Status of current underwater search. (Click on image to enlarge.)
Recent Activity
After a short stop in Fremantle for to re-fuel, re-supply, and change crew, Seabed Constructor, operated by Ocean Infinity, is back searching for the wreckage of MH370. Ocean Infinity is under contract with Malaysia to use its team of eight autonomous underwater vehicles (AUVs) to scan the seabed in the Southern Indian Ocean (SIO) in search of wreckage from the aircraft. Under the terms of the contract, Ocean Infinity will only be paid if the wreckage is found. The search is occurring in multiple six-week long “swings”, of which the first swing has been completed, and the second swing is just beginning. Subsequent swings will also require a stop in Fremantle for servicing.
For the first swing, Ocean Infinity began by searching the 25,000 sq km of seabed that ATSB and CSIRO have designated as the priority area (shown in white in the figure). So far, Constructor has scanned about 7,500 sq km of seabed, including 5,000 sq km within the priority area that was designated by CSIRO as the “primary area” (solid white). There remains about 20,000 sq km of the priority area that is unscanned (translucent white). Beyond that is the extended search area, which reaches to about 29S latitude along the 7th arc (translucent green), and is expected to be searched at a width of +/- 25 NM from the 7th arc. Under ideal conditions, the eight AUVs are capable of scanning about 1,200 sq km of seabed each day. Recognizing the possibility of weather and operational constraints, a more realistic expectation might be about 25,000 sq km per swing. However, until Ocean Infinity gains more operational experience, it is difficult to predict what scan rates are realistically achievable.
With the sparse and imprecise evidence we have, it is impossible to assign a high level of certainty to any impact site, as the satellite data and the drift models allow a broad range of possibilities. So, it becomes a numbers game–the more area searched, the higher probability of finding the wreckage. However, within that broad range, there are some “warm spots” that are based on assumptions about navigation inputs and other evidence.
What We Know So Far
In the previous post, I estimated the probability of finding the wreckage as 67%, assuming all of the priority and extended areas are scanned. (This probability will vary some depending on how far north the search reaches.) Considering that only 5,000 sq km of that area were scanned in the first swing, and assuming that there are equal probabilities within that total area, the probability of finding the debris field within the primary area would be about 4.4%. Considering this low percentage, it should come as no surprise that the wreckage has not yet been found, and we are far from the point of re-thinking the search strategy.
Within the area searched so far, there are three warm spots that CSIRO has designated as priorities, based on satellite images of objects that could have been MH370 debris, and from drift models that estimated the points of impact from the location of these objects. Last August, the highest priority location (CSIRO Priority 1) was described by CSIRO’s David Griffin in these words: We think it is possible to identify a most likely location of the aircraft, with unprecedented precision and certainty. Unfortunately, all three of these locations have now been scanned with negative results. Unless positive news is being withheld, the confidence expressed by CSIRO was unfounded. This is not a total surprise: The objects captured by the satellite images had too much surface area to likely be from MH370, and the location of the potential impact sites were not consistent with the high speed descent suggested by the final BFO values.
Two other warm spots have been at least partially searched in the first swing. The first is an impact location near 34.7S latitude that Inmarsat derived by minimizing the BFO error. More recently, Bobby Ulich proposed a location near 34.8S latitude that was based on a path of constant true heading (CTH). We should know soon whether or not these warm spots are completely eliminated.
Another warm spot that should be searched during the current swing is based on a great circle path between waypoints BEDAX and the South Pole. I first proposed this path in August 2014, and I still consider it to be among the best possibilities because of the excellent fit of the BTO and BFO data, and because of the simplicity of navigating in the direction of true south. That said, despite the attractiveness of this scenario, we don’t know whether the aircraft was navigated in this manner, so it remains one of many other possibilities.
As shown in the figure above, there are warm spots that reach as far north as 27S latitude that are based on certain navigational inputs. Although the match to the BFO data is not as good for paths ending that far north, the BFO error is still well within what was recorded for previous flights of the 9M-MRO airframe. The drift models also favor an impact point further south than 27S. However, for debris discovered on the beaches of Eastern Africa, there could have been a considerable delay between the time of discovery and the time of arrival near the shore, and this uncertainty reduces the accuracy of the drift models.
In a nutshell, although the previous search swing has eliminated some possibilities, we are still very early in the search process, and it is much too early to draw any conclusions.
Unknown Activities of Seabed Constructor
The figure below from Richard Cole shows the recent behavior of Seabed Constructor. At the end of the last swing, Constructor returned to the outer leg of the primary search area, which had been previously scanned. After following the pattern of a 5-km circle, it retraced what we believe was part of a previous path of an AUV, and then disabled its AIS data, which made it impossible to remotely track. When the AIS was eventually re-enabled three days later, Constructor had left the search area, and was traveling back to Fremantle. What activities occurred during these three days is not known.
Seabed Constructor’s path, as adapted from the work of Richard Cole. (Click on image to enlarge.)
At the start of the search for the current swing, Constructor again returned to the southern end of the outer leg of the primary search area, and seems to be actively searching the seabed in this location. The activities in the current area are likely related to activities that occurred when the AIS was disabled during the last swing.
Some possibilities that have been proposed by others to explain the behavior are:
- Constructor is re-scanning areas that had poor quality or missing data either because of malfunctioning equipment or challenging terrain
- There are one or more promising points of interest that are under being comprehensively investigated
- A search is underway to locate equipment that was lost in the previous swing
- Some combination of the previous possibilities
Whatever the reason for the unexplained behavior, it is noteworthy that there was no reference to the behavior in either of the last two weekly updates from Malaysia. As Malaysia has two observers on Seabed Constructor, Malaysia is certainly aware of the surrounding circumstances. Malaysia’s decision to omit pertinent information in the weekly reports further erodes the public’s confidence in the Malaysian-led investigation. Credibility is not possible without transparency.
Is much credence given would anyone know please to the debris drift analysis from Eric Jansen, reported variously and @ http://www.auntypru.com/forum/showthread.php?tid=24&page=27 that includes comments: –
“The new drift modelling shows that pieces of aircraft debris discovered in Réunion, Mozambique, South Africa and Rodrigues Island most likely originated from an area further north than the search area, between the latitudes of 28 degrees and 35 degrees south.
“Our simulation shows that the debris could also have originated up to around 500km further to the north. If nothing is found in the current search area, it may be worth extending the search in this direction,” said Eric Jansen, a researcher at the euro-Mediterranean Center on Climate Change in Italy and lead-author of the study.
“Our result is the first to calculate the movement of the debris that best agrees with all five of the currently confirmed discoveries. This should make it the most accurate prediction.””
@David
Regarding your comment in the previous thread: “I now throw in another possibility, that a rational, deeply despairing pilot might treat flying this way as a final challenge, “damn them all” being his frame of mind.”
I’m not qualified to comment on the psych aspects, but I believe the scenario you described has significant risk with little benefit. It would certainly make the aircraft go ‘dark’, but much of that can be achieved from the cockpit without having to select off all the generators. If this episode was meticulously planned to make the aircraft disappear, why risk losing control for the sake of de-powering a few extra lights in the cabin, when the chances of being spotted (let alone identified) by another aircraft were extremely low?
@formula: Welcome to our discussion.
The search zone recommended in that study is generally consistent with the current search area, as well as many other drift models. The previous search area is now generally recognized as too far south. Of course, we have more debris now than we did at the start of the underwater search, so drift model results are better.
@Andrew. Thank you once more. “…why risk losing control for the sake of de-powering a few extra lights in the cabin…” Yes, doesn’t sound optimal at all and not what should be expected.
I persist with the scenario as a possibility because it is the only one I know of which leads to a plausible explanation for the later SDU repowering: regaining autopilot.
@Andrew. I forgot to comment on your comment, “If this episode was meticulously planned….”
I did not have that in mind, more a spontaneous sort of anger.
@Victor Thank you for the post. I am curious about several things with the current search (as are all I’m sure). But a couple items in your post stand out to me.
First, I am curious to know folks’ thoughts on when the 90 contract days may be up? And, further, how much Ocean Infinity can really scan before bad weather drives them away? Of note, currently, it is believed SC is holding due to waves, since the AUVs cannot be safely recovered in current wave height circumstances. This situation will occur more and more as we move into March.
Second, while it is well understood that whatever “activities occurred during these three days [the AIS was off] is not known” by all here, I would posit there are likely several front and center participants in the non-governmental search community, including frequent commentators to this blog, who do know. I could be wrong, of course. But I think I am not totally wrong. And it should not go unsaid that not all of the secrecy surrounding this tragedy can be laid solely at the feet of the Malaysians. This wasn’t the import of your post at all, but I wanted to mention it as this is a common theme from most (or all) of us regarding the Malaysian government.
Let’s hope SC can get motoring soon and much more can be searched in the time allotted. Thanks again!
@ErikN: I hope that the 90-day term of the contract, if required, will be extended by mutual consent of Malaysia and Ocean Infinity. As for knowledge of Constructor’s activities during the AIS blackout, I’ll let others speak for themselves. Ultimately, it is the responsibility of Malaysia to provide that information to the public.
@Victor I also hope they will extend it. And I hope it’s 90 days under some definition (90 in which AUVs are deployed at least 12 hours (spit-balling?), or something). I agree it is Malaysia’s responsibility to provide information to the public. My intention was not to absolve them. My comment was simply to say it is not known by you or me does not mean it is not known outside the official investigation. Perhaps it is really neither here nor there, but my intuition says it is worth noting. Either way, this search is every bit as exciting and patience-testing as expected. Let’s hope some answers to the tragedy are forthcoming. Prayers for all.
comparison of search areas:
http://imgur.com/kUzCwVp.jpg
(PS: I could not participate in a while for health reasons)
@David
Many signs of the flight show well calculated intend with the purpose of limiting risk of detection/interception as best as possible.
I see no signs of reckless behavior by a pilot with a ‘damm it all’ state of mind. Flying the plane in the attitude you propose in your scenario-proposal would be quite reckless and unnecessary.
Suicides are often carefully planned. The rational behind it might seem/be very irrational often (to others) but the execution of the plan is often planned in very rational detail. And the plan is also often hidden from even the most closely related, professionals included.
Taking them by suprise when it happens at last.
I think many people know examples of this.
Many years ago a good friend of mine back then told me the way he planned to die when he would decide the time was right somewhere in the future. He was sure back then (which he repeated later several times) he was going to commit suicide somewhere in the future.
But it would be this way and nothing else:
He would sail away on his boat into the ocean without telling anyone. Not even his wife. Then somewhere at see he would sink his ship and go down with it. Making sure no fuel was left and nothing surfaced floating.
So no one would find him and everyone would keep wondering he had an accident or what happened.
He had planned it all in detail back then. Now some twenty years later he’s still alive and he still has his sailing boat. We never talked about it anymore and lost contact.
But I won’t be suprised he will be missing at see some day.
@Ge Rijn. Thank you. So this, “I’m mad as hell” scenario is unsupported.
It is incongruous that a suicide would take care not to be “reckless” just like some being totally callous about others. Your anecdote is illuminating thanks yet not entirely apt since in that instance he did not go ahead, like many contemplating euthanasia once having obtained the means I gather.
Also I note you say, “Suicides are often carefully planned”, the “often” leaving some room for doubt; but barring other input I will let this rest there.
@ Victor – many thanks for your reply. I hope SC is getting close.
@David
Yes ‘often’ means in this sence there are also many ofcourse who commit suicide in an impuls without a thoughtfull plan.
But than the situation and/or state of mind before the act is mostly clear to others and the act more or less understandable right away.
I think your proposal was well worth considering.
Deducing is also reducing isn’t it. Looking at every angle.
@Shadynuk and @Don. Thank you so much for your replies about debris and drift.
On psychology I found something else the may be worth considering.
It’s on (sea)burial of a Muslim.
First; it’s prohibited to cremate a Muslim.
(Muslims believe that the body still ‘feels’ for a while after death).
Second; when a sea burial can not be avoided the body has to be wrapped in sheets and a weight attached to the feet to make sure the body reaches the bottom of the sea (the earth).
If Shah was the culprit this maybe could have significance:
First; It could be his motivation to end the plane without fuel so risk of burning his body would be eliminated.
Second; It could be his motivation to end the plane in one piece as much as possible to make sure his body reached the bottom of the ocean.
Link: https://www.greenprophet.com/2011/05/muslims-buried-sea/
This makes interesting reading @Don. Thank you. Julia
https://www.flightglobal.com/news/articles/crashed-an-148-pitot-static-sensors-left-unheated-445881/
Hello. Interesting comments about muslim burial at sea. I have also been considering the flight from a potential islam perspective.
Is it possible the flight to the north east was pointed at Mecca to coincide with a particular prayer schedule (not sure how this is calculated inflight)…..after the prayer event the flight headed south.
The aircraft could also have been turned to face Mecca for on other occasions.. eg.morning prayers before returning to a course towards the 7th arc? (read somewhere about “lost time”.
Also on loss of the first engine due to fuel is it possible the aircraft once again was pointed towards Mecca before its final descent.
Regards
Chris
@formula
Thanks for referring to this Italian study again.
Like to mention clear indications on a crash north of 36S were allready made by me before july 15 2016 based on @MPat’s and CSIRO/Griffin’s drift models back then.
The discussion on lack of debris on the WA shores was allready going on for some time before that on JW’s blog.
Not to allude to myself but to indicate the ATSB never considered this information nor from others after that to adjust their search plan.
Extracted from Jeff’s article back then (with thanks to Jeff Wise):
“Indeed, after the report came out other pieces of debris were found, and drift modeling of these pieces be used to refine the search area. Indeed, after I published last week’s guest post by MPat, reader Ge Rijn pointed out:
Over those 20 years in MPat’s model only 7 out of 177 buoys landed in Australia. Those 7 all passed the search box under 36S… [this] points clearly to the trend the more south you go under ~36S the more likely it becomes buoys (debris) will land on Australia and the more north you go above 36S the less likely it becomes buoys~(debris) will land on Australia. This is also because the more south you go under ~36 the currents tend to go further east and the more north you go around 36S the currents tend to bend stronger to the north avoiding Australia. And this is exacly what the facts about found debris shows us till now.
