, “Data Gaps”, “700+ Anomalies”, and “Anomalies to investigate”, and from 18min24sec forward in the video), what prospect is there that OI will go back and cover those areas on completion of the northern search ?

  748. Peter Norton says:

    > Victor Iannello says:
    > he might have been aware that satellite calls that connect but are not answered
    > might look different to the calling station than satellite calls that never connect.
    > Disabling ACARS would still allow calls to connect. Until he was out of Malaysia,
    > he might have preferred for ATC and MAS to believe the plane had crashed.

    @DrB:
    I think Victor’s explanation is compelling and could also be added to your PDF as follows:

    « (4c) When the SDU is depowered, the satellite phones are inoperative and incoming satphone calls don’t connect – leaving mobile phones as the only remaining means of communication from the cabin. Alternatively the satellite phones can be disabled by depowering the IFE at the overhead panel, but with the SDU remaining operative incoming satphone calls still connect, informing the ground that the plane has not crashed. »

  749. Ge Rijn says:

    @VictorI @Richard Godfrey

    RE: “@Ge Rijn: Stop your whining. 97E longitude will be searched. You are not the first here to be quite certain about an endpoint to the exclusion of all others.”

    97E is ~15Nm beyond the +/-25Nm zone. That was/is my concern. But if you say it will be searched I’m comfortabled with that. BTW I never excluded other possibilities more north including yours. I advocated mostly on an area between ~96E/~97E and ~32S/33S. And yes I mentioned my favorite at ~32.6’S/97E. What’s wrong with that? You and others do the same.
    I realize many options are still open. Then don’t throw away a possible baby with the bathwater. It won’t look nice.

    RE: “I have spent weeks analysing, what I call the “Flaperon” and what you call the “Blue Panel”.

    It was this what got me off-guard. Why you called it the “Flaperon” while I was being more cautious by still calling it the “Blue Panel”?
    You’re ‘auto-correction’ seemed (and still seems) strange to me.

    But we’re making headway. The plane is also not close to the 7th arc between ~33S and ~32S as it seems at this point in time.

  750. HB says:

    @Andrew, re DC isolation
    It is interesting to know that the hot battery bus cannot be isolated. So a fault in there would rely on the electrical protection. This cannot even be isolated locally by manually switching the breakers? Or that area is not accessible?

  751. DennisW says:

    @all

    I am amused by the efforts of those who are advocating the accident scenario. It is too late to have any effect on the search activity, and the answers won’t be forthcoming until the FDR is recovered. What is the point of it?

  752. HB says:

    @DennisW, re What is the point?
    The ultimate objective for a number of people is not to find the plane but what has actually happened and how it happend to make the sky safer. It appears that there is enough data to rule out an accidental scenario. A detailed analysis is required for that. Then all focus can be on the most likely scenario which strangely enough has not progressed one inch in terms of clues since the RMP prelim report.

  753. Peter Norton says:

    ventus45: “no one here seems to have taken any serious interest in their contents”

    I did, but don’t want to get in the way of the flight ID discussion. Maybe on a quiet day and if allowed.
    (Or else past the 90 days when the question will have another status?)

    From a purely pragmatic point of view, OI will always deem more cost-effective to search in 100% unexplored areas than in 7%* unexplored ones. *(in some regions)

  754. Victor Iannello says:

    @HB said: Then all focus can be on the most likely scenario which strangely enough has not progressed one inch in terms of clues since the RMP prelim report.

    Why do you think that is? Why do you expect that to change?

  755. TBill says:

    @Ventus45
    I am still focus on that recent radio interview you posted. One of the things they implied was the Digital Flight Data Recorder, if not found, then other electronics (eg; flight computers) if found could be used to re-assemble the flight data. So that has me wondering if there could have been attempts to clean off that data prior to the crash.

  756. DrB says:

    @All,

    Revision 6 of “Observational MH370 Data” incorporating additional suggestions, is available HERE . There are numerous edits and additions. Any errors are mine alone.

  757. Nederland says:

    @DrB

    FI, p. 49: mentions “cabin telephony calls, were that function is available. In the case of 9M-MRO, the in-seat phones can only be used for seat-to-seat calling.”

    This would explain why there were no cabin satellite phone calls (outgoing or incoming), even with the IFE switched on. On the other hand, there was an SMS/email application, also powered through the IFE.

  758. David says:

    @DrB. See your 2b., reasons for transponder not transponding.

    A comment then re your 3d:, I think the left AC bus power loss would have been simultaneous with transponder failure had the back-up generators been switched off earlier.

  759. David says:

    @Don Thompson. A curiosity if you would.
    If the SATCOM system is logged off the SDU can be logged on manually, selecting “automatic” or by selecting a GES (AMM 23-15-00 p48).
    Since automatic log-on should have occurred automatically on any power up, what would be the circumstances where manual log-on would be available/necessary?

  760. HB says:

    @Victor,
    The most important in scenario analysis is to understand the WHAT and the HOW, WHERE and WHEN is secondary, the WHO is also secondary unless crime is involved. It is clear that the current strategy is not to progress other scenarios so long as an accidental scenario remains possible and as such the argument relies on finding the plane to find out WHAT has happened. Finding the plane may tell us WHAT and HOW but there is data already to tell WHAT and the HOW could be obtained with further analysis. If it is a highjack scenario, the HOW information may not even come from finding the plane or the black box unfortunately; hence the strategy that has focussed on the WHERE is flawed in my opinion and actually very frustrating. There is a good reason to progress the case on the highjack scenario as opposed to wait IF we find the plane but that’s my opinion. One can speculate why this has not progressed. Conflict of interests amongst the various departments and officials could be one reason, inexperience could be another one given this is the first case, fear of a public enquiry or a criminal law suit another, common factors are not excluded, who knows why this strategy persists. The excuse communicated in April 2014 by the DCA is that both investigations cannot be progressed unless the black boxes are found. Not sure whether all agree to this, i don’t. Maybe a public enquiry could unlock this status quo.

  761. William Shea says:

    Hijack. Who claimed responsibility?

  762. William Shea says:

    That’s what Hijackers do…

  763. William Shea says:

    That’s what they do…

  764. David says:

    @William Shea. My 5th paragraph of the below refers, http://mh370.radiantphysics.com/2018/03/18/mh370-search-update-mar-18-2018/#comment-14065 my April 1st, 8:32 PM

    @HB. “Maybe a public enquiry could unlock the status quo.”

    You have a point, particularly as to ‘what next’ should the search be unsuccessful, which should be known reasonably shortly. As you say, it may be that even were the wreckage found the outcome might be inconclusive anyway.

    Hopefully the final report might give a clue as to what further information should be sought; and that could relate to the aircraft captain in particular.

    However the police might not be seen to have the powers to compel evidence from Zaharie’s family, Fatima Pardi and others should they wish.

    They do have Royal Commissions in Malaysia. These have that power (no right to remain silent). See para 5. first sentence: https://asklegal.my/p/five-facts-on-the-royal-commission-of-inquiry-rci

    Currently there is one into a “forex event” of some time ago:
    http://www.bnm.gov.my/index.php?ch=en_press&pg=en_press&ac=4451

    As per the first URL they need not be manned and led by lawyers/judges though in Australia commonly they are. Not all are successful. For example, one in Australia replete with a judge and lawyers set up to examine a 1964 fatal naval collision was a case in point, one Crown lawyer in particular acting as a prosecutor, apparently from habit, as distinct from seeking the co-operation of witnesses. A repeat Royal Commission was better with the blame game but was short of the organic naval skills and experience needed to understand the close dynamic night naval manoeuvres’ scene comprehensively.

    My point is that Royal Commissions have the powers but not necessarily the wherewithal unless carefully constituted, their terms of reference broad and there is a means of expert review of their findings to advise government and parliament separate to the political grandstanding.

    Heavens knows how ICAO would react to that approach.

    I mean this as an adjunct to your question.