Note that 36 degrees south is just shy of the northern end of the current search area; as Ge Rijn observes, historical drift data suggests that if the plane had crashed south of this latitude, debris should have been found in Australia, which it obviously hasn’t.”
Jeff’s article:
http://jeffwise.net/2016/07/15/how-we-know-where-mh370-went/
Like to add that Griffin only stated recently that he had his ‘light bulb’ moment at the end of 2016 when he realized the crash area had to be more north because of the lack of debris finds in Australia.
@Ge Rijn
“Like to add that Griffin only stated recently that he had his ‘light bulb’ moment at the end of 2016 when he realized the crash area had to be more north because of the lack of debris finds in Australia.”
Funny how some people get lightbulb moments only after their feet are held to the fire.
If you were now to meet up with David Griffin and ask him if the lightbulb was still shining as brightly, his reply might be a tad unsettling. OI are still looking for the plane, and will continue to do so, fruitlessly, until someone gets another lightbulb moment.
The only reason they are still searching is because the ATSB/SSWG made the policy decision to rule out a piloted descent. It was a convenient ploy, intended to limit the originsl search area southern boundary. It’s backfired on them. The ATSB told Griffin and others (OI included) that retracted flaps spelt uncontrolled descent, spelt impact close to the 7th arc, which in turn had given birth to the new 25,000sq km search area. A compounding of one error upon another is now drawing out the search. There is still time to turn things around.
Money however, can be a powerful incentive. OI will need to search the area around S39.20 E88.36, if they are to claim their reward.
@Rob
Yes, it’s bit odd Griffin stated he had this ‘light bulb’ moment at the end of 2016. While the discussion about the lack of debris in WA started allready in late 2015 early 2016 and I made my observation in july 2015.
Based on MPat’s study but also on Griffin’s study from that time that showed only 12 historical buoys landed in Australia that also only past south of 36S.
I don’t know. Maybe he never read JW’s blog.
On your: “the ATSB/SSWG made the policy decision to rule out a piloted descent” I’m not so sure.
They initially took a search-width of ~45Nm to east and ~30Nm to the west between ~39S and ~35S.
It seems to me they at least took some kind of glide into consideration.
In this regard OI shows a more conservative search plan with +/-25Nm.
@Rob
I feel the current search area covers intentional pilot paths.
In particular, what I now see is 180S path and then a change to 135S heading (at 31S) to go southeast right over the top of Broken Ridge to around 32.5S end point.
PS-
Here is a approx. SkyVector path:
0894E 3194S 3295S
and envision that last southeast leg is right overtop of BR
Ge Rijn
“It seems to me they at least took some kind of glide into consideration.”
Yes, what they did was to take it “unofficially” into consideration. That was principally down to Simon Hardy and myself – we were both petitioning for the extended, piloted glide. The ATSB were constrained (straight-jacketed, if you like) by the Tripartite Ministers 120,000sq km. Primary search limit. To get round it some way, they took back some unused search area from the inside edge and added it to the outside edge, which is how the search area became asymmetric about the 7th arc. But whe they ran out of the pre-agreed quota of area, they had to call the search off. A tragedy, really.
On the other matter you raised, can you really blame anyone for not reading the JW blog?
@Rob
Re: “On the other matter you raised, can you really blame anyone for not reading the JW blog?”
I have to object to this. As you also should know (we discussed a lot there), Jeff atributed a lot with his blog before he got caught up extremely in (Russian) conspiracy thinking and his own confirmation biased approach based on this ‘theory’. This got worse over time.
It’s my opinion. It’s why I switched to Victor’s blog.
But I will keep appreciating many of Jeff’s earlier work.
@TBill
“I feel the current search area covers intentional pilot paths.”
TBill, you’ve done extensive work on your flight paths, and I admire what you have done to try and solve the mystery. But, personally speaking, I can’t see why he would have ever considered plotting a course to any particular seabed feature. It would have been extremely difficult, and pointless. I doubt very much if he had ever heard of Broken Ridge or the Dordrecht Hole. For once, I agree with Dennis.
@Ge Rijn
“Re: On the other matter you raised, can you really blame anyone for not reading the JW blog?
I have to object to this. As you also should know (we discussed a lot there), Jeff atributed a lot with his blog before he got caught up extremely in…..”
Fiddlesticks! Object as much as you want.
@Rob
Fiddlesticks is your 39S/88E confirmation.
In that regard you equal or even beat JW’s confirmation bias.
@ Ge Rijn most hypothesis support unlawful interference from the navigational way-points, turns and altitude changes however no proof as to who. If the implication is of Zaharie Ahmad Shah remember his Facebook interests in Atheism and other non Muslim interests. Aviation psychology is a very intricate subject matter often with the subject divorced and/or has blinkers on solely focused on their intentions, and can be bitter or discouraged. IMHO it is better to stay away from conjecture and focus of the search and seek closure for the families.
I stumbled upon your website through a Google Alert and thought I recognised your name. I have been following Duncan Steel’s website since the beginning and glad to read you are continuing on.
@David @All
I’m not qualified to comment on this particular situation vis a vis the airplanes configuration and any risk to benefit analysis that Zaharie may have considered. However, I would not necessarily say that flying the aircraft in the manner proposed would be ‘out of character’ and not in accordance with the Zaharie’s personality and objectives.
While it is true, as Ge Riin points out, that this does not APPEAR to be a reckless act insomuch as it was pre-meditated and seemingly successfully executed (as the aircraft is yet to be found)…I would confidently posit that inherently the act itself was desperate, reckless and rebellious. It was shockingly irresponsible and callous, executed by a man who had for his entire adult life been responsible for delivering pax safely to their destinations as they entrusted him with their lives.
He snuffed out the lives of the very people who were blindly trusting him to deliver them to safety. I say this only to impress upon readers of this blog that this is a man with significant psychopathy and that it is unknowable as to how he would choose to fly the aircraft.
I don’t think any maneuver or configuration can be ruled out based on risk/reward alone. I would include ‘zoom’ climbs (even if the benefit would be modest or not at all), for example. Maybe those extra lights in the cabin that Andrew mentions were something Z was just not comfortable with? Who knows? I think it was Zaharie who implored his political compatriots: There is a rebel in every one of us…let it out! Don’t waste your life on mundane life style. When is it enough?
It seems to me, in the end, he also walked the walk.
@Rob
I think your confidence in Zaharie not even bothering whatsoever with underwater features at his final destination is bit on the high side and misplaced, but I suppose possible. Given that he took great pains in sinking the aircraft (in your scenario, which i happen to agree with) , it would follow that he would also at least be curios as to what the topography would be at his final resting place, no?
That said, he does seem to have a fixation with ice, so drift analysis aside, the smart money may in fact be south and downrange.
@Donald. Thank you for your opinion.
@Donald. 2 points:
– I had it in mind that his purpose would have been not so much turning out the cabin lights but to go dark decisively, leaving no prospect of a break out resulting from something he was slow to remember to do or had not been apparent he should.
– The zoom you mention has been raised as to how there was a dive 2 minutes (after a stall) from fuel exhaustion.
@Martin
I agree it’s a sensitive subject and we have no right to judge/condemn Zaharie or anyone without any full proof (which there isn’t any yet).
But I believe we should not shut our eyes to the only one onboard we know of (yet) who could have performed all actions presummed to be taken in this flight.
And then it’s also allowed to dive into possible (psychological/religious/social) motivations imo. It could bring something that is helpfull in finding the plane too I think.
And this is what all want and the NOK most of all I suppose.
@Ge Rijn: It is almost impossible in a public forum to dispassionately review and discuss the indicators that suggest possible involvement of the captain, of which there are many. It’s not a bad thing that people defend the reputation of somebody that can’t defend his own reputation. That’s one of the many reasons that criminal investigations are not openly conducted.
@VictorI
It sure is not a bad thing people defend the reputation of the captain which has been (and still is) impeccable as far as I know.
But he was the captain on this flight.
I think it’s not right leaving him out of the discussion just to avoid sensitiveties. In one way or another he played a key-role in this drama.
In one way or another he must have influenced the flight and outcome, also if he wasn’t the culprit.
And we are no criminal investigators bound by legal bounderies.
I think, with carefull respect to the captain in mind, it can be helpfull to discuss possible motives too.
@VictorI
“It’s not a bad thing that people defend the reputation of somebody that can’t defend his own reputation. That’s one of the many reasons that criminal investigations are not openly conducted.”
What I think you are saying is that everyone is entitled to air their views, which is as it should be on a blog. But I also see a potential dilemma. This is not a court of law; is not the primary and overriding goal here to help find the plane? Continuing to debate on whether or not the pilot was a heartless, psychopathic criminal, is in my view not going to help find the plane. There surely has to come a time when one has, to say “enough is enough, we are pretty sure we know who was responsible, let’s get on and narrow down the alternative scenarios? The question of Shah’s culpability has been discussed for almost 4years now. We don’t want to still be here 4years from now, agonising over the same issues.
Or am I missing something?
@Rob: Yes, there is a dilemma. I wish people could dispassionately investigate the facts without the discussion going off the rails, but I don’t think it’s possible in this public forum unless I ban a lot of people, and I don’t think that would be helpful. I recognize that fact and accept it, because I am powerless to change it. That doesn’t change my opinion of what scenario is overwhelmingly most likely. For better or worse, it does limit what I am willing to speculate about publicly.
@Rob
In the “picture is worth a 1000 words” department, it is not hard to see underwater features in flight sim work.
For you, below a possible flight path end point graphic, that I preliminarily feel fits BTO/BFO well. This is Flight Sim PSS777 and I am using a common add-on utility to link the flight path to Google Earth so I can see where the aircraft is going. What I am saying is, it is something flight simmers often do is show the aircraft on Google Earth. Part of the reason I originally started studying PSS777 was to understand if this visual link could be done.
Path picture: goo.gl/LfMT52
Notes: This is 180S CTH flight with heading change to 125S CTH at 31S. Presumed active pilot has descended and slowed down starting 22:50. I manually added the path history (cyan line). The red aircraft icon is being linked over from PSS777 in real time as the flight progresses. Approaching Arc6
with Arc7 coming up (only need about 300 ground speed at this trajectory to fit the Arc6/7 gap).
PS- You have to cut and paste the LINK into your browser
@VictorI
Thank you for being honest. It is a bit like walking a tightrope.
I think I have come uncomfortably close to being banned on more than one occasion (you don’t have to answer).
I am along with many others I’m sure, very appreciative of having a platform like this to air our views.
@Rob
Nice post.
@David
From 17:21 (or perhaps earlier) until he arrives at the ‘FMT’ was certainly the most critical and challenging part of the flight. I have always believed that a ‘one time’ action to effectively cause the aircraft to ‘go dark’ would have been the best course of action as Z would presumably have been task saturated.
Also, and this is perhaps significant, it would be normal for an inordinate degree of paranoia to have infested his mind during the planning and executing stages. He may have become convinced that going as ‘dark’ as was humanly possible was the only way to ensure the aircraft could journey untracked. He probably had a fairly decent understanding of potential tracking systems but was left with a gnawing worry that his understanding was insufficient. The irony, of course, is that this concern proved to be well founded. Those pesky satellites discovered him in the end.
And there is always the breadcrumb idea. That is, the power restoration took place to revert back to a fully functioning aircraft AND leave a trail behind.
The overwhelming consensus, however, is that he would not have had a clue about the satellite stuff (despite AF447) and believed himself to be scot free once he rounded the tip of Sumatra, or thereabouts. I can go along with this, but I still wonder!
@TBill
I cannot get your link to work.
@DennisW
Thank you Dennis
@Donald
“And there is always the breadcrumb idea. That is, the power restoration took place to revert back to a fully functioning aircraft AND leave a trail behind.”
I wondered about that possibility as well. The consensus, as you say, is that he would have been unaware of the hourly handshakes. I still think that’s most likely true. For certain he didn’t want to be tracked as he progressed south; he was careful to keep ACARS deselected.
It’s significant I think, that he reconnected the LH AC bus as soon as he was out of radar range (that cannot be coincidental). The aircraft was flyable with a dead LH AC bus – he could have kept it that way if he thought there was any chance it could give him away. The first logon was him saying “I’m still here, this is no accident, this is something else entirely).
As you said, he walked the walk. He betrayed the trust put in him as an airline captain and what’s more, he deliberately ran out of fuel in the remote wastes of the SIO. To run out of fuel must be every pilot’s secret nightmare, but he went there by design and let the world know it with the second logon (not everyone will agree on that particular point, of course)
The man was a perfectionist.
@Rob @Victor1 I agree. We are fortunate to have such a high quality forum available. I am sure that operating this forum comes with its attendant frustrations and time demands.
@Rob “The man was a perfectionist.” Perhaps. But the more I read about this the more convinced I am that he was also an idealist – a naive idealist who has apparently been beaten by the authorities.
@Donald. Thanks.
A summary from my perspective.
This discussion originated from the aim of finding the wreckage. Sequence, cause and motive, currently indeterminate, follow as best they can from that. Just one example would be finding burnt cabin oxygen cylinders indicating that most likely there was depressurisation.
There can be slightly more confidence in the search width if an unpiloted final descent can be assumed. Most likely IMO this would home in on an electrical configuration included by the ATSB in recent simulation results. That leads backwards to how that configuration could have arisen. I have put forward an early all-normal-generators-off selection by a pilot as a means of going dark followed by an IDG restoration to regain autopilot, that causing the first log-on. @Andrew commented that a pilot with a plan would be unlikely to select power off like that. I then raised the question of whether one driven by spontaneous anger would.