  765. Ge Rijn says:

    SC seems allready to have come within ~50km of the 32.26`S/97E. This point reflex the ~middle of a +5100m hole at the beginning of another deep trench towards the east.
    I’m truly hopefull now SC is also going to search the surroundings of this lat/long point if they don’t find the plane where they are searching at this time.
    I have a feel they could find it anytime now. Fingers crossed.

    https://twitter.com/LabratSR

  766. David says:

    Search Weekly Report 10 out. Favourable weather and seas the week ahead.
    http://mh370.gov.my/en/mh370-underwater-search-2018

  767. Barry Carlson says:

    @David,

    I could have got the same advice from the Bureau of Meteorology.

  768. David says:

    @Barry Carlson. Maybe that was a source Barry though its not just the winds and seas: I imagine they are informed about the residual swells by OI.

  769. Barry Carlson says:

    @David,

    I was attempting to place a degree of sarcasm on the MY weekly report – about nothing. IOW they haven’t told us anything we don’t already know.

  770. Ge Rijn says:

    @David

    One thing I notice among all weekly reports till now is how calm the weather and ocean have been from the start.
    This is in contrast to the general perception often mentioned in the past of a wild and turbulent SIO.

  771. Andrew says:

    @HB

    RE: ”It is interesting to know that the hot battery bus cannot be isolated. So a fault in there would rely on the electrical protection. This cannot even be isolated locally by manually switching the breakers? Or that area is not accessible?”

    The area is not accessible.

  772. TBill says:

    @Ge Rijn
    I think you are thinking the crash site might be in the deep feature just to the southeast of the cross-hatch area in my BR search area sketch.

    https://pbs.twimg.com/media/DZqq8tkWAAUUxn0.jpg

    It does look interesting, but there are plenty of volcanos, trenches, rocky slopes, and perhaps better resolution on Google Earth (for pre-planning) in the +-25 nm area.

  773. Victor Iannello says:

    @HB: I think you are advocating that investigators abandon the possibility of an accidental disappearance and concentrate on intentional diversion of MH370. First, there is not unanimous belief that the disappearance was intentional, and there probably never will be, even if the debris field is found. If intentional, the captain becomes the prime suspect, and that option is just too emotionally charged for all people to accept without an overwhelming amount of evidence. Second, if intentional, the disappearance was a criminal act, and Malaysia is charged with leading the criminal investigation. I don’t see any circumstance in which this would change, and I don’t see any circumstance in which Malaysia will impartially conduct a criminal investigation. So considering this, I think finding the plane is our best hope in solving the mystery.

  774. HB says:

    @Victor, i am not advocating any abandon. There are two investigations and there is no reason to wait for the plane to be found to progress them. That s what i am advocating.

  775. Ge Rijn says:

    @TBill

    Yes, I mean that +5100 deep hole around 32.26’S/97E.
    One argument is it is easily found on Google-Earth by someone planning a disappearance this way and with a good chance of result. He would only have to aim near this coördinates. I don’t think a pilot planning something like this would bother aiming at a very specific rocky slope, part of a trench or a volcano. I think it would be too detailed with too many factors involved and therefore too difficult to accomplish.

    Another argument is the area could fit with reverse drift (imo) applied on the ‘Blue Panel’ and associated spotted debris field on 28-3-2014 by the RNZAF at 32.4S/97.8E(degrees). Roughly counting back in time the ‘Blue Panel’ and debris field could have been in this area on 8-3-2014 or closer to the 7th arc.

    Now it seems nothing has still been found closer to the 7th arc between 32S/33S this option to the east is still open and is still a serious option imo.

  776. Victor Iannello says:

    @HB: Yes, the investigation should diligently proceed on all fronts. I don’t think there would be any disagreement among contributors here. Unfortunately, we have little influence, other than to provide input on where OI should scan, and encourage the search to continue as long as possible.

  777. Rob says:

    @VictorI

    Why did the ATSB and the Malaysian government sit on Shah’s simulator data? The data indicates that he was considering a great circle path directly into the SOI, flying at high altitude I would venture to suggest, towards a manually-inserted waypoint.

    That suggests minimum maneuvering between the SLOP and fuel exhaustion, doesn’t it? That tends not to give support to the loiter scenario doesn’t it? That also incidentally, supports the DSTG minimum manoeuvers assumption iro the Bayesian priors?

    So just what is the technical justification for searching further north?

    It seems to me that the primary and only impetus for searching further and further northward along the 7th arc was the ATSB First Principles Meeting, when they proposed the 25,000sq km priority search zone.

    They proposed the priority zone on the following grounds

    1)Aircraft not found in the Indicative Search Zone

    2)Their interpretation of flaps retracted on impact

    3)Their interpretation of the condition of the debris

    4)Their interpretation of steep descent suggested by final BFO, two minutes after fuel exhaustion.

    5)New drift modeling by CSIRO pointing to 3 priority hotspots.

    As soon as the 25,000sqkm search area became established as the focus of future attention, the effects of confirmation bias then rejected any logical arguments against the ATSB’s original points 2, 3 and 4 out of hand.

    What price now the CSIRO drift modeling? I remember headlines such as “experts say they are now confident they know where MH370 crashed”

    What a complete mess.

  778. Ge Rijn says:

    @TBill

    Another argument to add is the coïncidence of a flaperon-like piece as the ‘Blue Panel’ together with ~13 other spotted pieces of debris being near the 7th arc at 8-3-2014 (according my reverse drift estimations) is at least quite remarkable.

    But this also counts for @VictorI’s debris spotted on 29-3-2014 by the RNZAF around 29S west of the arc.
    Although no debris in those captured pictures show any direct resemblance with possible aircraft debris this would be the next best option imo.

    More north the drift analysis predict weaker and weaker possibilities but I surely encourage a search till ~28S.

  779. Victor Iannello says:

    @Rob: We’ve been through this too many times. I’m not doing it again.

    You need to ask yourself why, despite your persistent pleas and long-winded comments, you fail to gain support for your theory. I can’t think of any contributor here other than one that agrees with your logic. The one exception that comes to mind is @Byron Bailey. Unfortunately, he disappeared when he was challenged on some of his basic assumptions (like a rapid descent was required for RAT power). So is there a conspiracy to disagree with your theory? Do we all suffer from the same confirmation bias? How do you explain this?

  780. Ge Rijn says:

    @Rob

    It’s not the CSIRO drift modeling that failed. Those are what they are based on the best scientific input. And therefore they are still relevant.

    It’s just the link they made made with the satelite images that proved to be wrong.

  781. TBill says:

    @David
    “Search Weekly Report 10 out. Favourable weather and seas the week ahead.”

    Does the link to the document seem broken now? I was able to read a fuzzy version.

  782. Ge Rijn says:

    @Rob

    But I understand your spectrum-condition limits you to see other views.
    Still I wonder, because you are aware of your condition, you are still not able to consider your views could be wrong. Is it so deeply ingrained?

  783. Rob says:

    @Ge Rijn

    Yadda yadda yadda.

    Ge Rijn, you are in thrall to collective confirmation bias, a bias making you and and the other members of your group close ranks in order to defend your beliefs and assumptions. It’s a well known psychological phenomenon that has baffled psychologists for years; why does a group close ranks and ignore any rational arguments that contradict their beliefs. Core to this phenomenon is the human need to belong to a group. Anything that threatens the group has to be attacked or rejected. Ge Rijn, imagine (and you are going to need a powerful imagination here) imagine that all your peers and associates signed up to the ATSB Priority Search Zone, and accepted uncritically all it’s assumptions and tenets, all except you. Only you were like the boy in the Hans Anderson story about the Emperor’s New Clothes. Only you could see the Emperor was naked because you were the only one not to be brainwashed by the con-men. You would feel very lonely wouldn’t you?

    Same situation. Your desire to be a member of the group blinds you to rationality.

    And the rest of the group are supporting you in your delusional behaviour. They have to. They aren’t free to tell you the truth about the Blue Panel, the panel lost from a ship rather than a plane. You all support each other in your delusional thinking. No one can contradict the holy of holies the experts, the ATSB. That is verboten.

    This is the route to hell. This is the route to disappointment and humiliation.

  784. Ge Rijn says:

    @Rob

    As you might have noticed no-one here is obviously supporting me in my opininions/behaviours. In this sence I’m a howling wolf many times just like you.
    The difference between you and me is I can consider other opinions and facts and change my mind-set according new information.
    I know I’m almost as stubborn as you but I’m open to new information and are willing to adapt my views accordingly.
    If you are not able to see different possibilities I can imagine you feel quite lonely at times.
    But this doesn’t make you less imo. It’s maybe just one ability you don’t posess.

  785. Richard Godfrey says:

    SC continues to make good progress in searching the Diamantina Escarpment towards the Broken Ridge.

    https://www.dropbox.com/s/qa4oawqtwxk2a0f/SC%20Track%2003042018.pdf?dl=0

  786. TBill says:

    @Richard
    Thank you…what does the solid yellow “center” line represent? Looks like the search center is offset from that line.

  787. TBill says:

    ALSM has tweeted that OI has moved north of BR now

  788. Victor Iannello says:

    @Rob: I can’t take your lunacy any longer. If you think people in this group are closing ranks to protect each other, you really are delusional. I will only allow further comments from you that add to the discussion.

  789. David says:

    @TBill.”Does the link to the document seem broken now? I was able to read a fuzzy version.”
    Yes the DCA page no longer is available here though others are.
    My downloaded version:

    https://www.dropbox.com/s/0kdlyqazev0u8jd/MH370%20Search%20Weekly%20Report%2010_English.pdf?dl=0

  790. David says:

    Expect a new MH370 documentary in Aust, Britain, France:

    http://www.c21media.net/intl-nets-examine-mh370-mystery/

  791. Don Thompson says:

    @David

    MH370 documentary…

    Produced in UK by Knickerbockerglory, an independent production company launched in 2011 by Jonathan Stadlen.

    Promo photograph shows Defence Minister Hishammudin, DCA/CAAM DG Azharuddin, and former RMAF chief Rodzali Daud on set (I don’t recognise the individual at extreme left).

    Wouldn’t it be a freaky coincidence if executive producer Jonathan Stadlen was actually the brother of Paul Stadlen, long rumoured to be the lynchpin of Najib’s PR operation?

  792. Barry Carlson says:

    @All,

    The Malaysian government on Monday 2nd April made it illegal for any person to publish in any format “fake news” that defames Malaysia and/or its citizens….

    The law defines fake news as “news, information, data and reports which is or are wholly or partly false” and includes features, visuals and audio recordings.

    It covers digital publications and social media and will apply to offenders who maliciously spread “fake news” inside and outside Malaysia, including foreigners, if Malaysia or a Malaysian citizen were affected.

    Refer to the Reuters article.

  793. DrB says:

    @David,

    You said: “I think the left AC bus power loss would have been simultaneous with transponder failure had the back-up generators been switched off earlier.”

    My understanding (which may be incorrect) is that the L Transponder is powered by the AC Standby bus, which is normally powered by the L Transfer Bus which is normally powered by the L Main AC Bus. I thought if both Backup Generators were turned off, the Standby Bus would automatically switch over to be powered by the R Transfer Bus in the case you described where the L Main AC Bus is also turned off and both Backup Generators are turned off. Maybe I am not correctly understanding how this works, but it looks to me like the AC Standby Bus (and therefore the L Transponder) is extremely redundant in power supplies and requires loss of both IDGs, both Backup generators, and the standby power system consisting of the RAT and the main battery plus its static inverter, to lose power in flight.

  794. DrB says:

    @Nederland,

    You said: “In the case of 9M-MRO, the in-seat phones can only be used for seat-to-seat calling.”

    You are correct. The Boeing manual says that only “interphone” calls to specific seats or handset locations are possible. I could not find anything to confirm the SMS or email capabilities using the IFE. Can you provide a reference for that?

  795. DrB says:

    @All,

    Concerning the ACARS software settings, I received some additional information I would like to confirm/deny if possible:

    1. The ACARS version (of two possibilities) in 9M-MRO is the earlier one indicated as [Not AIMS 2003] in “All Versions” manuals.
    2. There are three mutually exclusive ACARS Modes: VHF, SATCOM, and AUTO.
    3. It is possible to check either one of these three boxes or none of these three boxes. You can’t check two or three of them.
    4. AUTO uses VHF first; if that fails it switches to SATCOM.

    To stop all ACARS data from being sent, one could do one of the following:
    (a) uncheck all three boxes in the ACARS Mode page, or
    (b) if the ACARS Mode is set to VHF, that brings up the VHF Manager Page, and if DEFAULT RADIO MODE VOICE is set for the DEFAULT RADIO unit, no ACARS data will be transmitted, or
    (c) if the ACARS Mode is set to SATCOM, shut down power to the SDU.

    In MH370 all ACARS messages were received via SATCOM, so SATCOM most likely was the ACARS Mode setting through 17:07.

    Does anyone have information contrary to this?

  796. Mick Gilbert says:

    @DrB

    Re: ‘2. There are three mutually exclusive ACARS Modes: VHF, SATCOM, and AUTO.
    3. It is possible to check either one of these three boxes or none of these three boxes. You can’t check two or three of them.

    That’s not quite right. On 9M-MRO there were only two selections available on ACARS Manager Page 2/2; VHF ENABLE and SATCOM ENABLE. There is no AUTO selection available. However, when both VHF and SATCOM are selected this permits ACARS to automatically use VHF or SATCOM (if VHF is unavailable).

  797. Andrew says:

    @DrB

    RE: ”3. It is possible to check either one of these three boxes or none of these three boxes. You can’t check two or three of them.”

    There are three modes (VHF, SATCOM, or AUTO), as you described, but the ACARS Manager page only allows two selections, VHF ENABLE and SATCOM enable. The AUTO Mode is enabled when both the VHF ENABLE and SATCOM ENABLE boxes are checked.

  798. Andrew says:

    @Mick Gilbert

    SNAP!

  799. David says:

    @Dr B. “My understanding (which may be incorrect) is that the L Transponder is powered by the AC Standby bus, ….”

    I cannot find a direct reference in the AMM and have not looked at the Training Manual etc as yet but I was supposing it to be the left transfer bus, direct as per Don Thompson, (28th 5.55 pm), “The ATC Transponders are supplied by via the Transfer Buses.”

    Also I referred to your 2b which alludes to the left getting power from there ie implicitly with no standby AC backup.

    With no power to the left AC bus (thence to the left transfer) or from backup gens it would not work, I believe. The right transponder would, powered by the right IDG, that powering the right transfer. The right transfer would not be connected to the left in that configuration.

  800. David says:

    @Dr B. Confirmed. More on the transponder system in that section:

    https://www.dropbox.com/s/r1mgrj6w73zi2ot/Transponder%20power%20supplies.pdf?dl=0

  801. David says:

    Dr B. Clarification. Your view confirmed I believe.
    Your 2b might need a change though.

  802. Andrew says:

    @David
    @DrB

    @DrB is correct, the L transponder is powered by the AC Standby bus (see TM 34-53-00 p.13).

    The AC STBY bus is normally powered by the L AC XFR bus. If the L AC XFR bus is unpowered (voltage less than 80V), the AC STBY PWR relay connects the AC Standby bus to the static inverter, which is powered by the standby power system via the Battery bus. (see TM 24-33-00 pp.211/214).

  803. David says:

    @Dr B. A casualty is that my touted notion that a pilot in switching off all engine generator AC would ensure the aircraft went dark, so I can cross that off my possibilities list.

    A side benefit of the work put into your listing thanks.

  804. DennisW says:

    @David

    @Dr B. A casualty is that my touted notion that a pilot in switching off all engine generator AC would ensure the aircraft went dark, so I can cross that off my possibilities list.

    A side benefit of the work put into your listing thanks.

    I can only shake head at the level of commentary here.

  805. HB says:

    @Andrew, is there is not power to the X buses the RAT deployment will be automatic as i understand. If the intention (whatever the reason response to an electrical fault or intentional to remove the Flight ID or whatever else) is to shutdown all the DC power, one possibility is to isolate the power sources and wait until batteries are discharged. However, this won’t be possible due to RAT deployment.
    My questions are
    (1) whether RAT deployment could be prevented or whether power from the RAT can be isolated?
    (2) what is the endurance time of the main batteries? is the timing could match any observed re-log on?
    I presume the transponder will remain powered in that period also from the standby bus so long as the batteries are available.

  806. Ge Rijn says:

    It seems clear SC has skipped the oportunity to search around 32.26’S/97E based on the ‘Blue Panel’ and associated spotted debris.
    And they were only ~50km away from this area. It would have cost them probably only two or three days to scan this area.