The trouble with any conclusion that a pilot was suicidal is that it depends on the assumed scenario IMO. In this instance that appears circular, with no independent support. Therefore there is no independent means of distinguishing between the distantly possible and probable.
My query was intended to be more about the psychology of spontaneous anger arising from exasperation and despair and how that might lead to someone breaking away from the discipline of their training and practice – road rage being distantly related perhaps.
To me this hypothetical scenario remains possible in the sense that it is not disproven, yet unlikely, there being no evidence by precedent to support it, or corroboration through demeanor.
It looks doubtful that we will get much further, so continuing is unwarranted IMO. Back to finding the wreckage?
@Shadynuk said: We are fortunate to have such a high quality forum available. I am sure that operating this forum comes with its attendant frustrations and time demands.
Frustrating at times, yes. But also very rewarding at other times. The re-start of the underwater search and the extension past the 25,000 sq km recommended by ATSB/CSIRO was at least partially due to the efforts of contributors here to sustain the interest in MH370’s disappearance and to challenge the official narrative. Our success in doing this also makes contributors here and the blog in general a target.
@Donald. My second line, for ‘cabin oxygen cylinders’ please read ‘cabin oxygen generating cylinders’.
@David
I didn’t address the possibility of a ‘road rage’ type incident because I feel the simulator data is evidence enough of a premeditated and pre-planned event. I think it highly unlikely that a triggering moment occurred while already in flight.
I do think that the Anwar sentence and perhaps some personal problems could have precipitated implementation of the plan.
I agree that this is not helpful in finding the aircraft, but I do find it strange that some commenters on this forum are so readily dismissive of the glide/ditch hypothesis. I personally have little, very little, doubt that the PIC was alive and in control until fuel exhaustion and beyond and that pilot commands induced the dive that led to the much talked about BFO data point. The pull-out and subsequent glide is of course all very speculative, but imo worthy of serious consideration.
For example, Don Thompson called this EOF scenario put forward by Ge Rijn and others (as at least being a possibility) ‘inane’. I would be eager to hear his rationale. Perhaps I am taking his description out of context or am misunderstanding? Unless the condition of the debris or something else definitively rules out a glide and/or low to medium energy impact (or higher, depending on AofA), it seems to me a bit irresponsible to be labeling the idea as sheer lunacy.
But like I said, maybe I am misunderstanding his comment?
For VictorI. You might want to log the JW comment to me below for future use here when things get bad.
By continually repeating your cherished piece of misinformation you are deliberately muddying the waters and preventing the community from reaching an understanding of the investigation. In general I value your contribution but I feel I am doing readers a disservice in allowing you to fog the discussion in this way, so I am gong to set your comments on manual approval and will delete future comments that mislead in this fashion.
@Victor “It is almost impossible in a public forum to dispassionately review and discuss the indicators that suggest possible involvement of the captain, of which there are many. It’s not a bad thing that people defend the reputation of somebody that can’t defend his own reputation. That’s one of the many reasons that criminal investigations are not openly conducted.”
I could not agree more. And also could not agree more with sentiments to effect of gratitude for creating and driving this platform. Bravo.
On the discussion, I am always glad to see investigative discussions outside the engineering and sat sphere here. I believe some of Victor’s Best posts have been on the investigatory side (sim path for example). One of the yet to be written post mortems of all this, I believe, may be how much treasure and toil was spent without a true, full investigation by those with the power to do so. We will see.
Thank you, again.
@Donald. “I do think that the Anwar sentence and perhaps some personal problems could have precipitated implementation of the plan”.
The premise is conjectural (to me) but yes that narrative does overcome timing and flight selection issues.
On the glide/ditch hypothesis an impression I have is that a principal reason doubt is entertained about a glide extending search width (if not, like high AoA, it has no consequence in finding the wreckage) is the notion that a pilot would first delay for almost 2 minutes, dive then recover; or zoom, dive (or stall) then recover, in either instance the recovery then leading to a low angle descent.
To me, like other scenarios there needs to be a reasonable explanation for such apparent contradictions or their standing is weak. But I do not necessarily speak for others.
With regard to above statement about Muslim burial practice- isn’t it essentially cremation when one ‘blows themselves up’ with a bomb vest? Why would this type of death be honored then?
Would there be a way to check when cell phone connection was lost and if so use this time factor to help narrow down the search path(latitude)? I just can’t believe no radar picked up this thing!!!
.. or picked up by sonar if crashing into ocean, or realtime satellite imagery for disturbance in that area? Heat signatures..?? I apologize for asking these questions if they have already been discussed.
I don’t want to believe it was the pilot.
What if Innmarsat data is wrong? The eye witness in Indonesia have a different location. Right?
@Donald
“inane: lacking sense or meaning: silly
The term used in your comment above, “sheer lunacy“, is a misrepresentation. Had I intended an allusion towards insanity, I would have used an appropriate adjective.
Refer to a recent comment where I attempted to use analogy to explain “ditching because flaperon”. Confirmation bias established by the finding of the flaperon, & the unfounded furore concerning ‘extended flaps’, has persisted despite all subsequent debris exhibiting damage outcome from catastrophic forces, and ATSB’s analysis of the Pemba-Kojana flap that concluded with high confidence that the flaps were in the retracted position.
“Ditching because flaperon/flap because trailing edge damage” is a conclusion based on superficial observation which requires reinforcement that imagines a procedure to ensure the debris is quickly obscured by submerging. Such a conclusion ignores the detail that is apparent:
a) the trailing edge wedge is not a structural component of the trailing edge control surfaces (flaperon & outbd flap), it’s inherently weaker. A range of scenarios could cause the apparent damage;
b) the structure of the flaperon is intact, the entire inboard section of the flap structure is intact;
c) the fracture lines evident on the attachment parts transect those parts where the material offers significant strength against an upwards, rotating, moment acting on those parts.
Considering b) and c), primarily the condition of the flap & flaperon attachment parts, allows an entirely different conclusion (other than attempted ditching) to be drawn for the detachment of those control surfaces.
@David
We are never going to get to the stage where it’s safe to assume an unpiloted descent. There are numerous reasons why.
That assumption has already been responsible for one failed search, and could possibly endanger another.
Rob writes “We are never going to get to the stage where it’s safe to assume an unpiloted descent. There are numerous reasons why. That assumption has already been responsible for one failed search, and could possibly endanger another.”
Great example of confirmation bias (ditch because flaperon) while ignoring the scope of considerations that produce plausible latitudes along the 7th arc vs distance NW/SE from the arc.
@Rob
This statement does not sound logical to me.
The plane either intersected with the 7th ring in the old area or in the new area (or, less likely, outside the two). The last search may therefore have failed because the plane is in the new rather than in the old area, but not because of pilot inputs.
Also, the assumption is that the plane is close to the 7th arc as the two final BFOs and debris suggest, either because of pilot input/vertical dive or no pilot input, but not just because of “unpiloted descent”.
What you mean is the implausible event that there was pilot input, resulting in vertical dive, and subsequent recovery (without engine power) and then an extended glide.
First of all, thanks for all the wonderful work here, I appreciate this very much.
Now to my question: Is there a specific spot in the timeline of the flight that without doubt needs an active human being in the cockpit to perform the proposed course which is given by the Inmarsat data, e.g. the SDU reboot or the final turn south. Or is it possible that the whole scenario was entered into the flight computers even before the turn after IGARI?
@Nederland
“What you mean is the implausible event that there was pilot input, resulting in vertical dive, and subsequent recovery (without engine power) and then an extended glide.”
That is in no way what I mean, and I would appreciate it if you didn’t try to distort or misinterpret what I said purely to make it fit your own biassed interpretation of events.
Your logic is obviously different to mine.
I will get back to you later today with a fuller critique.
@David
re:”..the notion that a pilot would first delay for almost 2 minutes, dive then recover; or zoom, dive (or stall) then recover, in either instance the recovery then leading to a low angle descent.
To me, like other scenarios there needs to be a reasonable explanation for such apparent contradictions or their standing is weak. But I do not necessarily speak for others.”
I think the 2 minute delay now rather indicates a pilot controlled descent.
The simulations (Boeing, @ALSM, @Andrew) show no steep descent at the right time. Bank angles do not exceed 20 degrees (@Andrew-sim) or recover to 20 degrees: @ALSM-sim).
This suggests to me a pilot took some time (~2 minutes) to prepare for the steep descent after fuel exhaust.
A reasonable explanation could be he wanted to descent as fast as possible to lower/denser altitudes to avoid decompression complications and to shorten the gliding time as much as possible to a certain destination. A long glide from high altitude would not have been an objective I think considering a higher change of detection and unnecessary lengthening of his own end.
And remember the captain was a skilled paraglider.
As one of few who did it before he could have performed an unpowered steep descent like this and a recovery and glide after.
Gerald asked: Or is it possible that the whole scenario was entered into the flight computers even before the turn after IGARI?
Some of the manoeuvers that have been proposed (e.g., leaving a holding pattern, route offsets) require pilot input after IGARI. However, I’d say that it is possible to program the flight computers with a route that matches the radar data and the satellite data with no pilot inputs after IGARI, although that route would have many custom waypoints.
@DennisW: That’s an unfortunate situation, but his comments and actions should not surprise you. There is only one scenario he will ever accept.
@David said: On the glide/ditch hypothesis an impression I have is that a principal reason doubt is entertained about a glide extending search width (if not, like high AoA, it has no consequence in finding the wreckage) is the notion that a pilot would first delay for almost 2 minutes, dive then recover; or zoom, dive (or stall) then recover, in either instance the recovery then leading to a low angle descent.
If we hypothesize that the pilot was alive until the impact (a possibility that I don’t dismiss), then the two minute delay has less significance. For instance, the pilot might have actively balanced the fuel in the left and right tanks, so that the timing (and even the sequence) of the two engine flameouts is unknown. If the right tank went dry first and the left IDG and left backup generator were isolated, that would cause the APU autostart. The pilot might have chosen to perform a steep descent immediately after the flameout of the left engine, which could have been delayed around two minutes from the right engine, and happened to coincide with the SATCOM log-on.
@VictorI
Yes, no biggie. He clings to:
1> The plane is not in the SIO.
2> The fact that the plane has not been found means that the ISAT is flawed somehow.
@Nederland
You said “The plane either intersected with the 7th ring in the old area or in the new area (or, less likely, outside the two). The last search may therefore have failed because the plane is in the new rather than in the old area, but not because of pilot inputs”
I say the last search failed because the plane crossed the arc close to S38 and was pilot-glided beyond the southern boundary of the old search area. The plane never entered the new area.
Why am I so confident? Here is why:
1) There is no evidence whatever to suggest that the pilot lost control at any stage between the takeover at 1721 and crossing the 7th arc at 0019 the next day. On the contrary, the exercise has all the hallmarks of being meticulously pre-planned, and the plan then carried out to the letter.
2) The currently accepted interpretation of the final 0019 BFO remains open to question. A 0.65g downward acceleration during the 8 seconds between 0019:39 and 0019:27, ie coinciding with the exact time of the 2nd logon is I would respectfully suggest unlikely on statistical grounds alone. If it is genuine, the aircraft would most likely have been found close to the 7th arc, and any recovered debris would have been highly fragmented, and originating from many different areas of the plane. In any case, even if the BFOs are a genuine indicator of descent acceleration, they do not rule out a piloted glide. MEFE could have been at FL380, with insufficient aerodynamic damping for adequate control under RAT power alone. The aircraft could have become unstable and gone into a steep descent. A subsequent controlled glide from FL300 with glide ratio 13:1 could still end up outside the old search area.
3) The new search area is based on false assumptions. False assumptions about the validity or accuracy of drift studies, and about the validity of BTO/BFO interpretation. Look again at Fig 5 in Flightpath Analysis Update of 8/10/14. The green Data Error (RMS) optimization zone is distinctly flat topped. All flightpaths in the is zone have an equal chance of being true, the PDF is indeterminate. Put another way, no single path has any distinct chance of being true. In contrast, the Constrained Autopilot Dynamics PDF has a very marked peak at around S38. The DSTG carried out a validation exercise on 6 actual 9M-MRO flights. In each case, the actual ACARS flightpath terminated beneath the PDF peak. In all cases, the actual flightpath was within the top 85% region of the PDF. A pretty impressive demonstration of a statistical method slated by many. But I would point out, only slated after the initial search failed, and the initial search only failed because the ATSB were unable to go sufficiently far downrange of the DSTG hotspot.
Now I have identified a flightpath that crosses the 7th arc at S37.84 E88.62, a flightpath that would have intersected S41 E88, if it had not run out of fuel. A flightpath at M0.82 at FL400 that has precisely optimal range and endurance that the fuel would have allowed on that night, and rendezvoused with sunrise, which is no mean achievement. The sunrise area is a constantly moving target, both annually and daily.
@Rob
The new search area is based on false assumptions. False assumptions about the validity or accuracy of drift studies, and about the validity of BTO/BFO interpretation.
The question is why these assumptions/interpretations are “invalid”? It seems like you are simply declaring that because they do not support your terminus. Give us some meat here that supports your “invalidity” assertion relative to the drift studies and BFO data.
@Rob
You are of course entitled to your opinion.
However, as to fragmentation of debris, those pieces that are not entirely fragmented normally have pull rather than crush damage, suggesting that the type of primary damage is not from direct impact (perhaps pieces from the aircraft that didn’t come off during descent, other than the vertical stabiliser, are too small to identify? Blaine Gibson et al have found various very small pieces possibly from an aircraft.)
Yes, the Flightpath Analysis Update of 8/10/14 has a hotspot around 38S. However, this is based on a) FMT before 18:40 and b) no pilot input (e.g. change of heading or speed) thereafter (notwithstanding pilot input during end of flight). The northern area (not assuming FMT before 18:40) always had better BFO fits than the southern area (on this the earlier ATSB analysis updated on 13/8/2014, this is also what I took from my own experiments with the BFO calculator).
Still, you have to assume either final BFOs are entirely wrong/not genuine or steep descent (either willful or unwillful) and thereafter recovery.
@DennisW
Please try this alternate link:
MH370_PATH
This is how I see an active piloted 180S CTH path probably ends consistent with meeting BTO/BFO. View is PSS777 linked to Google Earth.
@Rob: You equate choosing assumptions that result in a peaked PDF as validation that those assumptions are correct. It’s also possible that those assumptions over-constrain the problem.
@Nederland
Re: “However, as to fragmentation of debris, those pieces that are not entirely fragmented normally have pull rather than crush damage, suggesting that the type of primary damage is not from direct impact (perhaps pieces from the aircraft that didn’t come off during descent, other than the vertical stabiliser, are too small to identify? Blaine Gibson et al have found various very small pieces possibly from an aircraft.)”
Like to mention all the pieces found so far don’t show significant crush-damage but tension/pull damage. Even the engine cowling pieces and the Rodrigues-panel shows intact honeycomb and most of all of the nose landing gear-door piece which would be the first to be crushed in a nose down/dive impact. The lcd-mounting frame is hardly deformed/damaged too.
Also the small panel-pieces found by Blaine show intact honeycomb.
No debris shows clear signs of a high speed nose down impact. On the contrary. These facts should not be ignored.
@Tbill
Thx. That worked fine.
@Ge Rijn,
Yet again, a wilful misrepresention of the condition of the physical evidence so far recovered.
Your apparent expectation that honeycomb core should be crushed so as to indicate a high force impact has no basis in reality.
@GeRijn
“Fiddlesticks is your 39S/88E confirmation. In that regard you equal or even beat JW’s confirmation”
Thanks GeRijn. I’m in good company then. I will take that as a compliment.
@DennisW
@Nederland
Dennis, what you are saying in effect is “where’s the beef?”
Look no further than CSIRO’s efforts to comply with their master’s wishes; the discredited Priority 1, 2 and 3 spots.
@Nederland
Thank you. I have taken on board what you say. Yes, we are all entitled to our opinions, be they right or wrong. Time will tell, and I don’t think we will have to wait too long.
Thanks
Rob
@GeRijn,
Re debris found so far— Let’s not forget that any debris crushed by an impact might not float. So only the parts that have failed in tension are likely to be found by Blaine on the shorelines.
Re the ‘nose gear door’— this may have been torn off during the final high-speed dive. Interestingly, this part might be pivotal in working out what happened initially. Consider this, if it can be established that the door would not rip off in a high speed dive, then maybe it had opened during the initial event. Something like the crew oxygen bottle rupturing next to the nose wheel bay and initiating a decompression through the nose wheel bay might do this….so the door remained open for the duration of the flight, only to be torn off during the final high speed dive.
@Ge Rijn
You always seem to forget the vertical stabiliser state and damage. If any piece is found and identified despite high speed impact, you don’t have to look any further.
SC finished dealing with “technical issues” and is now in position to launch a few AUVs to complete the 25nm width on the NW side, adjacent to Sortie #3.
-35.67379 91.11619
Mike
@Rob. “We are never going to get to the stage where it’s safe to assume an unpiloted descent.”
This was not my expectation Rob. I believe a piloted descent is possible. It isn’t that. It is that the combo of the 2 minutes delay in the plunge and a recovery from that to a high L/D glide need a reasonable explanation. The 2 mins maybe could have been taken in a zoom and consequent, though that would hardly be consistent with a recovery.
@Victor has drawn attention to one engine continuing past AC loss, the ATSB electrical unpiloted configuration, but now with pilot. That would explain a delay but likewise not the apparent mismatch of a recovery. Another but more remote possibility would be that the pilot shut down that engine which had the IDG operative when fuel remained in its tank still. The APU then auto-started and immediately he shut it down. After the second engine suffered fuel exhaustion he put the nose down to windmill start that with fuel remaining. The steep descent continued, the final log-on being triggered at low altitude at the requisite descent rate. However again this hardly would be consistent with a recovery.
A BFO-consistent descent leading to a continuing plunge is certainly possible. I for one have posted about that and I note similar remarks from @ Nederland above. Manning is not so much the problem, that being why he would enter a steep descent only to recover? Lose altitude just to regain much of it?
To me a long glide bespeaks the final BFO’s being wrong. However if it is not to be ruled out entirely how then should we prioritise the search?
A ‘long glide’ could have been in any direction, even a circle while evaluating sea conditions. It could also have been truncated by a dive.
As to direction, an assumption of it continuing on the aircraft’s previous course would limit search area but what are the chances of that? More likely a general case would be made which supposed that all the area within an agreed (somehow) radius should be searched. That entails a search close to the 7th arc still. In that event the current search would not have been ‘wasted’. Even if unsuccessful the next step would be to assess the best mix of widening and lengthening. Who knows, OI or another might agree to continue the search for the same reward, utilising coverage from those previous.
Unfortunately though it is yet more complex than that if you take other piloted possibilities into account, including various cockpit selections available to him. For example, under the ATSB’s 2016 electrical configuration the remaining engine could be operating still during the dive descent, the BFOs still being APU initiated. After any recovery a pilot could then fly a distance under single engine power using some combination of fuel saved by electrical load reduction in the southern leg (eg no loads from IFE, galley, packs), step climbs and transfer of fuel in favour of the surviving engine. Had engine bleed air been off for a time that would have saved fuel too. He might re-engage the autopilot under APU power and if the left engine fly say twenty minutes, over 100 NM; or set the A/P to complete that.
With a possible long glide in the same direction to follow or an unmanned descent, how far to extend in that direction in a search?
The problem in any search continuation speculation is to draw limits such as has been done to date to maximise search prospects per dollar in the situation of the time. It could well be that a piloted descent would continue to be left aside, not because that is impossible but that its inclusion makes is unhelpful. A hypothetical. Suppose the probability of finding the wreckage was .85 without considering the pilot possibility and 1.0 with, but the area entailed to lift the probability by that remaining 15% would multiply the total search area by 10 times, which would you choose supposing to search to 85% would take a season?
As to your expectation put to @DennisW and @Nederland that, “Time will tell, and I don’t think we will have to wait too long”. If time tells that the wreckage is not in the new search area either, that does not tell us where it is. There will be no sign I think pointing to your preferred area and thus your assertion will be left unrequited, as it stands.
About your post for @Nederland, “MEFE could have been at FL380, with insufficient aerodynamic damping for adequate control under RAT power alone”. Interesting thought although I think it has been established that all other flight controls would join those powered by the RAT except those few PCUs controlled by the L2 ACE, through engine windmilling.
@Ge Rijn, Much of the above is intended for points made either by Rob, you or both, supposing fiddlestickings permit.
Your, “The simulations (Boeing, @ALSM, @Andrew) show no steep descent at the right time.”
The ATSB, Boeing and SSWG imply Boeing’s does. The yaw from that asymmetric untrimmed engine thrust might well be sufficient to create the descent rate. That applies to neither @ALSMs nor @Andrew’s. More specifically as to timing I think all one can say is that apparently the published graphs do not support that.
“A reasonable explanation could be he wanted to descend as fast as possible to lower/denser altitudes to avoid decompression complications….”.
Cabin altitude increase due to leakage after engine failure would be comparatively slow I would think. The outflow valves would close if on ‘auto’.
“…..and to shorten the gliding time as much as possible to a certain destination.” If he is shortening the glide much will he exceed 25 NM in that descent? I am supposing you mean no recovery, ie conversion of descent speed to altitude, since that would take time?
@Victor. A complementary note to what you say. An engine powered post AC loss could be either. The pilot could have balanced fuel as you say but also running all electrics on one engine would burden its fuel consumption compared to the other, so one starting with more could finish with less after several hours, all else equal.
@Victor. A complementary note to what you say. An engine powered post AC loss could be either. The pilot could have balanced fuel as you say but also running all electrics on one engine would burden its fuel consumption compared to the other, so one starting with more could finish with less after several hours, all else equal.
@Victor. On my point above, unpiloted, assuming the right consumes fuel at the greater rate with loads balanced, worsening that by having it carry the total electrical load will bring its fuel exhaustion forward and leave the left with more fuel than otherwise. The 7th arc would be brought forward.
With the right carrying an electrical load of say, say 80 kW at the engine, say 30% engine to generator output thermal efficiency for 7 hours the extra fuel consumed would be around 100 lbs, 45 kg so maybe not that much assuming that load and efficiency were around right.
I expect the fuel consumption difference between engines, loads balanced, would be greater than this via PDAs etc so that were the left carrying the electrical load it would decrease that but not reverse the sign.
Thus most likely it would be the right to fail first still. in which case there would be no yaw after the left’s failure due to the right then continuing, no A/P or TAC.
Pertinent as you know, the Boeing simulations included,”In an electrical configuration where the loss of engine power from one engine resulted in the loss of autopilot (AP), the aircraft descended in both clockwise and anti-clockwise directions”.
On the face of it, Boeing has included left-engine-fails-first(s) for completeness but that is not what would have happened.
@Victor. The first para above repeated by accident.
Time for a cup of tea, or somesuch.
The question about the debris is not “ditching or not-ditching”, it is about “controlled input or not”. For me, whichever way i look at it, there is a high probability of controlled input. Taking the flaperon for instance.
* it has been mentioned that mid air separation could be due to high g load or flutter.
– If the plane is unpiloted, my understanding is that the AoA is constant and, unless i missed this, the maximum g I have seen in all the simulations with “no pilot input” is below the manufacturer test load. High g could however develop if there is manual input.
– the only one argument that can tally with a mid air separation and no pilot input is flutter. However, the flaperon is in a relatively structural stable (stiff) area of the plane (ie not very sensitive to flutter) and even if there is, flutter will be dampened due to the fail safe nature of the hydraulic system. It is also hard to conceive that flutter will lead to damage at that trailing edge location (aspect ratio too short) nor torsional hinge failure (vibration mode will be bending). In my view, the likelihood is very low.
* Now if there was no mid air separation, the absence of structural damage on the component frames such as flaperon indicates a lower energy impact, big enough though to cause debris. This also most likely the result of a controlled input to lower the impact energy.
I am not sure whether controlled input could affect the plus or minus 25NM band, but, as far as i know, there has been no official calculations done to demonstrate that flutter was a possible damage mode on this component and, in my view, calculations re flutter damage are key to inform whether there was a controlled input or not.
Saying that flutter caused flaperon damage due to exceeding safe flight envolope is not a valid argument. Not all components will be subject to high degree of flutter if flight evelope is exceeded. Also the speed needs to be relatively constant for flutter to stay in the natural frequency range of the component.
@Nederland
I did not forget about the vertical stabiliser piece and agree this is harder to explain considering a relatively low speed/low AoA nose up impact.
A bit like the No-step piece both are leading edge pieces from the tail section. In a nose up impact though this section would hit the water first taking the brunt of the initial impact (speed). Leading edges would be subjected to those forces too at this locations. Then seconds later, seperating engine- and trailing edge wing parts could also have hit the leading edge of the vertical stabiliser. Or the tail section seperated all together on impact like Asianna 214 which showed similar damage on the V.s. at the same location.
It’s no proof ofcourse. But the piece is no proof of high speed impact either (the Malaysian report on the piece did not draw a conclusion on this also).
And I don’t agree with your statement: “If any piece is found and identified despite high speed impact, you don’t have to look any further”.
If ~20 pieces only show tension damage and no significant crush damage the one piece found that shows significant crush damage surely does not prove all the other pieces are the result of a high speed impact.
It only tells this piece has to be explained differently.
More important is to explain the intact leading edges and trailing edge damage of the flaperon and the outboard flap section, the nose-gear door piece and the undamaged forward edge of the big right hand engine cowling piece with a high speed nose down impact. And the fact that ~90% of the pieces found are trailing edge wing/controle surface and engine related pieces.
All pieces you could expect to seperate in a ditch-like, low AoA/nose up impact.
Imo this cannot be explained by a high speed nose down impact.
Only by a ditch-like impact.
@David,
You show comprehensively how complicated the assessment of the final sat data becomes when we allow for the possibility of pilot input. Whoever piloted the plane (personally I believe it was Shah) had quite a few options for handling the situation, which unfortunately multiplies the number of flight-end scenarios.
And if we accept that the BFO which indicates the plane was descending at a high speed rate, is reliable – well, it’s only a snap shot of what happened during these few seconds. It doesn’t allow the firm conclusion that these seconds of high speed descent actually ended in the plane’s ultimate crash near the 7th arc.
Like you, I have problems, though, with the idea that the pilot entered voluntarily into a steep descent, only to recover and then transit into a glide. It’s probably not impossible, but those who propose this scenario, need to come up with a convincing explanation why the pilot initiated a steep descent prior to the glide. It seems to be a contraproductive move with the inbuilt risk of compromising the plane’s integrity during a high speed descent, which subsequently could compromise the glide. As long as there is no plausible explanation for such a maneuver I prefer a scenario where the plane crashes near the 7th arc – even if the high speed of the descent, which is suggested by the BFO, seems to hint at a final pilot input before the crash – as in pointing the nose down manually.
You also point out correctly that all gliding scenarios – with or without a recovery from a dive – can’t answer what this glide may have actually looked like. It could’ve gone into any direction, depending on the intentions of the pilot. And yes, it could also have been truncated by a final dive. Even if there was a glide, it doesn’t follow at all that there was necessarily a ditch!
The problem of any search is, that some limits have to be set, since the search can’t go on indefinitely. It does make practical sense to use the 7th arc +/- x nm of a tolerance zone as a restricting limit, even if final pilot input can’t be excluded anymore, or is even deemed likely.
If the pilot really executed a protracted glide, then the chances of finding the wreckage are indeed very slim, especially if the pilot managed to put a considerable distance (more than 25 nm) between the plane and the 7th arc. But we don’t know if that really happened. Right now it’s only one of many possibilities. Therefore it’s definitely worth looking for the plane within certain limits.