    I’m kind of baffled the ‘Blue Panel’ and associated debris is not handled with the same high priority status as the CSIRO/Griffin satelite images have been and @VictorI’s debris pictures west of ~29S will be.

    But I assume the unproven and controversial assumption of a high speed impact close to the 7th arc probably is still the main reason not to look beyond the +/-25Nm bounderies. I think this could turn out to be the basic mistake causing ultimate failure of the search.

    To me it’s completely clear that if the flaperon and outboard flap section did not seperate in flight (during the descent) they only could have seperated during a ditch-like event. Which means the plane must have glided for some distance after the steep descent. The damage (and hardly damaged leading edges) on the flaperon and outboard flap are indicating this clearly among the other fact that ~90% of the pieces found are trailing edge/surface control/wing related/engine cowling related pieces and including a nose gear door item with no compression damage.

    Still all of this seems to get ignored. Also by OI.
    The only realistic option left now IMO is @VictorI’s point around 30S based on the spotted debris west of the arc. If the plane is not found within +/-25Nm by then odds become very poor it will be found farther north and odds increase dramatically it must be outside of the +/-25Nm area.

    I hope I’m wrong but if not I would advise OI to turn back to the 32.30’S area and go searching east till around 97E while there is still time.

    Anyway I wish them goodluck and hope they find the plane as soon as possible.

  807. Nederland says:

    @DrB

    The reference for the cabin SMS/email app is also found in the FI, p. 52:

    “The SATCOM provided the Satellite link for the following functions: …
    Cabin Packet Data (Data-3) – Interface via the Panasonic System 3000i IFE equipment:
    – SMS/e-Mail
    – BITE-offload” [Built-in Test Equipment]

    and p. 49: “CPMU is Cabin Passenger Management Unit, which provides an interface between the Panasonic IFE and the SDU, for any Data-3 SMS/e-mail messages.”

    with p. 53:

    “The IFE equipment set up two ground connections over SATCOM (for the SMS e-mail application and Built-In Test Equipment (BITE) application) after the SATCOM re-established the link at 18:25 UTC, 07 March 2014 (normal), but not after the SATCOM re-established the link at 00:19 UTC, 08 March (abnormal). At no time during the flight was any user data sent over the link by means of the SMS/e-Mail application.”

    Presuming that the IFE was switched off from the cockpit at some point before 18:25 (presumably to prevent user data being sent), the lack of such a connection at 0:19 could indicate the aircraft had crashed within 90 seconds – the time it takes to set up the connection (p. 54-5)

  808. Ge Rijn says:

    @Nederland

    RE:(IFE)..the lack of such a connection at 0:19 could indicate the aircraft had crashed within 90 seconds – the time it takes to set up the connection (p. 54-5)”

    This is another assumption which is not based on evidence.
    Nobody still knows why the IFE did not log-on. Maybe the APU ran out of fuel before the IFE log-on could occure. If the APU started at all after second engine fuel exhaust. In @Andrew’s simulation the APU did not start.
    Maybe there was a brief engine re-light that stopt before the IFE log-on could be send. Maybe there was another reason. I think no-one can know for sure but still many consider this as additional proof the plane must have ended in a high speed impact close to the 7th arc. Excluding contradicting debris indications.

    I hope they are right but the plane still isn’t found close to the 7th arc.

  809. TBill says:

    @Victor
    hmm…well OI still seems to be in BR territory this morning.

    >It would be nice to get some assessment of how well OI was able to search the difficult terrain. Are they just doing a partial try and may have to go back? Or are they doing complete job?

    >Are they skipping the BR slopes (the blue BR section of the prior maps)? The new report from MY is different – they eliminated the blue BR zone that I was trying to sketch out.

  810. Victor Iannello says:

    @TBill: Before reaching Broken Ridge, OI had some level of confidence that they could “power through” the difficult terrain. I am not aware that their opinion has changed. I long ago stopped trying to understand the Malaysian maps. I view them as a qualitative reference.

  811. Andrew says:

    @HB

    RE: ”My questions are
    (1) whether RAT deployment could be prevented or whether power from the RAT can be isolated?
    (2) what is the endurance time of the main batteries? is the timing could match any observed re-log on?
    I presume the transponder will remain powered in that period also from the standby bus so long as the batteries are available.”

    ELMS controls the automatic deployment of the RAT when both XFR buses lose power. Deployment cannot be prevented by the pilot(s), nor can the RAT generator be isolated.

    The B777 Engineering Training Manual states: “The main battery can supply standby system power for at least five minutes.”

  812. Richard Godfrey says:

    SC continues to make good progress in searching the Diamantina Escarpment towards the Broken Ridge.

    https://www.dropbox.com/s/2dkbp8pnzqbq5u8/SC%20Track%2004042018.pdf?dl=0

    @TBill The yellow line is the 7th Arc calculated at an altitude of 20,000 feet, courtesy of Mike Exner and as agreed with Ocean Infinity.

    Ocean Infinity are searching a total width of 56 NM, 24 NM inside the 7th Arc and 32 NM outside the 7th Arc.

    I expect the search width to narrow as Ocean Infinity progress further northwards.

  813. DrB says:

    @David,

    OK. Now we know the L Transponder and the Dual Transponder Panel both get their power from the AC Standby bus, so shutting down the L IDG and both Backup Generators won’t cause the L Transponder to fail. It takes more failures of power supplies than that, so loss of power is extremely unlikely. Similarly, both the ADIRU and the SAARU feed it air data, so this possibility is also extremely unlikely. The only single-point failures I have identified are the L Transponder LRU itself or its antenna connection.

    I will modify 2.d in the next revision to remove the power loss option.

  814. DrB says:

    @Nederland,

    Thank you for supplying information on the possibility of SMS/email traffic from the cabin. I will modify the Observational Data list to remove the cabin phone call option and add the SMS/email options.

  815. DrB says:

    @Mick Gilbert,
    @Andrew,

    Your responses regarding the ACRAS Manager Mode options appear to be consistent with the “AIMS 2003” software.

    9M-MRO was delivered on 31 May 2002 as 777-2H6ER Serial Number 28420. That is prior to the 2003 release of modified AIMS software. Therefore 9M-MRO was delivered with the first, earlier version (i.e., “Not AIMS 2003”).

    How do you know that the later version was actually installed in 9M-MRO?

  816. TBill says:

    @Richard
    Good info (thank you!) on the yellow “centerline” location. If they narrow the search, I suggest they need +-20nm up to about 31.5S.

  817. Andrew says:

    @DrB

    RE: ”How do you know that the later version was actually installed in 9M-MRO?”

    FCOM Pt 2, page 5.40.51 has a description and diagram of the ACARS Manager page 2/2, as installed on 9M-MRO. The diagram clearly shows two check boxes, VHF ENABLE and SATCOM ENABLE. The text states that AUTO mode is enabled when both boxe are checked.

  818. Andrew says:

    @DrB

    As I noted in a previous comment:

    “My apologies, 9M-MRO was an AIMS-I aircraft. I believe it was an earlier block point version of the AIMS software that required at least one ACARS transmission pathway to be enabled. Later block point versions deleted the ‘Auto’ selection and allowed both VHF and SATCOM to be disabled. The AIMS software load on 9M-MRO must have been one of the later versions, given that the FCOM applicable to that aircraft states that both VHF and SATCOM can be disabled.”

  819. Richard Godfrey says:

    SC Update as there has been some fake news on Twitter.

    SC did wait outside the search area for 6 hours earlier today, as it is too early to collect the current cycle of AUVs.

    SC did not launch the ROV.

    SC is not heading to Fremantle.

    https://www.dropbox.com/s/gtdc6o3inympaub/SC%20Track%20040420182.pdf?dl=0

  820. sk999 says:

    Richard Godfrey,

    Do you know what BTO is being used, in conjunction with the 20,000 foot altitude, to compute the 7th arc?

    Thanks in advance.

  821. David says:

    @Victor. On the JW site @buyerninety has raised whether David Griffin completed the supplementary drift study he had agreed with you to do. If so it might be useful to have that on the record even if it would not affect searching plans directly.
    Any news on that please?