The current search limits seem to be reasonable, since they allow for a number of plausible crash zones. And maybe, the search will be extended if it comes up empty. OI wants to get paid after all 🙂 It might depend on how much of the designated area has actually been covered after 90 days. That the plane might be lost forever because of bad luck or because that’s what the pilot might have intended with his actions, is a risk which needs to be accepted.
@David
Thank you for your considered reply to my earlier post. Much appreciated.
A couple of your points I would like to respond to:
1) “A ‘long glide’ could have been in any direction, even a circle while evaluating sea conditions. It could also have been truncated by a dive.”
Yes I grant you, in theory a long glide could be in any compass direction. However, the great circle flight path toward S41 E88 crossing the 7th arc near S38, ie. the flight path I’m proposing, required the most efficient flight setting possible for constant Mach, namely M.82 at FL400, which to me implies that the pilot was seeking to cover the greatest distance possible in a straight line, which in turn implies (to me at any rate) that a piloted glide would be continued in the same direction, to further maximise the total distance achieved. Why take the trouble to fly as economically as possible (range wise) for over 5 hours, which I believe he did, and then squander it at the end by not continuing on as far as possible in the same direction. Not to do that makes no logical sense to me.
There is also wind direction and sun direction to consider, as well as the fact that the aircraft wasn’t found in the original search area. This flight path also synchronised with sun elevation, sun 6deg above eastern horizon. All the above factors suggest to me that on balance, a glide in the same direction ie. continuing toward S41 E88, would be the most likely.
To be continued…
@Ge Rijn,
Asiana 214: it may have escaped your attention, the engines came to rest forward of the fuselage. Large heavy objects, inertia, sudden deceleration.
This idea for crushed honeycomb: do take some time to understand the compression strength of the honeycomb material used within a composite panel.
A terminal case of bias confirmititus, symptoms exacerbating, seek help.
@Don Thompson
In Asiana 214 the engines were under power still producing thrust that’s why they ended up further on the runway.
Anyway this would not happen in a ditch. Engine(s) and/or pieces will seperate while the fuselage moves fast forward subjecting the aft of the plane to impact damage from those pieces.
Do take some time to come with arguments instead of the usual put-downs.
@David
Continued from above…
So to recap, I think a glide in the same southerly direction was most likely, which is useful when it comes to sizing the required search area. Parking the slightly controversial final BFOs for just a moment, a 16:1 glide from FL380 is technically possible, with a corresponding splash point at about S39.20 E88.36. A circle of radius 20nm, centred on S39.20 E88.36 is about 4,300sq km in area, and could be covered in 5 days. This area extends far enough south to encompass a 20:1 glide, arguably the max theoretically possible, and extends back far enough toward the original search southern boundary (in fact it conjoins it) to beble sure of encompassing a 13:1 glide. A 13:1 glide is possible, even if the final BFOs are a valid indicator of descent acceleration confined to the 8sec snapshot. For reference the Gimli Glider managed about 12:1, the TransAt flight would have been capable of 13.5:1, if it hadn’t needed to loose energy in order not to CX overshoot the target airport.
2) “Time will tell, and I don’t think we will have to wait too long”. If time tells that the wreckage is not in the new search area either, that does not tell us where it is. There will be no sign I think pointing to your preferred area and thus your assertion will be left unrequited, as it stands.”
Fortunately, I don’t share your pessimism. The M0.82 FL400 flight toward S41.00 E88.00 is specific enough and unique enough to warrant attention from OI in due course, once financial considerations serve to concentrate their thinking.
@David
I rather could have given the argument of oxigen-complications instead of decompression-complications as one reason for the steep descent to arrive at save altitudes as soon as possible.
Wouldn’t this be the first thing a pilot would do after all engine flame-out at high altitude?
You could give the argument he then just put up his oxigen mask but maybe by then a lot of the pilot-oxigen was allready used up in the earlier stages of the flight and this was a better alternative.
@Ge Rijn writes “Do take some time to come with arguments instead of the usual put-downs.”
Discussion has been attempted, often. Re-read your own comments, typical form: “blah, blah, blah. Imo this cannot be explained by a high speed nose down impact. Only by a ditch-like impact.“.
@Ge Rijn
Re: ‘… but maybe by then a lot of the pilot-oxigen was allready used up in the earlier stages of the flight …’
Highly unlikely. The crew oxygen system was recharged to 1800 psi just prior to the flight. That is enough oxygen for one person for around 26 and half hours in a completely unpressurised environment at 36,000 feet.
@Ge Rijn
In Asiana 214 the tail section hit a wall?
But there is no wall on the Ocean.
@Sabine Lechtenfeld. Thank you. Obviously we are on the same page.
@Rob. Thanks for that. “The M0.82 FL400 flight toward S41.00 E88.00 is specific enough and unique enough to warrant attention from OI in due course….”
I wish you luck with that though uncertain of your chances. You may need some patience.
Incidentally my remark about fidddlestickings (I left out the, “..at thirty paces”) was intended to be light hearted though I do not have the skills to get a smiley in there. Pleased that neither you nor Ge Rijn seem to have taken offence.
@Ge Rijn. Thanks. Yes, an oxygen mask but pressure I think would have dropped slowly all the same. The aircraft has 2 large bottles (each 3150 litres of oxygen) which should have been ample for one user, even with earlier consumption by him. I see Mick Gilbert has given some consumption figures confirming this.
@Nederland
Yes, in Asiana 214 the tail section hit the slope to the sea just before the runway. But hitting a medium sized wave with ~150mph would have had the same effect.
Hitting water under a high angle with such speeds is about the same as hitting a concrete wall.
@Mick Gilbert
Yes you are right there should have been more then enough oxygen.
But then I still think a steep descent to lower altitudes would be more conveniant in this situation (after flame-out) only requiring short use of the oxygen mask (if necessary at all).
@David
Patience and perseverance are two watchwords of mine. I remain confident.
I think it was Robert Louis Stevenson who coined the phrase “better to travel hopefully than to arrive”.
I’m sure GeRijn and me will remain on friendly terms. I will provide the smiley face for you 😀, even though some participant take a dim view of such frivolities
@David, I also have only limited emoji-skills, but sometimes I find the smileys useful.
They will appear automatically if you type in the following signs : ) but without a space in between 🙂
@Ge Rijn
So, basically the vertical stabiliser, even though this is the best protected part in any crash scenario, got hit by a giant wave and therefore shows crush damage, but is still identifiable. And the screen frame was washed out of the interior and, in the process, got badly distorted?
How about the other parts from the aircraft frame or interior? Did they all sink largely intact because of a halfway sucessful ditching (despite giant waves) and have therefore not been found on beaches? Is this your main reason to say a ditching is more likely than high speed crash?
@Nederland
I did not mention giant waves. In a nose-up ditch scenario the tail section would hit the water first at the highest angle of attack and highest entry speed. Hitting a short wave or even a flat water surface could have seperated the tail section depending on AoA and impact speed.
The leading edge of the vertical stabiliser could also have been hit by seperating wing and engine pieces.
Also an aft door could have seperated (like in Asiana 214). The hull also could have been breached at other places allowing some broken none structural panels and not firmly fitted parts (like the screen mounting) that broke and seperated on impact to float out of the plane.
Take a look at the Asiana 214 interior pictures after the crash landing. They speak for themselves (easy to find on the net).
Several screen mountings seperated and numerous interior panels.
Then consider Asiana 214 stayed mainly intact. Wings did not seperate and besides the tail section the fuselage remained in one piece even after cartwheeling across the runway.
Yes I think it’s quite possible most of the interior content of MH370 went down with the plane and only some none-structural items escaped the hull.
Which are the ones that have been found so far.
@Ge Rijn, @Nederland
“But hitting a medium sized wave with ~150mph would have had the same effect.
Oh yes, an imaginary “medium sized wave”
N106US impacted the Hudson River at a ground speed of 128kts, 147mph, a pitch angle of 10º, and RoD 750fpm. Its empennage, the horizontal & vertical stabilisers, remained intact and attached.
HL7742 impacted the SFO sea wall at a ground speed of 105kts, 120mph, a pitch angle of 12º, and RoD 500fpm. Its entire empennage detached aft of the pressure bulkhead with the vertical and horizontal stabilisers departing.
Rock, water, wave. All the same when ‘opinion’ comes into play.
Allowing as seems reasonable the assumptions that the pilot (I) was unaware of the Inmarsat tracking capability so believed after leaving radar range there would be no record available to others of the flight path and (II) wished to provide as best he could that the aircraft would never be found, then would anyone be able to say please if he himself guiding the aircraft in its final minutes could be expected to make much material difference over a pilotless decent in minimizing the debris field such that it could be made less likely to leave traces? (I am mindful of @Victor’s comment under the previous post to the effect (as I recall, if I understood correctly) that although if flying on autopilot so appropriately trimmed etc. for much of the time, post-flame out the A/P would disengage itself and the aircraft would automatically revert to the last pilot-set trim – that I suppose (in my ignorance!) would be capable of being set to make the aircraft perform a steep dive which I understand might be a good way of minimizing the debris field.)
If he considered that the location was undetected and undetectible, would he be much bothered to stay awake and alive for an extra five or so hours after the FMT with nothing much to do but contemplate the enormity of his deeds and his own less than pleasant death from high impact or drowning when the only apparent purpose (for a rationally-minded individual!) would be to possibly reduce the amount of debris that he assumed would never be found anyway? (I grant he might have acknowledged that currents might well cause parts to wash-up as they have on distant shores which may explain why the crash site was away from areas where currents would all the sooner deposit debris on the coast of Australia.)
If he would not, then causing the final half hour or so of flight to be after sunrise, thereby to provide light enough to make easier a piloted decent, might not have been a consideration. Rather, it may have been thought better to navigate to allow the final moments of the flight to be in darkness, all the better to hide the aircraft from any chance mariners. Would anyone know please if a crash site in darkness is at all plausibale, being consistent with the BTO/BFO data and debris tracking?
@Ge Rijn wrote “The [MRO] hull also could have been breached at other places allowing some broken none structural panels and not firmly fitted parts (like the screen mounting) that broke and seperated on impact to float out of the plane. Take a look at the Asiana 214 interior pictures after the crash landing.”
Review the NTSB Survival Factors Group report that details the post-impact condition of HL7742: interior ‘monuments’, at worst, dislodged from fixed locations but none fragmented in the way that is evident from the recovered fragments of 9M-MRO. The report contains 128 images detailing HL7742’s condition.
Even HL7742, significant as it was, does not present any damage characteristics that illustrate the extent of forces wreaked on those parts recovered from 9M-MRO.
@formula
“Would anyone know please if a crash site in darkness is at all plausibale, being consistent with the BTO/BFO data and debris tracking?”
You are touching on a contentious subject with as many different opinions as participants here.
The BTO does not support a flight ending in darkness. The best BTO fit is a flight CV ending shortly after sunrise. For the best BFO fit, take your pick, but all flights end in full daylight.
Personal opinion only – a daylight ending would be preferable, not least to be able to make sure you weren’t coming down next to a ship or yacht, and to be able to see the surface. An uncontrolled impact would by general agreement, produce the most debris
@Ge Rijn
“Hitting a short wave or even a flat water surface could have seperated the tail section depending on AoA and impact speed.
The leading edge of the vertical stabiliser could also have been hit by seperating wing and engine pieces.”
OK, so the whole tail section came off due to impact with water. How does this account for crush damage specifically to the vertical stabiliser, when this is the last piece to hit the ground (unless the ditching was performed upside down)
I also think there is a difference between a tail section hitting a (vertical) wall (Asiana) and a (horizontal) ground.
I am then trying to visualise the direction it would have taken for wing or engine pieces to crush the vertical stabiliser and rip it into large pieces, as well as the size of these pieces.
@Nederland
Maybe this video can help you visualise what could have happened to the tail section. It’s not exactly a B777 but a Boeing still and a fairly large one in that time:
https://www.youtube.com/watch?v=WMk8z3h2QC8
@Don Thompson
I’ve seen all those images and have to disagree. Take a close look at the dislodged closet f.i. and the seats. You can see several fragmented panels and lcd-mountings seperated. Throughout the plane many interior panels seperated. And also the left aft door seperated on impact. Concerning other damage on wings etc. this was a ground crash-landing ofcourse with quite other dynamics than a water crash-landing.
@David
Yes @Rob and I have discussed ‘fidlesticks’ too often to take offence after discussing for +three years.
@ Rob
“Personal opinion only – a daylight ending would be preferable, not least to be able to make sure you weren’t coming down next to a ship or yacht, and to be able to see the surface. An uncontrolled impact would by general agreement, produce the most debris”
Thanks for that and prior, much appreciated. I understand.
@HB: If you allow the possibility of pilot input after fuel exhaustion of one or both engines (a scenario I don’t dismiss), the damage and separation of the right flaperon could have occurred if the pilot turned the wheel (or applied aileron trim) during a high speed descent, which also could have been initiated by nose-down input. This is similar to what was seen in SilkAir 185, where empennage parts separated before impact, likely due to high speeds and high loading from pilot input. The parts recovered from the debris field at the point of impact were mostly shattered, small fragments, similar to the many parts recovered for MH370. Meanwhile, the parts of the empennage, recovered some kilometers away, were larger with intact leading edges and tattered trailing edges, resembling the right flaperon of MH370. Radar data showed that SilkAir 185 descended from an altitude of 35,000 ft to 15,500 ft in 24 s, which represents an average downward acceleration of 1.7g. (No data from the FDR was available, as the FDR was disabled prior to the final descent.)
As most of you know, the Malaysian weekly report was just published. The items of interest are:
* Seabed Constructor left Freemantle on the 12 February and had arrived at the search area at 2245 UTC on 15 February 2018.
* Upon arrival, the vessel was on standby due to adverse weather.
* On standby, Seabed Constructor initiated calibration of equipment before resuming the search operations.
* Weather is reported to be marginal for the week ahead.
No mention is made of what activities took place when AIS was disabled in the search area at the end of the first swing, nor what ongoing activities in the (current) second swing require Seabed Constructor to re-visit areas that were already supposedly scanned and cleared.
It’s not clear why Malaysia and Ocean Infinity feel the need to hide whatever issues are occurring. Ocean Infinity is using new technology that has great promise to drastically improve the economics of seabed searching. Reasonable observers would expect that there would be issues that need to be addressed before the full capability is achieved.
It makes me think that even when MH370 is found, the Weekly Malaysian Reports will be the same. No mention of anything significant
@Don Thompson @Others
A link to the Asiana 214 interior damage:
http://www.chinaaviationdaily.com/news/27/27594.html
Scroll through the pictures and you’ll find the dislodged closet and fragmented/missing panels and the seperated lcd-mountings (two if you look well)
@Rob
The video I posted might interest you too.
https://www.youtube.com/watch?v=WMk8z3h2QC8
The captain waited till oncomming daylight with his ditch to see the swell and to burn fuel during the time before to prevent fire.
@David
Regarding your scenarios, at what point in the continuum of loss of normal electric systems does the Flight Data Recorder finally stop recording?
Looking at this video again I realize this could be hard for the NoK to watch who are following @VictorI’s blog.
Please excusse me, NoK for any distress it might cause.
My only objective is to help find the plane and some closure/peace of mind to you too.
@Ge Rijn
I don’t think it is particularly surprising that the aircraft frame separates on impact, but I couldn’t see any specific damage to the vertical stabiliser in that video. (I couldn’t see the tail section at all to be fair).
@Nederland
Oke. Now it’s not suprising to you anymore the tail section could seperate on impact. In a controlled ditching.
Whatever happened to the vertical stabiliser remains unsolved to me also.
The crash-landing of Asiana 214 proves the vertical stabiliser can remain intact with only pieces ripped out at the leading edge at the HF antenna area which can be considered as a ‘weak spot’ in the leading edge. It’s at the exact same spot as the MH370 piece:
http://www.alamy.com/stock-photo-ntsb-investigators-view-the-tail-section-of-the-asiana-flight-214-57988670.html
Ethiopia 961 is a better example. Which is also viewable on YouTube. Although that hit a Reef before breaking up & of course it was closer to shore….
How long did the wreckage of Pan Am 6 stay afloat for?
@GeRijn
GeRijn, thanks for sharing the video of PanAm 6, and the Asiana 214 interior photos with us. The Asiana interior pics serve to show how badly the interior can be messed up in a “relatively” low energy impact. The state of the cubicle partitions were interesing.
The Boeing Stratocruiser engines and propellers were notoriously unreliable. Only 50 or so were ever built, but several were written off over what turned out to be a very short service time. One disappeared over the Brazilian jungle. Makes you wonder howw people could be persuaded to fly in them! Those were the days.
@Ge Rijn,
Above, you provide a typical example of diversion behaviour when serious engagement is attempted.
You were invited to review a specific set of images, compiled as part of an NTSB investigation, and relevant to the current debate. But no, you diverted attention to another set of images. This is typical: diverting to any of HL2264 (Asiana), N106US (USAirways), N4522V (China Airlines), ET-AIZ (Ethiopian Airlines), and even N90943 (PanAm in 1956).
This habit appears to be some form of defence mechanism, a need to divert attention and reframe the context of discussion into your realm of control. It’s certainly a mild trolling behaviour, it is provocative and anti-social. You claim psychology competence, please explain it.
@TBill. “…..at what point in the continuum of loss of normal electric systems does the Flight Data Recorder finally stop recording?”
Good point. The manuals show it powered by the right AC transfer bus, so it will stop recording on AC power loss. To power and restore it in the air an IDG, backup generator or APU would need to be on line.
The voice recorder the same (AMM).
@Sabine Lechtenfeld. Emoji.
Thanks. This is the outcome. Will see how it posts.
It was OK before shifting it to the site from Word.
I need lessons, or my a contraption needs a good talking to.
No matter thanks.
@David,that’s strange 🙁 I just type the two symbols without a space in between, and the emojis appear. But I know another website, where I also get these pecuiar crossed rectangles.
Doesn’t matter – we will figure out what you want to tell us 🙂
@Ge Rijn
The problem with the Asiana flight 214 crash photos is that it is unclear when they were taken. If you view the full sequence of photos on the NYC Aviation website it is apparent that some of them were taken after the fuselage was broken apart subsequent to the crash and some of them were taken after the wreckage was relocated. Photo 11 of 33 shows the rear fuselage being lifted as a slung load with slings through the No 3 and No 4 doors; that photo also shows the dislocation of the lower fuselage below the cabin floor. Accordingly, it is unclear whether we are looking at crash damage or damage caused by the separation, lifting and relocation of that section of fuselage.
In any event, as has been pointed out repeatedly, Asiana flight 214 was not a ditching; it was a crash into a rock and concrete seawall. After the main undercarriage separated on striking the seawall, the bottom of the cabin section in the photographs you refer to bore the brunt of the crash impact. The impact forces were sufficient to cause the empennage to separate at the pressure bulkhead. Not surprisingly, the rear cabin (Zone C – seat rows 29 – 42) had the highest proportion of serious injuries (30 out of a total of 49 serious injuries occurred in that section) and the only in-crash fatalities.
As to ‘Hitting water under a high angle with such speeds is about the same as hitting a concrete wall., good grief! No, it is not! That is complete and utter nonsense! For fear of having to state what should be blindingly obvious, concrete is a solid that strongly resists deformation; water is a liquid that weakly resists deformation but that allows displacement. It all comes down to rate of deceleration; strike concrete and deceleration is effectively instantaneous, not so with water.
Further, and again as has been pointed out repeatedly, the damage to the vertical stabiliser of Asiana flight 214 is very clearly strike damage. Moreover, the section missing in the 214 photo is largely from the leading edge. Debris item 22 is most assuredly not from the leading edge (it is a section of vertical stabilizer panel from behind the leading edge) and it shows no apparent strike damage.
Separately, the chain of correspondence regarding your reasoning for a pilot-controlled dive subsequent to fuel exhaustion is perhaps yet another example of why some contributors find your style perplexing.
1. You offer up a hypothesis – that the rapid descent as evidenced by the final BFOs was to avoid issues with depressurisation and hypoxia (need to use crew oxygen system) subsequent to fuel exhaustion.
2. The factual basis for that hypothesis is refuted – the airplane would not depressurise rapidly subsequent to fuel exhaustion and there would have been an abundant supply of crew oxygen.
3. You ignore the factual rebuttal and restate your hypothesis – ‘Yes … But then I still think a steep descent to lower altitudes would be more conveniant in this situation …’
Your approach is somewhat worse than ‘faith-based’; if faith is belief in the absence of evidence, you seem to exhibit belief in contradiction of evidence. Now, you are most certainly not the only contributor to doggedly ignore contradictory evidence while advocating your hypothesis but your advocacy is regular, repeated and strident and I’d suggest that it is that which makes it annoying to at least some of us. If nothing else we (or at the very least, ‘I’) tire of having to restate the same factual rebuttals every time you repeat a largely unsupported hypothesis.
@victor,
You are correct. Damage is possible due to high g due to controlled input. My argument is that it is very unlikely without controlled input. There is no likely damage mechanism to produce the debris patterns observed if there is no controlled input (i don’t necessarily imply ditching). If flutter calculations could be done officially, this could be one conclusive evidence in the case. The next question would be whether the behaviour is consistent with a trained pilot or not.
@HB: We need to find the FDR to better understand what occurred. Unless we believe that our interpretation of the final BFO values is wrong, or there was an extraordinary recovery from the descent followed by a long glide, the wreckage should be close to the 7th arc, with or without pilot input.
@Chris
In a comment you made several days ago you asked about Salah, the 5 daily prayers that Muslims recite. Interestingly the first prayer of the day happens at dawn (fajr), which occurs between the first light and sunrise. This is the first prayer time that would have occurred for mh370. This is also when mh370 went down. The person would face Mecca when praying which would have been to the northwest. The plane may have turned to face that direction.at that time.
@Ge Rijn
Your recent comments regarding IFE components are misleading to the less informed or casual observer. Several people have tried to coach you on this (any many subjects) multiple times, but you keep repeating the same misleading propositions without providing counter arguments supported by data.
The recovered MH370 IFE debris is NOT an IFE exterior decorative bezel. The MH370 IFE support frame debris is INTERIOR to the head rest and requires significant force to dislodge and deform vs the exterior bezels you observe in the Asiana Air 214 photos. Ignoring this (and other facts) repetitively is counterproductive and distracting to the core of positive dialogue.
If you can show that the recovered MH370 IFE debris is the exterior bezel please do so. If you can show an example of ejected and deformed Asiana Air 214 interior IFE support frame please do so. Until then, attempts to equate the two different components for the sake of underpinning your overall EOF theory remains invalid.
@Mick Gilbert
Item 22 is at the leading edge right behind the HF antenna cover:
https://imagr.eu/up/XFmmp_Cqz5FkaUIAAyMZV.jpg
On hitting water at high speeds:
http://www.askamathematician.com/2012/07/q-why-is-hitting-water-from-a-great-height-like-hitting-concrete/
@Kenyon
Take a look at picture 3, 4 and 5 in the link:
http://jeffwise.net/2016/06/09/blaine-alan-gibson-finds-3-possible-mh370-debris-pieces-in-madagascar/comment-page-2/
The piece is definitely a IFE exterior cover. The cloth hook proves this too. It’s clipped on the seat and not much force is needed to remove it.
I cann’t show an ejected IFE exterior cover but the interior pictures I posted show at least two of those covers seperated.
@Michel John
The commentator says the plane sank right after the last occupants were evacuated.
I guess it stayed afloat for ~20 minutes or so maybe a bit more. Just a guess.
@Ge Rijn
What do you mean with leading edge of a 777 vertical stabiliser?
The point of a ditching is to avoid hitting water at high speed, thus water is not the same as concrete.
@Ge Rijn
For goodness sake, read the official debris report! Item 22 is part of the right vertical stabilizer panel from between the auxiliary and front spar. It is not from the leading edge of the vertical stabiliser.
And read the article you’ve linked regarding water and concrete or one of the myriad of others including the Mythbusters episode dealing with the phenomenon. Water, although non-compressible and quite viscous, is a liquid that allows for displacement; this lowers the rate of deceleration. Concrete is a solid and does not accommodate displacement; deceleration is therefore essentially instantaneous.
And debris Item 11 is most assuredly not the exterior cover. Again, look at the official debris report! It is clear that the recovered component is exactly as Kenyon says, it is the seat back trim panel for encasing the IFE monitor; it is normally covered with upholstery (you can see the Velcro attachment points). The exterior frame fits into the recovered component (on 9M-MRO the exterior frames are dark grey, again, if you’d bothered looking at the official debris report you would see that).
Seabed Constructor has recommenced scanning on the NW side of the Primary Search Area presumably to widen the coverage to ensure the space out to 25 NM from the 7th Arc is completed.
The link below is to a map of the SC track in the search area to date.
The weather in the area is dominated by high pressure with good visibility, but there is a 21 knot wind and the combined swell and waves are 4.2 m high.
https://www.dropbox.com/s/uvws50x9okh681h/SC%20Track%2021022018.pdf?dl=0
@Ge Rijn writes “The piece is definitely a IFE exterior cover.”
Incorrect: the IFE display component recovered by Blaine Gibson is the seatback display chassis. It is firmly fixed to the seat frame. The chassis provides the mount for the LCD panel and enables the LCD panel to be tilted.
The visible part of the IFE seatback display, the frame around the screen, is the display bezel. The display bezel is a separate, cosmetic part of the installation.
The NTSB’s ‘Survivability Factors’ report catalogues 128 images of HL2264: not one* depicts a bezel or the display chassis detached from a seat frame.
(* considered those that are not fire damaged)
@Don Thompson
Take a look at picture 37. In the middle at the second row of seats the left seat is missing its encasing. At the right on the fourth row also the left seat IFE encasing has seperated. It’s a different lay-out of the encasing but the principal is the same. It’s only an encasing. The LCD-screen is not mounted onto it.
http://www.chinaaviationdaily.com/news/27/27594.html
The MOT debris report concludes therefore:
“The item is confirmed to be part of the seat back trim panel for encasing the IFE monitor.”
to add: at the fourth row in the middle, the middle seat shows a partially detached monitor encasing. Here you can see clearly it only surrounds the monitor but is not attached to it.
@Ge Rijn writes “The MOT debris report concludes therefore “The item is confirmed to be part of the seat back trim panel for encasing the IFE monitor.””
Key word: ‘part’.
The recovered MH370 item is the seatback display LCD fixture, you are mistakenly confusing/misattributing the seperate ‘bezel’ part to be the same item.
The China Aviation Daily image to which you refer shows that seat 41D is missing the BEZEL associated with the IFE seatback display fixture. However the LCD panel, its fixture, and the fabric cover, all remain in place.
The images presented in the NTSB report show the back of seat 41D reclined back on seat 42D. Therefore, the seat back has been disturbed from the condition found by the NTSB investigators, before the China Aviation Daily images were acquired.
Collating images, annotating, posting to a second site, is time consuming. I may get to that later so as to illustrate comprehensively.
@Trip
I do not include religious considerations in my flight paths, but I have used the word loiter or delay in talking about the fact that the distance between Arc5 and Arc6 is too close, depending how you model it. I feel twilight at around 22:51 and high winds aloft below 22S and sunrise and the 23:14 sat all are all potentially big events for a pilot looking out the window in any case. I currently envision descent and slow down after Arc5 to reduce visual profile and maybe to get out of the high winds.
If the current search area fails, then I will probably go back to see if I can make Ed Baker’s path work on BTO/BFO (~22S end point). He does envision a religious moment. I also have a non religious path to 22S, very similar to Ed’s proposal, based off Victors/Richard McMurdo path.