  822. DrB says:

    @Andrew,

    Thank you. I found the specific info for 9M-MRO as you said in the airline’s FCOM2 Page 5.40.51.

    I found one point interesting on that page. The FCOM says: “ACARS is set to auto mode (both boxes selected) at power-up or during a manual data communication system reset.” Thus, the pilots must have changed(at the gate) the default of both boxes being checked, and unchecked the “VHF ENABLE” box, leaving only SATCOM ENABLE. Perhaps this was SOP for flights into China.

  823. David says:

    Dr B.”I will modify 2.d in the next revision to remove the power loss option.”
    Thanks.
    I note in passing as you will have that had the right transponder been selected there would have been transponder failure at loss of transfer bus power.

  824. DrB says:

    Revision #7 of “Observational Data” is available HERE . I have made some changes to put things in chronological order and incorporate suggested improvements. Any errors are mine alone.

  825. Victor Iannello says:

    @David: Thanks for asking about that.

    David Griffin wrote me a nice email on Nov 29, 2017. In a nutshell, he thought my use of the KMZ files nicely captured the predicted drift behavior. At my request, he performed some drift calculations for points of impact (POIs) along the 7th arc between 29.5S to 30.5S. Around Mar 19, the line of points starts to stretch apart, as the video I produced shows. This occurs because of the fanning-out of the sea surface height contours. David believed that the spreading of the debris suggested by the location of the photographs could have resulted from a single origin (near 29.7 or 29.8S) on Mar 8. He produced a nice series of images for each day after the impact until Mar 29 to show this. If there is interest, I could ask him if I can share those links.

  826. Mick Gilbert says:

    @Victor Iannello

    Re: ‘He produced a nice series of images for each day after the impact until Mar 29 to show this. If there is interest, I could ask him if I can share those links.

    I’d be interested in seeing that work, thanks Victor, if Dr Griffin is amenable.

  827. David says:

    @ those interested in other recent underwater searches.

    The Australian submarine AE1 was lost with Australia at war in 1914 east of Papua New Guinea and was found a few months ago by Fugro. A short account of that search:

    https://www.aspistrategist.org.au/finding-men-ae1/?utm_medium=email&utm_campaign=Daily%20The%20Strategist&utm_content=Daily%20The%20Strategist+CID_8219d2fd5de590f5272b59dea801e847&utm_source=CampaignMonitor&utm_term=Finding%20the%20men%20of%20AE1

  828. David says:

    and the video in case you did not get that far.

    https://youtu.be/vfJMMwQ7J1A

  829. Peter Norton says:

    Richard Godfrey says:
    SC Update …

    It’s amazing how fast they are progressing. Seeing that demonstration in speed, it will be difficult for Malaysia not to extend the contract beyond the 90 days.

    During the previous search you got the impression that the Ocean is way too big to continue. But now it seems only like a question of money, not any more a question of what’s humanly possible – granted the financial resources, OI could even search the entire 7th arc (brute force attack) in a conceivable amount of time.

  830. HB says:

    @DrB,
    You may want to add no ELT signal captured by Satellites.

  831. airlandseaman says:

    sk999 says: “Do you know what BTO is being used, in conjunction with the 20,000 foot altitude, to compute the 7th arc?”

    We do not know for sure, but if they are searching symmetrically each side, it appears that they are using an arc about 4 nm further out than the 20,000 foot 18390 usec arc we recommended in January, based on a weighted combination of the corrected 00:19:29 and 00:19:37 BTO values. Boeing’s 2014 arc was 18440 usec, apparently due to the use of only the relatively noisy 00:19:29 600 b/s BTO value. ATSB suggested 18400 in several papers. 18390 looks like the best estimate based on the work Victor did in January.

  832. Andrew says:

    @DrB

    RE: ‘The FCOM says: “ACARS is set to auto mode (both boxes selected) at power-up or during a manual data communication system reset.” Thus, the pilots must have changed(at the gate) the default of both boxes being checked, and unchecked the “VHF ENABLE” box, leaving only SATCOM ENABLE. Perhaps this was SOP for flights into China.’

    The ACARS system has an airline-modifiable list of VHF datalink service providers (DSP) and frequencies, known as a ‘scan mask’. The scan mask prioritises the DSPs and the frequencies that will be used when ACARS is using VHF datalink, according to the airline’s contract(s) with the DSP(s) (eg SITA, ARINC, AVICOM) and the relative cost of using VHF vs SATCOM. If both VHF and SATCOM are enabled, the ACARS system scans the prioritised list of VHF frequencies first and tries to set up a VHF datalink. If the appropriate VHF frequency for the aircraft’s location isn’t listed in the scan mask, the system automatically switches to SATCOM.

    I can think of several possible reasons why MH370 did not use VHF datalink:

    1. VHF C, normally used for data communication, was tuned to another frequency (eg the company operations frequency) and was not subsequently returned to DATA mode.

    2. The DEFAULT RADIO MODE on the MFD VHF Manager page was selected to VOICE.

    3. VHF ENABLE was deselected on the MFD ACARS Manager page.

    4. The airline’s scan mask was not set up to use the SITA AIRCOM Far East/Asia frequency. That may have been the case if the VHF datalink costs in the Far East/Asia region were more expensive than SATCOM, as occurs in some other regions (eg Japan).

  833. Peter Norton says:

    > HB says:
    > @DrB, You may want to add no ELT signal captured by Satellites.

    @DrB:
    If you are going to add this, here is the lengthy discussion about ELTs streching across 3 blog posts:
    http://mh370.radiantphysics.com/2017/04/23/important-questions-about-most-likely-mh370-crash-site
    http://mh370.radiantphysics.com/2017/06/04/mh370-end-of-flight-with-banked-descent-and-no-pilot
    http://mh370.radiantphysics.com/2017/06/12/the-unredacted-inmarsat-satellite-data-for-mh370

    If you highlight all occurrences of “ELT” and scroll down the page, you find all the postings.

  834. Peter Norton says:

    > DrB says:
    > I found the specific info for 9M-MRO as you said in the airline’s FCOM2 Page 5.40.51. I found one
    > point interesting on that page. The FCOM says:
    “ACARS is set to auto mode (both boxes selected) at
    > power-up or during a manual data communication system reset.” Thus, the pilots must have
    > changed (at the gate) the default of both boxes being checked, and unchecked the “VHF ENABLE”
    > box, leaving only SATCOM ENABLE. Perhaps this was SOP for flights into China.

    Don Thompson said that the checkmark in the “VHF ENABLE” checkbox was removed at 15:54 UTC during pre-flight preparations:
    « The full satcom log records an ACARS Media Advisory message at 15:54 that shows ‘lost VHF’ and only SATCOM remaining available: LV155453S
    with subsequent Media Advisory messages showing ‘established SATCOM’ and, again, only SATCOM available, eg: ES155607S
    If VHF was available but not used, the trailing characters in these msgs would read ‘VS’ not simply ‘S’.»

    Andrew later confirmed this:
    « The ‘Lost VHF’ message that Don mentioned seems to indicate that the VHF ENABLE check box in the ACARS Manager was unchecked, so ACARS was unable to send downlink messages by VHF. »
    « The ‘Lost VHF’ message sent by SATCOM at 1554:53 UTC (when the aircraft was on the ground) seems to indicate that VHF had been de-selected in the ACARS Manager. ACARS transmission by VHF would not have been possible, even if the aircraft was within VHF range. »

  835. Andrew says:

    @Peter Norton

    Thanks, I’d forgotten that earlier discussion.

    @DrB

    Please ignore no. 4 in my list above.

  836. Peter Norton says:

    > Andrew says:
    > I can think of several possible reasons why MH370 did not use VHF datalink:
    > 1. VHF C, normally used for data communication, was tuned to another frequency (eg the company
    > operations frequency) and was not subsequently returned to DATA mode.
    > 2. The DEFAULT RADIO MODE on the MFD VHF Manager page was selected to VOICE.
    > 3. VHF ENABLE was deselected on the MFD ACARS Manager page.
    > 4. The airline’s scan mask was not set up to use the SITA AIRCOM Far East/Asia frequency. That may
    > have been the case if the VHF datalink costs in the Far East/Asia region were more expensive than
    > SATCOM, as occurs in some other regions (eg Japan).

    @Andrew:
    (1) + (2) I don’t know but would like to know, too …
    (3) We have previously discussed this. Don said VHF was deselected pre-flight (see above).
    (4) You previously stated the following:

    « Re: “Early on it has been suggested that deselecting VHF is normal practice on MAS flights to Beijing. MAS’s ACARS Service Provider SITA has no VHF receivers in China. Receiving ACARS messages through a third party (ARINC or a state-owned Chinese company) would incur roaming charges.”
    Such a procedure is not documented in the airline’s manuals. The Operations Manual states “Ensure VHF and SATCOM are both selected…during ACARS usage”. There are no exceptions mentioned for areas with high VHF data charges. Where I work, the airline programmable part of the ACARS software is set up so that ACARS automatically uses SATCOM in areas with high VHF data charges. There is no requirement to manually deselect VHF. »

  837. Peter Norton says:

    Andrew says: “Please ignore no. 4 in my list above.”

    Why ?

  838. Andrew says:

    @Peter Norton

    The scan mask I mentioned in point no. 4 is the ‘airline programmable part of the ACARS software’ in the earlier comment you quoted above. The ACARS system will only use those frequencies permitted by the scan mask. If the datalink provider’s ‘local’ frequency is not included in the scan mask, then the system will automatically use SATCOM if it is enabled.

    In MH370’s case, the VHF datalink was lost at 15:54, while the aircraft was on the ground. That suggests the VHF datalink was previously available. The local SITA frequency must therefore have been permitted by the scan mask.

  839. Peter Norton says:

    > Drb says:
    > (1) I am not aware that ACARS reports can be sent via HF/VHF radio.
    >
    > Peter Norton says:
    > They can. […] And it’s an important point for other reasons, I will come back to this another time.

    @DrB:
    Okay, I wanted to save this for later (when the flight ID discussion has ended), but since you are already right in the middle of it:
    When I said above, that the option to send ACARS via radio is an important point, it’s because this option was deselected pre-flight, which could potentially be a huge red flag.

    Since the missing flight ID may be also a red flag for the pilot, these two topics are somewhat related anyway.

    If it turns out that it was NOT MAS SOP to deselect VHF for ACARS and only occurred on MH370, I would consider this the biggest red flag of all – for the very simple reason that this was done on the ground, when everything was still hunky-dory.

    All other suspicious events took place in-flight after IGARI.
    (Or was there other suspicious stuff going on already beforehand?)
    And theoretically they could be part of a freak accident.

    But this is different.
    This was not an accident.
    It was either MAS SOP.
    Or it was the pilot working through his pre-flight-pijacking-to-do-list 47 minutes before takeoff.

    Unfortunately, so far we haven’t been able to decide this one way or the other. Opinions differed:

    Andrew: “There is no requirement to make selections on the ACARS Manager page during the pre-flight. It would certainly be unusual for VHF to be deselected – I’ll leave the ‘suspicious’ bit for you to decide.”

    Don Thompson: “Red flag? Possibly, possibly not, you might it fruitful to mine the web for hobbyist ACARS tracker logs with examples of MAS traffic over VHF (there are plenty of sites).”

    Gysbreght: “Early on it has been suggested* that deselecting VHF is normal practice on MAS flights to Beijing. MAS’s ACARS Service Provider SITA has no VHF receivers in China. Receiving ACARS messages through a third party (ARINC or a state-owned Chinese company) would incur roaming charges.”
    * suggested by whom ?

    Andrew: “@Gysbreght: Such a procedure is not documented in the airline’s manuals. The Operations Manual states “Ensure VHF and SATCOM are both selected…during ACARS usage”. There are no exceptions mentioned for areas with high VHF data charges. Where I work, the airline programmable part of the ACARS software is set up so that ACARS automatically uses SATCOM in areas with high VHF data charges. There is no requirement to manually deselect VHF.”

    Maybe we can get to the bottom of this ?

  840. Peter Norton says:

    I tried to find out, whether the same event (ACARS VHF deselection) also happened on MH371.
    (This would make the red flag disappear entirely, I guess?)

    I got this reply from Don Thompson back in October 2017:

    « The Media Advisory ACARS messages (label SA) transmitted during the MH371 sector show only SATCOM available, an example below:
    (soh)2.9M-MRO(nak)SA4(stx)S61AMH03710ES010740S(etx)(chksm)(del)(nul)(nul)
    the key portion of the string being: ES010740S
    Enabled SATCOM at 01:07:40 only SATCOM available.
    As I stated above, had VHF also been available then the trailing characters would indicate that condition (as ‘VS’)
    VHF coverage over China is good, with VDL Mode 2 RGS deployed, but not so good with plain old ACARS over Analogue VHF in many territories. The latter is slow, exacerbated by growing traffic. »

    What does this ACARS message mean:
    – VHF not available (i.e. solely a status message without any information as to the prior status) ?
    – VHF was available, but became unavailable (i.e. someone just unchecked the checkbox) ?

    What I am trying to get at:
    Does this ACARS message tell us whether the MH371 pilot also deselected the VHF checkbox pre-flight ?

  841. HB says:

    @Peter Norton

    If I may add to the previous discussion,
    * The FI has totally misrepresented the reliability of the ELT, the recent types as the ones installed on MH370 are reliable. They have averaged all incidents. The previous types were unreliable.
    * The fixed ELT was armed but the portable ELT was set to off according to the FI. The portable ELT could only have been triggered manually but according to the FI, pilots were not trained to trigger the ELT in case of a distress situation before impact.
    * This type of ELT can transmit a signal even if the antenna is broken.
    * Not sure where this 50s reference come from. if a satellite cover the area at that particular time, the signal should be picked up. Coverage is not 100% of the time.
    * The FI strangely enough did not make an explicit statement that no ELT signal was picked up
    * The ELT is designed to absorb shock up to 100g and triggers signal at 2.3 g. Judging the debris, the g-switch should have been triggered, 2.3 g is not a lot in terms of crash impact. 100 g is unlikely to be reached in crash impact, possible but unlikely, is requires a penetration from a sharp object. Mounting and lanyards unlikely to resist this kind of loading.
    * The ELT (fixed and portable) should be with a floater and water detection but not sure on MH370 what was the configuration, the previous discussion seems to indicate otherwise.
    * One uncertainty is whether that part of the plane where the fixed ELT is attached can float or not and whether the ELT will be entrained underwater, But if the ELT detaches, it should float (pending clarification on previous point).
    I would say on the new types of ELT, there is ~90% chance that it would have emitted a signal in that case. The chance varies depending on the actual satellite coverage at that given time/location.

  842. Peter Norton says:

    > DrB says:
    > the pilots must have changed (at the gate) the default of both boxes being checked, and unchecked
    > the “VHF ENABLE” box, leaving only SATCOM ENABLE. Perhaps this was SOP for flights into China.

    @Drb: You are right. But how did you know that this must have occured at the gate (as opposed to in-flight at the same moment when the SATCOM checkbox was deselected) given that you didn’t seem to know about the “Lost VHF” ACARS message at 15:54:53 UTC (when the aircraft was on the ground) ?

  843. Peter Norton says:

    > HB says:
    • The ELT (fixed and portable) should be with a floater and water detection but not sure on MH370 what was the configuration, the previous discussion seems to indicate otherwise.
    • One uncertainty is whether that part of the plane where the fixed ELT is attached can float or not and whether the ELT will be entrained underwater, But if the ELT detaches, it should float (pending clarification on previous point).

    @HB:
    I think I was able to show that 9M-MRO’s portable ELT is equipped with a floater:
    http://mh370.radiantphysics.com/2017/06/04/mh370-end-of-flight-with-banked-descent-and-no-pilot/#comment-3986

  844. flatpack says:

    @Peter Norton

    “Or it was the pilot working through his pre-flight-pijacking-to-do-list 47 minutes before takeoff.”

    Would 47 minutes be as soon as the pilot entered the cockpit? (plus any power-up time)

    Presumably there would be timestamped airport video of the crew boarding.

    Wasn’t there an hour long phone call made?

  845. Andrew says:

    @Peter Norton

    RE: “What does this ACARS message mean:
    – VHF not available (i.e. solely a status message without any information as to the prior status) ?
    – VHF was available, but became unavailable (i.e. someone just unchecked the checkbox) ?”