@TBill: I don’t see excessive error for the leg between the 5th and 6th handshakes if errors at each handshake are properly included and autothrottle modes are properly modeled.
Relative to path modeling, the one thing I am confident about is that any confidence about any particular path is unfounded. Modeling paths leading to impact sites along the 7th arc north of around 36S require relaxation of assumptions that are inherent in the DSTG Bayesian analysis, and with that relaxation of assumptions comes a broadening of the posterior PDF along the arc.
@Don Thompson
Re: “The recovered MH370 item is the seatback display LCD fixture, you are mistakenly confusing/misattributing the seperate ‘bezel’ part to be the same item.”
I’m sure not I but you are confusion the item to be the LCD fixture.
But if you can prove the item to be the LCD fixture this would be welcome and important for it would mean conclusive proof (to me) of a high speed impact.
To seperate the fixture and monitor would need tremendous impact forces. Seperating only the encasing (which it is imo and called by the MOT) could be done by hand force.
@Ge Rijn,
Take a few days off from posting, friend. Go for a run, read a book.
Sincerely,
A long time reader of this forum
@Sabine Lechtenfeld
@David
@Rob
In “normal” conditions which are a pilot with good intentions to ditch properly, the recommendations are the following:
Quote from Operating Manual and Skybrary:
“In ideal conditions (smooth water or very long swells) land into the wind. This will ensure the minimum possible touchdown speed and help minimise impact damage.
Where the swell is more marked, it may be advisable to ditch along the swell accepting a crosswind component and the higher touchdown speed, thus minimising the potential for nosing into the face of the rising swell.
The optimum attitude for touchdown in most aircraft types is approximately 10 degrees nose up.
End of quote.
On March 8th, 00:00Z (UTC) in the current search area the wind was coming from 220° at 15kn.
Thus moderate but with waves. The best direction to touch down would have been around 130° and if possible on the top of a wave. Basically this direction is South-East and orthogonal to the 7th Arc in this area.
Independently of the people in command’s intentions, one could imagine that due to the waves the tip of a wing touched first the water making the aircraft rotate and forcing the nose to plunge or maybe it could have been a nosing in a wave first. This would make the plane rotate very fast on a circle mostly in a vertical plane and centred on the nose in the water. Thus the satcom antenna (located towards the rear of the a/c) would move much faster explaining a negative BFO on the downwards part of the rotation.
I made some computations at the current search location and a BFO=-2Hz is included in the range of possible values using coherent Vref for ditching.
To my view this would also contribute to explain why so little debris. It is a kind of “slow” plunge compared to other hypotheses expressed here.
We will develop this in a new version of our analysis report (to come at http://www.mh370-captio.net).
@Jean-Luc Marchand: Due to symmetry, the sensitivity of the BFO to vertical speed is the same anywhere on the 7th arc. That means that the 184-Hz drop in measured BFO from 182 Hz to -2 Hz in 8 seconds represents a change in vertical speed of about -10,479 fpm anywhere on the arc. Over the 8-second interval, that corresponds to an average downward acceleration 0.68g. Without a doubt, the pilot was NOT performing a ditching during this time interval.
@Joel: Actually, I think we might have made some progress.
@Ge Rijn said: But if you can prove the item to be the LCD fixture this would be welcome and important for it would mean conclusive proof (to me) of a high speed impact.
Perhaps @Don Thompson or @Mick Gilbert can provide proof.
It reminds me of that line in the Blues Brothers…”I’ve seen the light!”
Perhaps that applies to @Rob’s sunrise thesis as well 😉
@Richard Cole believes that only 6 of the 8 AUVs are currently scanning the inner leg of the primary area. Perhaps there are ongoing problems with one or two of the AUVs.
@Paul Smithson
“It reminds me of that line in the Blues Brothers…”I’ve seen the light!”Perhaps that applies to @Rob’s sunrise thesis as well”
Paul, I saw the light a long time ago. 😉
@Victor,
“@Richard Cole believes that only 6 of the 8 AUVs are currently scanning the inner leg of the primary area.”
I’d agree with that. The launching was very messy, and it was only after the SC had followed them up, was it possible to identify the tracks leading back to all the launch points. Looks like the current width between tracks is about 1.45km.
Concerning AUVs & present mission: if the necessary swath can be achieved with 6 AUVs why expend the time involved in launch and recovery for 2 more?
@Don Thompson: Assuming all the AUVs are functioning, the decision to launch more than 6 would depend on the perceived value of the incremental data. The incremental time to launch and recover is much less than a mission duration of approximately 48 hours, so even if the data has diminishing value, it still might make sense.
6 or 8 AUVs at a time, it is still a staggering quantum jump in ocean bed search rate capability. I am in awe.
@Victor
You are right if the plane was falling. But such a dramatic change in 8 seconds can occur if the plane is in rotation because of nosing in the water for ex.
At 00:19:37Z, if the a/c was close to touch down at a ground speed of about 150kn, on track 130°, and if the a/c is nosing in the water it would rotates fast. Towards the end of its kind of flip over the antenna can reach instantaneous vertical speed of -18,000 fpm leading to a BFO of -58Hz according to Yap’s model. This means that the measured -2Hz is likely reachable in this hypothesis (at 200kn the vertical speed could reach -24,000fpm).
It means also that one has to consider that 8 sec earlier the plane was already at sea level. This would mean it has descended earlier. Gliding is a strong possibility.
@Jean-Luc
OK good…I was about to ask if your paper addressed the -2 BFO. I propose a somewhat similar flight end to you, except mine is happening at around 32.5S. Previous posts here I have showed a 123-135 deg approach to Arc7. Your work is helping me finish my unfinished path 180S CTH from ISIX.
Where does said glide end? on Arc7?
@Jean-Luc Marchand
Do please explain this ‘flip over’ in more detail. Over what period does this occur, is this a longitudinal or lateral ‘flip’, a flip to a complete inversion?
@Jean-Luc Marchand: Your explanation of the final BFO values in the context of a ditching requires:
1) A nose-down flip during a ditching, where a nose-up attitude certainly would be attempted.
2) The final log-on transmission coincided almost exactly with the moment of impact.
3) There was no loss in the required navigational data and antenna pointing capability after impact.
I don’t understand how you believe this explanation is even remotely possible.
@Ge Rijn
@Victor Iannello
@Ge Rijn said: But if you can prove the item to be the LCD fixture this would be welcome and important for it would mean conclusive proof (to me) of a high speed impact.
@Victor Iannello said: Perhaps @Don Thompson or @Mick Gilbert can provide proof.
Okey doke, I don’t know whether this constitutes ‘proof‘ but let me have a go at that.
For starters, let’s set aside the false analogy with Asiana flight 214; the crash airplane, HL7742, is fitted with different seats with different seat-back IFE components to the Malaysia Airlines B777-200ER fleet. Let’s instead look at a completely analogous airplane to 9M-MRO, let’s look at 9M-MRD, which as we know was blown out of the sky while operating as MH17.
Here is a photograph of the part in question, specifically the seat back trim panel for encasing the IFE monitor, still covered in the seat trim upholstery and still attached to the seat back; the LCD monitor has been detached. And here is a photograph of the same component this time with the monitor still attahced but sans seat. At this point you might reasonably be forming a view that the part in question, the seat back trim panel for encasing the IFE monitor, is somewhat robustly attahced to the seat-back.
If you were not, this photograph might get you over the line; it shows a pair of seats, the seat on the left still has the seat back trim panel attached and the monitor is still in place, the seat on the right has the seat back trim panel partially detached in such a fashion that you get a sense of the design of its attachment points and method of attachment. Again I’d suggest that it shows that the recovered part would have been somewhat robustly attahced to the seat-back on 9M-MRO.
If you are now left wondering do those parts ever come loose, well yes they do as evidenced here and here.
@Ge Rijn said: The piece is definitely an IFE exterior cover. (WRONG.) The cloth hook proves this too. (WRONG, ACTUALLY PROVES MH370 DEBRIS IS NOT THE EXTERIOR DECORATIVE BEZEL.) It’s clipped on the seat and not much force is needed to remove it. (HUH?)
@Ge Rijn said: I can’t show an ejected IFE exterior cover but the interior pictures I posted show at least two of those covers separated. (WRONG, SHOWS EXTERIOR DECORATIVE BEZELS POPPED OFF, INTERIOR SUPPORT FRAMES REMAIN IN TACT.)
@Ge Rijn said: But if you can prove the item to be the LCD fixture this would be welcome and important for it would mean conclusive proof (to me) of a high speed impact.
Ge Rijn perhaps the following linked PDF can shed light on your understanding of the various IFE frame and bezel components. Page 2 and 3 use photos from MH17 (B777) seat debris to convey identification of various IFE parts as they relate to the MH370 recovered, deformed, interior (aft) support frame. While studying, try to contemplate the mid air & ground impact forces involved with MH17 that rapidly shed 560 mph (900 kmh) of speed and the seats fell from 33,000 feet (10000m).
IFE interior aft support frame
I hope this is welcome to your ongoing understanding.
IFE interior aft support frame
Sorry I don’t know how to post an active link:
https://www.dropbox.com/s/xj9vd8qsa6130mh/IFE%20interior%20aft%20support%20frame%20.pdf?dl=0
@VictorI
@Jean-Luc Marchand: Your explanation of the final BFO values in the context of a ditching requires:
1) A nose-down flip during a ditching, where a nose-up attitude certainly would be attempted.
2) The final log-on transmission coincided almost exactly with the moment of impact.
3) There was no loss in the required navigational data and antenna pointing capability after impact.
I don’t understand how you believe this explanation is even remotely possible.
I had the exact same reaction, but did not comment. I felt like I must be misunderstanding the description.
@Ge Rijn: @Don Thompson, @Tom Kenyon, and @Mick Gilbert all have provided solid evidence that the MH370 part that Blaine Gibson found is NOT the IFE cover, but rather the IFE frame. By your own admission, this is proof of a high speed impact rather than a ditching.
@Dennis. Yes, and having come innow , this might be time for your prized “whacko”/”not-whacko” of the day award.
@ikr
I am overwhelmed. 🙂 I think it best to simply stay in the observer mode for the time being.
The Operational Search for MH370 (3 October 2017); page 101 has note 29 in relation to the final descent rates as follows:
“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.”
What if the track and speed labels were invalid?
@Pilatus
What if the track and speed labels were invalid?
What do we do if they were? Should we not try as JW suggests? Should we ignore them? It is the best we have at the moment. If they are invalid, we have nothing. At this moment all we can do is trust the data. There is no compelling reason not to do so.
Seabed Constructor has continued scanning on the NW side of the Primary Search Area and has reached the NW corner and presumably will start collecting the 6 AUVs that were deployed.
The link below is to a map of the SC track in the search area to date.
The weather in the area is dominated by high pressure but with reduced visibility, there is a 17 knot wind and the combined swell and waves are 3.4 m high.
https://www.dropbox.com/s/a7hlavykfc6eogi/SC%20Track%2022022018.pdf?dl=0
@Pilatus
What if the track and speed labels were invalid?
Should we not rather consider what happens WHEN they are invalid, since default expectation is that pitot heat is gone? What do we think an unheated (for ~2 mins) pitot might show speed to be and how might that affect the doppler compensation/BFO observed?
@Paul wrote “Should we not rather consider what happens WHEN they are invalid”
Concerning the SDU, it gets 7 data words concerning position and attitude: Latitude, Longitude, Ground Speed, Track Angle, True Heading, Pitch, and Roll
An interesting question, however, none of these data words are derived from air data. Therefore, the SDU’s calculation of the frequency compensation should be unaffected by loss/degradation of air data.
@Mick Gilbert @Kenyon @VictorI @ALSM
Thanks a lot for those MH17 pictures. They allow me to have more clear impressions. I’ll take some more time to reply in detail on the subject.
I would appreciate it if @ALSM would take a closer look too.
You will understand I still have my doubts but more on this later in detail.
@Paul Smithson
RE: “Should we not rather consider what happens WHEN they are invalid, since default expectation is that pitot heat is gone?”
Don already mentioned that the data words are not derived from air data. Nevertheless, the loss of pitot heat is a bit of a moot point, because it would be restored as soon as the APU began generating AC power.
@Joel
Thanks for your concern.
Don’t worry, I’m doing fine.
@Ge Rijn said: You will understand I still have my doubts but more on this later in detail.
At this point, we can be sure that the piece found by Blaine Gibson is the IFE frame and not the cover. You said previously that this would imply a high speed impact. Whether or not you wish to now back off this claim, it’s time to incorporate the identification of this piece into your theory and move forward.
@Mick Gilbert @Kenyon @Victorl @Don Thompson
My statement to @Don Thompson was:
“But if you can prove the item to be the LCD fixture this would be welcome and important for it would mean conclusive proof (to me) of a high speed impact.”
@Mick Gilbert’s link to this picture:
https://www.flickr.com/photos/jeroenakkermans/15207775193/in/album-72157649358752251/
clearly shows the found item is not the LCD-monitor fixture. The monitor is attached on a metal fixure by pivots. This metal monitor fixture-frame is attached in the seat headrest with screws or rivets (last photo in @Kenyon’s dropbox-link).
The item is only the monitor encasing (and the bezel is an integrated part of it) which is fitted on two rails which probably allow it to pivot the encasing sidewards to allow servicing to the monitor and/or easy removal of the encasing by sliding it off (see also last photo in @Kenyon’s dropbox-link).
The attachment of the encasing to those rails is thin plastic on plastic. Not a strong attachment either I would say. Subjected to breaking easily.
Then I see this encasing consists of two shells put together.
The found item is the outer shell.
This leaves the question how this outer shell seperated from it’s inner shell while both shells were covered by the same fabric.
I assume the fabric detoriorated over time while drifting in the ocean and after the fabric was gone sufficiently the two shells seperated from eachother.
To summarize; the found item is not the LCD-monitor fixture and I still see no proof of a high speed impact in this piece.