    The message shows that only SATCOM was available. As per Don’s previous comment, if VHF was available and then became unavailable, then there should have been a media advisory message LV{timestamp}S to indicate VHF was ‘lost’, as there was with MH370. As I understand it, there was no such message, which suggests that VHF was not available rather than ‘lost’.

  846. Peter Norton says:

    @Andrew:
    You conclude from the missing “lost VHF” message that on MH371 “VHF was not available rather than ‘lost’”.

    Ok so:
    On MH370, VHF was lost pre-flight.
    On MH371, VHF was not available (rather than “lost”).

    But if “ACARS is set to auto mode (both boxes selected) at power-up or during a manual data communication system reset” (FCOM), MH371 would have had both boxes selected initially.
    If subsequently VHF was “not lost”, why would it then be “unavailable” ?

  847. Andrew says:

    @Peter Norton

    RE: “But if “ACARS is set to auto mode (both boxes selected) at power-up or during a manual data communication system reset” (FCOM), MH371 would have had both boxes selected initially.”

    Not necessarily. First, if the aircraft was only on the ground for a short time between arriving and departing Beijing, then it would have remained powered, so no ‘power-up’ would have occurred. Second, MAS procedures had no requirement to perform a data communication system reset before a flight. If the VHF was deselected during the flight up to Beijing, then it would have remained deselected for the flight back to KUL.

  848. HB says:

    @Peter Norton,
    Thanks for clarifications. This makes sense.
    The log from the satellites could be an evidence. I am not sure whether that log was screened based on location/time or not. It could have been missed on that basis but i am not sure how it works after a valid signal has been received and which authority is responsible to handle it. There is absolutely no reference to this in the FI.

  849. HB says:

    @Peter Norton,
    The fate of the fixed ELT is the most important as it was armed. The portable one was off according to the FI as it was used as a portable one. The fixtures of the fixed certainly cannot cope with 2.3 g. The fixed one is identical to the portable one ie should be designed to float and trigger a signal at 2.3g or on contact with water. I am not aware about The 50 s delay, not sure where that comes from. If armed, there should not be any delay and signal reception depends on the presence of satellite at that time. Happy to see the reference for this delay if you have.

  850. Mick Gilbert says:

    @HB
    @Peter Norton

    Re: ‘The fixed one is identical to the portable one ie should be designed to float …’

    They’re not identical. The portable ELT was an ELTA Model ADT 406 AP PN: 01N65910. It is essentially the same as the fixed ELTA Model ADT 406 AF/AP PN: 01N65900 but with one notable difference; the 01N65910 is fitted with a float whereas the 01N65900 is not.

  851. Mick Gilbert says:

    @HB

    And just by the bye, neither the ELTA Model ADT 406 AP PN: 01N65910 or the ELTA Model ADT 406 AF/AP PN: 01N65900 are triggered by contact with water.

  852. Paul Smithson says:

    Dr B.”I will modify 2.d in the next revision to remove the power loss option.”
    Is this because there is so much redundancy that you believe that power loss per se is not possible?

    Is it not conceivable that an electrical fire could cause so much disruption/alarms/tripping that pilots deliberately shut down/isolate multiple systems?

  853. TBill says:

    @ALSM
    @Richard
    @sk999
    “We do not know for sure, but if they are searching symmetrically each side,..”

    I thought Richard indicated the search to date was -25nm/+32nm (not symmetric).

    I like Richard’s yellow line, but if that is not what OI is using, I’d be a little concerned the inside the arc is getting short-changed, especially if OI narrows to +-20

  854. HB says:

    @Mick, correct, FI only mentions the slide raft ones are activated on contact with water. The fixed one should have been activated on impact. The chance of reaching a satellite would depend on the satelite presence and whether the device was submerged or shiielded at the time of transmission. A delay of 50 s was mentioned in previous discussion but i cant find reference to that. It also appears that only the AMSA MCC was consulted according to the FI. No evidence of which time frame was checked nor that other MCCs were consulted. The FI is also silent on the detection, it instead describes wrongly it’s reliability averaging events with old types.

  855. Mick Gilbert says:

    @HB

    There was a pretty detailed discussion on ELTs in this forum quite some time back – you might want to see if you can find it.

    The problem with fixed ELTs and high energy impacts relates to the antenna; it’s not integral to the unit. On the B777 there’s a couple of metres of cabling running from the ELT itself to the external antenna; if the cabling is severed in the impact the ELT isn’t going to get a signal out. The other problem with a water impact relates to immersion. Even if the ELT remains connected to its antenna if the antenna is submerged then getting a signal out becomes problematic.

  856. Ge Rijn says:

    @All

    With all respect it’s not about ELT’s being triggered or not anymore.
    In a ditch ELT’s possibly won’t reach the 2.3G load necessary to trigger them. There is an example of a crash-landing (on land) where the ELT’s did not activate due to not reached G-loads (posted the example long time ago but cannot find it anymore). Did the Hudson-ditch trigger the ELT’s?
    I never read they did.

    It’s now about the search width OI is taking along Broken Ridge.
    They’ve been within ~30km of the 32.26’/97E region.
    If they skipp this hole it could turn out to be very costly IMO.
    But lets move on asif this possibility does not exists..

  857. Peter Norton says:

    > Mick Gilbert says:
    > @HB There was a pretty detailed discussion on ELTs in this forum quite some time back – you might
    > want to see if you can find it.

    it’s here

  858. Peter Norton says:

    In my view all someone-hid-the-plane theories must explain 2 things:
    – the missing ELT signal (e.g. unreliability)
    – how the perpetrator made 100% sure ELT activation does not occur

    Historically, ELTs activated in 22,5% of accidents
    This percentage relates to crashes both on land and on water. A study of ocean crashes alone would perhaps be more meaningful. However, the main element of differentiation, namely the antenna getting submerged by water, was only once found to be the reason for ELT failure in a study of 173 accidents over the past 30 years.

    If the goal was to hide the plane, the perpetrator would not have gone to such great lengths in planning a highly sophisticated vanishing trick that includes disappearing at FIR boundaries, depowering busses, etc. and then take any chances of ELT activation at the crash site thereby blowing his entire sophisticated plan, well, out of the water …

    For these reasons,
    I would recommend adding the ELTs to DrB’s fact sheet.

  859. Nederland says:

    @Peter Norton

    The ICAO advised that the ELTs can be switched off manually from the cockpit (although tape may be needed to fasten the springload switch to the reset position).

    “All onboard surveillance and surveillance sources e.g. ADS-B
    /X’PDR/FMS can be turn off inflight by switches or via circuit breakers. The aircraft Emergency Locater Transmitter that is installed on all B777 (a recent mandate) failed to provide a signal. Apparently this can also be deactivated if the switch is left in ‘reset’ position by the crew.” (p. 3)

    https://www.icao.int/APAC/Meetings/2015%20APSARTF3/WP05%20ICAO%20Brief%20on%20the%20SAR%20Response%20to%20MH370.pdf

  860. Ge Rijn says:

    @Peter Norton

    Exactly. The perpetrator going all this length to avoid detection would certainly do everthing to avoid ELT’s being triggered.
    A soft as possible ditch is the best way to avoid this.

  861. Peter Norton says:

    > Mick Gilbert says:
    > The problem with fixed ELTs […]
    > if the [antenna] cabling is severed in the impact, the ELT isn’t going to get a signal out.

    HB claims (above) that “This type of ELT can transmit a signal even if the antenna is broken.”
    I have no idea what this claim is based on. Maybe HB can clarify ?

    > The other problem with a water impact relates to immersion. Even if the ELT remains connected to
    > its antenna if the antenna is submerged then getting a signal out becomes problematic.

    The study referenced in my previous posting determined this to be the reason for ELT failure in
    only 1 out of 173 accidents (0,6%) over the past 30 years. Three possible interpretations:

    • The 173 accidents include a very low number of crashes on water (I was unable to find the ICAO study, otherwise I would have verified that).

    • 34% of ELT failures occurred for “unknown reasons”. Maybe lots of submersion cases hide in this category?

    • Or else, for whatever reason, failure due to submersion simply might not be a scenario occurring very often in real life.

  862. HB says:

    @Peter Norton, Mick
    I think the historical reliability data is misleading as this type is reliable compared to previous types. There are just some circumstantial factors that can affect the signal. It wont be this low number today. The FI has misrepresented this and on that basis did not present further evidence eg MCC logs etc. The signal frequency has increased also. I would give it a good 80 to 90 pc success rate overall for an accidental scenario.
    The FI has also not explicitely mentioned there was not detection from all the relevant MCCs in the relevant time frame. Not a big issue but it would be good to see this in the final report.
    For non accidental scenarios, the ELT may be in off position of course. Not sure if the ELT in armed mode send any signal to the satellite system that can be traced. Does it leave any trace? Any way to know the status of the ELT?

  863. Peter Norton says:

    > Nederland says:
    > The ICAO advised that the ELTs can be switched off manually from the cockpit
    > (although tape may be needed to fasten the springload switch to the reset position).

    Thank you for the link. Yes, this was also discussed in the ELT discussion I referenced above.
    As an alternative to your tape, chewing gum or glue were the previously proposed methods.

    I’m not saying at all that this isn’t doable. My point is that it the missing ELT signal should be added to DrB’s fact sheet, because the someone-hid-the-plane theories need to explain why Mr. Someone was 100% sure the ELTs would not activate. Simply saying “soft ditching” is not going to cut it, as in keeping with a sophisticated plan precautionary measure must have been in place in case the ditching fails (quite likely in rough seas) and triggers an ELT signal revealing the exact crash location.

  864. Rob says:

    @Peter Norton
    @Ge Rijn

    “In my view all someone-hid-the-plane theories must explain 2 things:
    – the missing ELT signal (e.g. unreliability)
    – how the perpetrator made 100% sure ELT activation does not occur”

    Notwithstanding the reliability question, and I think it’s safe to assume that no ELT signals were picked up?

    Ge Rijn advocates a soft ditch, but a hard ditch would also do the job if the aircraft sank within 50secs of impact. Last time ELTs were discussed in detail, I think it emerged that the portable ELTs have a limited range. If correct, then to someone intent on hiding the plane, it would be considered desirable to end up in as remote an area as possible.

    Relying on taping the cockpit switch in the reset position would be risky. A planned soft ditch could turn into a harder one if the pilot misjudged it, and or a wing caught the swell, and hard enough to both free the switch and injury the pilot in the process.

  865. Peter Norton says:

    > HB says:
    > I am not aware about The 50 s delay, not sure where that comes from. If armed, there should not be
    > any delay and signal reception depends on the presence of satellite at that time. Happy to see the
    > reference for this delay if you have.

    The only sources I have for the 50 seconds is a news article and Brian Anderson’s posting:

    « The two 406 ELTs on MH370 are designed to trigger upon a deceleration of about 2.5G or more. However, the data transmission protocol delays the first transmission for approximately 50 seconds after triggering, and then continues to transmit the data block of 520mS at 50 second intervals. If the ELT antenna is underwater before the 50 second delay elapses then no signal will be radiated. Repeating the data block every 50 secs is necessary to allow for the possibility of one of the monitoring satellites to be passing [roughly] overhead. A larger delay may be experienced before this occurs. In the case of a crash into the sea, even if the ELT does trigger, the likelyhood of a successful transmission being received is close to zero. »

    In response to this I asked:

    >> the data transmission protocol delays the first transmission
    >> for approximately 50 seconds after triggering.
    >> If the ELT antenna is underwater before the 50 second delay
    >> elapses then no signal will be radiated.
    >
    > Then why wait 50 seconds ?

    Unfortunately the reason for delaying the transmission by 50 seconds during which the antenna can become inoperable was never explained.

    @HB: Could you please in return source your claim that
    “This type of ELT can transmit a signal even if the antenna is broken.”

  866. DennisW says:

    @Ge Rijn

    The perpetrator going all this length to avoid detection would certainly do everthing to avoid ELT’s being triggered.

    So flying a 777 within cell phone distance of Penang was going to great length” to avoid detection? Do you have any idea how silly your avoiding detection assumption is? The evidence points to an FMT followed by a autopilot path into the SIO ending with a spiral dive close to the 7th arc. There were at least 400 pieces of debris created when 9M-MRO hit the ocean, and that number does not include the probability of a beached piece of debris not being found or the decay of debris (waterlogged and sinking) over time. Most likely there were thousands of pieces of debris.

  867. Peter Norton says:

    > HB says:
    > The fixed one [and] the portable one should […] trigger […] on contact with water.

    They don’t, but I agree that they should.
    I don’t see any reason for fixed+portable ELTs to lack a water sensor,
    but good reasons for adding water sensors to both of them:

    • They are not expensive.
    • It would be helpful if fixed+portable ELTs activated automatically upon contact with water.
    • This function is outsourced to the life raft ELTs, but that’s an unsatisfactory substitution for 2 reasons:
    (a) A lack of G-force triggering is not equivalent to life rafts being deployed and their ELTs activated.
    (b) Life raft ELTs don’t transmit on satellite-monitored 406 MHz (contrary to fixed/portable ELT).

  868. Victor Iannello says:

    @All: I have received permission from David Griffin to share his email that discusses modeling of the drift of the objects spotted by the RNZAF on March 29, 2014. I have appended that email message to the bottom of the prior post that discusses these objects.

  869. Peter Norton says:

    > HB says:
    > The fixed ELT was armed but the portable ELT was set to off according to the FI.
    > The portable ELT could only have been triggered manually but according to the FI,
    > pilots were not trained to trigger the ELT in case of a distress situation before impact.

    To be precise, the FI report doesn’t categorically state that the portable ELT was set to OFF on MH370,
    but that “It is normally in the OFF position” (p.32).

    If indeed the portable ELT was in OFF position and not in ARMED position, I find this highly absurd.
    Especially with regards to “the problem with fixed ELTs and high energy impacts relat[ing] to the antenna; it’s not integral to the unit. On the B777 there’s a couple of metres of cabling running from the ELT itself to the external antenna; if the cabling is severed in the impact the ELT isn’t going to get a signal out.”

    • The fixed ELT does not float. The portable ELT floats.
    • The fixed ELT suffers from a high risk of severed antenna cables. The portable ELT doesn’t have this problem.

    Yet the portable ELT was turned off ?
    Is this even allowed by regulations ??

  870. Ge Rijn says:

    @DennisW

    IMO we’re talking about a max of 400 initial pieces comparable with pieces found able to beach within ~4 years. I think it’s on the high end still.
    ~30 pieces have been found. The rest is probably still drifting in the ocean and some buried on beaches along African shores and islands.

    If you would count all the possible pieces that were possibly released on impact you might as well count the 100.000’s of mangosteens.

    Ofcourse there were much more (small) pieces of debris after impact. But they sunk soon after or were undetectable/indentifiable by any means.

    I consider your Weibull-analysis is based on likely debris. Not on all possible initial (small) debris which we don’t know and therefore has no meaning in the analysis imo.

  871. Peter Norton says:

    > HB says:
    > @Peter Norton, Mick: Not sure if the ELT in armed mode send any signal to the satellite system
    > that can be traced. Does it leave any trace? Any way to know the status of the ELT?

    I always assumed that unless the ELTs are activated (manually or by the sensor), they remain completely passive and don’t transmit/communicate. I doubt that there are “ELT pings” analogous to the SDU pings, if that was the question. I don’t know of any trace. Can someone confirm?

  872. Peter Norton says:

    > flatpack says:
    > @Peter Norton
    > “Or it was the pilot working through his pre-flight-pijacking-to-do-list 47 minutes before takeoff.”
    > Would 47 minutes be as soon as the pilot entered the cockpit? (plus any power-up time)
    > Presumably there would be timestamped airport video of the crew boarding.
    > Wasn’t there an hour long phone call made?

    I don’t know about this call nor when the pilot entered the cockpit or why the timing of this would be of importance.

    Can someone better handle these question directed at me ?

  873. Ge Rijn says:

    @VictorI

    Very nice from Griffin. But again nothing about the ‘Blue Panel’ and associated spotted debris field at 32.4S/97.8E on 28-3-2014 by RNZAF.

    But again; I truly hope you are right.