Another argument to consider in this regard imo is that the MH17 seats fell out of the sky and probably had an impact speed of less than ~150mph. Which would be comparable to a ditch-like impact.
@VictorI @Others
Please first consider my argumentation seriously.
@Ge Rijn: Your continually shifting logic is exasperating to many of us. There is no sin in modifying your theory over time. In fact, that shows objectivity.
@VictorI
I got fiercly attacked on my statement this item found by Blaine is not the monitor-fixture but only the encasing.
I proved with arguments I was right. If those who attacked me on this this way are also man enough to say so, I have no problems at all. It’s all complicated I understand.
If not they have to come with better arguments to convince me.
If they keep silent I’ll know where I stand and where they stand.
I’ll await their reaction.
If someone shows conclusive proof to me of a high speed impact I would be glad to see it and right away change my assumptions.
@Ge Rijn: Nobody attacked you. You made statements that were wrong. Several contributors factually corrected you. You are stubborn beyond belief. Unless you have new evidence, it’s time to move on.
@Ge Rijn wrote “The item is only the monitor encasing (and the bezel is an integrated part of it) which is fitted on two rails which probably allow it to […]”
It is certainly not “fitted on two rails“. It is located with two spring-clips, the spring-clip fixings are then locked by means of the vertical bars.
I’m gradually compiling a detailed description, your impatience is just another deficiency.
@Don Thompson
I saw a spring-clip. This was my earlier explanation of the attachement of the encasing. Which would explain an even easier way of detaching with little force needed.
I’m ofcourse eager to see you more detailed description.
And I’m not impaciente. Take your time.
And please leave the concept of deficient projection out of the matter.
Just objective arguments.
@Don Thompson, @Tom Kenyon, @Mick Gilbert, et al: You are going to have your hands full when Larry Vance publishes his book with his interpretation of the evidence that shows the MH370 was ditched.
@ALSM
I would really like to hear your opinion on this subject.
You maybe don’t want to get your fingers burnt but your opinion would mean a lot to me.
Just to keep Ge Rijn waiting: [MODE B MOVIE ON] The screen case might have been ripped off the seat back when Z climbed onto the seat back to open the hatch in the ceiling giving access to the flight recorders. [MODE SUPER B MOVIE ON] And when Z was gliding the plane after recovering from the steep dive, both recorders were thrown by an opened cockpit window many miles before the plane came in the sea: Perhaps the March ’14 pings were from the recorders? And they were missed because there was nothing else there big enough to be seen?
@Don, @Andrew.
I’m definitely out of my depth on data words, source of these and the chain through which they pass so I am glad to be “schooled” on this 🙂
There was a lengthy prior discussion that I think demonstrated that the speed/track data are not coming from GPS but from ADIRU. As we all know, GPS can produce anomalous velocities during boot as more satellites are acquired and the position is settling. If that is all irrelevant because there is no direct GPS feed (only ADIRU), then where does the ADIRU get its data by default and in backup in case of invalid data from primary source? Is the airspeed (and a sometime unheated pitot) not implicated in any way here?
I should declare that I remain highly skeptical about the BFO and this skepticism has only been reinforced as expert opinion mulls over how difficult it is to match the timing and magnitude of downward acceleration with the control state expected.
@ Ge Rijn said: @Mick Gilbert’s link to this picture:
https://www.flickr.com/photos/jeroenakkermans/15207775193/in/album-72157649358752251/
clearly shows the found item is not the LCD-monitor fixture. The monitor is attached on a metal fixure by pivots. This metal monitor fixture-frame is attached in the seat headrest with screws or rivets (last photo in @Kenyon’s dropbox-link).
Ge Rijin you are really struggling to observe features in photos amongst other issues. Each time you respond you make an incorrect observation and then equally incorrect conclusion immediate follows. Your posts can and will confuse casual observers\followers of Victor’s blog. I only write in response to your incorrect observations and subsequent conclusions to help the casual observer, you personally appear to be unreachable with facts, data and logic. If you catch on to the rebuttals that would be a bonus for your “efforts” and for all having to put up with these tiresome attempts to get through to you.
There are multiple subcomponents to the IFE system. I conclude by your recent statements you are virtually unable to accurately envision how and why mechanical subcomponents fit together.
One more piece of advice, study how plastic components temporarily deform, break, and most importantly permanently deform. If you are able to grasp the basics you will be able to move on regarding the IFE internal frame.
I’m tapping out on the IFE discussion with you (I tried) but reserve the right to chime in the next time you offer misleading observations on other subjects. Again I’ll chime in for the benefit of the casual observer/follower of MH370 so they don’t get side tracked by wrong information/conclusions.
Tom
@Kenyon
This is no logical responce, just bullshit. You know I’m basically right but refuse to admitt it.
Factual arguments please. No side-walk distractions.
@Ge Rijn: For some time, a number of people have privately asked me to ban you. I’ve defended you, attributing your comments more to stubbornness and confusion rather than malice. However, with this last series of comments, it is becoming more difficult to defend your comments. There is no conspiracy to attack you. Those that have responded to you have provided facts that dispute your claims, which seem to evolve over time.
My advice is to take a break, as was earlier suggested, before you say something you don’t really mean.
@Paul Smithson
“I should declare that I remain highly skeptical about the BFO…”
Paul- are you saying all BFO or just 00:19 BFO’s in question? I think you are saying 00:19 BFO’s may be iffy, which is my simple interpretation of what Jean-Luc is suggesting.
@TBill: No, that’s not what Jean-Luc is suggesting. He is proposing that the high descent rate suggested by the final BFO value was caused by the plane flipping.
@TBill – Victor is right. I’m talking about the 0019 BFO.
@Paul asked “where does the ADIRU get its data by default and in backup in case of invalid data from primary source”
In the case of ground speed, track angle and heading, the ADIRS processing relies on its gyros and accelerometers and the velocities integration function for these data.
ADIRS is reliant on initialisation while the aircraft is stationary, on the ground, usually taking a GPS derived position as reference but a manually entered gate position is equally valid.
@Victor
I agree with what you said, but in my mind the net effect is questioning what the final BFO represents. Seems to me rapid descent is a reasonable leading proposal.
The overall question is trying to define how a proposed flight path needs to end to be considered a complete and valid path proposal. Does the proposal have to include -2 BFO as part of the flight, or can the -2 optionally be argued away somehow? Also if we accept -2 BFO, what altitude constraints does that put on the end-of-flight? I am thinking maybe even a descent from 5000-ft could meet the -2 BFO, so that is not too restrictive.
I am assuming Jean-Luc is having difficulty fitting the -2 BFO into their scenario, if so logic dictates the MH370-Captio.net may need to discount the final BFO. That is the nature of developing an intentional flight path, one has to fit the data or else come up with an alternate explanation. The alternate explanation can be considered a new vision (only if correct), or unlikely, in which case the scenario probably needs rework.
@Don
One of the most interesting mems applications I was ever exposed to was demonstrated to me by a couple of guys in a San Francisco startup. They developed tiny inertial systems that could be mounted on the top of a first responder’s shoes (about 25mm square). The problem with backpack systems is drift without the ability to do any zero velocity updates. In a shoe, each step ends with zero velocity, and an update can be performed. Your path through space then becomes a series of step vectors without the inherent drift associated with the backpack approach. Their demos were incredibly impressive tracking users inside buildings, up and down stair wells, and so on. It never made it to market for some reason. Not unlike a lot of things I think are cool. 🙂
One of the inherent advantages of factory installed navigation systems is that we do include an accelerometer (to allow velocity integration), a rate gyro (for turns), and an odometer (distance) input. With an aftermarket Garmin you are lost entering a parking garage. With the added components you are golden. You can return to your parked car, enter a destination, and go. When you get to the exit of the parking garage you are told the correct way to turn. With the aftermarket Garmin like systems you are still “searching for satellites”.
@Paul Smithson
RE: “There was a lengthy prior discussion that I think demonstrated that the speed/track data are not coming from GPS but from ADIRU. As we all know, GPS can produce anomalous velocities during boot as more satellites are acquired and the position is settling. If that is all irrelevant because there is no direct GPS feed (only ADIRU), then where does the ADIRU get its data by default and in backup in case of invalid data from primary source? Is the airspeed (and a sometime unheated pitot) not implicated in any way here?”
Following on from Don’s comments above, the ADIRU has two separate functions: air data and inertial reference data. The SDU only uses the ADIRU’s inertial reference data. The inertial reference data is not affected if the air data is invalid because of a pitot heat failure. As Don mentioned, the ADIRU is initialised with a known position while the aircraft is stationary on the ground. Once it is initialised, the inertial side of the ADIRU is essentially a self-contained unit that does not require data from other sources.
Thanks, chaps
@Ge Rijn
You seem to be attempting to move the discussion away from your initial patently false analogy (specifically, the contention that the detachment of an external seat-back IFE monitor frame from the crash of Asiana flight 214 is evidence that part of the seat back trim panel for encasing the IFE monitor could be detached in a ditching) to yet another patently false analogy; ‘… the MH17 seats fell out of the sky and probably had an impact speed of less than ~150mph. Which would be comparable to a ditch-like impact.‘
That is complete, utter and errant nonsense! Once again, and just as in your ‘water equals concrete’ patently false analogy (are you seeing a pattern here?), the important consideration is the rate of deceleration. Those seats from MH17, falling at around 130-150 mph, were decelerated to a stop almost instantaneously when they hit the ground. In a ‘iditch-like impact‘ the deceleration is most assuredly not instantaneous, deceleration occurs over a short period of time; even an unsuccessful ditching like Ethiopian Airlines flight 961 has a deceleration period of a good 6-7 seconds. In short, they are not comparable events.
Further, you should note that despite the impact forces experienced by the MH17 seats, in most of the recorded images the seat back trim panel did not separate from the seat back. Moreover, on the recorded occasions when the panel did separate from the seat back there is no evidence of it splitting into its constituent front and rear parts. Accordingly, even if the impact forces were analogous to a ditching (which they are not) they were still insufficient to cause damage that is even remotely similar to what is required to produce debris item 11.
You therefore have no evidence at all to support your contention that debris item 11 is in any way consistent with a ‘iditch-like impact‘. So in terms of who is waiting on whom to produce convincing evidence, the burden now rests with you to demonstrate that a ‘iditch-like impact‘ could liberate debris item 11.
Mmm … ‘iditch’ is not an iPhone app, that was me ballsing up the italicisation.
@Mick
Problem solving sessions represent an interesting study of human behavior. In my working life I preferred a no holds barred open attitude where direct reports could freely tell me I was full of shit (and they often did). You have to step away from your ego, and the normal desire to be right, and just let the ideas fly. It really does work.
Having said that I am on the brink of re-evangelizing my CI scenario (just kidding Victor). 🙂
@all
There is a high likelihood the terminus is at ~30S near the 7th arc. My geometric algebra work (which no one else has commented on) supports that location. Integrated Doppler calculations support that location (again weakly). Richard’s drift studies support that location. You heard it here.
@Dennis,
“…You have to step away…”
Agreed. Decided to do that after a bunch of dimwits tried to demonstrate their massive egos on the other blog. What was I thinking, commenting there? Hopefully I can resist the temptation to comment there again.
@Mick Gilbert
I cannot demonstrate a ditch-impact could liberate item 11 although I wished I could enter a B777 and pull an encasing off.
And I’m not steering away from the issue.
The issue was you all agreed the item was the monitor-fixture which I demonstrated it is not.
And I assumed a high speed (nose down) impact would be necessary to seperate the monitor-fixture. In that case I would except this as evidence of a high speed impact. Item 11 is not the monitor fixture but the encasing which is clipped onto the fixture with spring-clips (@Don Thompson) so it cannot serve as evidence to me for a high speed impact. That’s all.
And I fully recognise there’s no conclusive evidence of a ditch-like impact too. Only some clear indications imo. Like there are also indications of a high speed impact but no conclusive evidence yet also.
@Ge Rijn
‘The issue was you all agreed the item was the monitor-fixture …‘
I most assuredly did nothing of the sort, I referred to item 11 by the description used in the debris report, viz ‘part of the seat back trim panel for encasing the IFE monitor’.
@DennisW
G’day Dennis,
I’m not particularly invested in being right or even being seen to be or thought of as being right (given the number of times I have by my own admission been wrong that would represent a somewhat poor investment given my position as an inside trader). What I do have a problem with is unsubstantiated conjecture parading as fact and false analogies being trotted out as reasoned analysis, more so when silence is often deemed to be agreement. The airplane may well have been ditched by an active pilot or there may have been an unpiloted low angle of attack, low speed impact with the water but I’ve yet to see anything that strongly supports either of those contentions and there’s plenty about that tends to contradict them.
I don’t disagree that problem solving, in particular idea generation, benefits from unconstrained thinking. That’s all well and good but at some point you’ve got to put those ideas through the filter of practicality. While there may be ‘no dumb ideas’ from the perspective of creative thinking, when it comes to practical and implementable solutions we all know that there is most assuredly a hierarchy of ranging from just plain bad to excellent.
Forums such as this can be incubators for ideas but shouldn’t they also be crucibles where ideas are tested? I am not a scientist but I am an advocate of the scientific method. By all means float ideas as ‘what if?’s but if you’re going to advocate them as ‘this must be it’ hypotheses surely you should have verifiable objective evidence and reasoned thinking to support them. ‘IMO’ and flawed or false analogies just don’t cut it in that regard.
I’d hazard a guess that while the atmosphere you fostered with employees was one where they felt okay telling you you’re full of shit, for it to be a productive discussion best they have a better argument to support that observation than simply ‘IMO’.
And just by the bye, I like 30S too (and have done since about this time last year).
@Victor, Don Thompson. A change of subject. More on search width.
A summary of the two pages below is that establishing/verifying the maximum time for IFE initial connection would rule a post 7th arc transit beyond 25 NM in or out, IMO.
https://www.dropbox.com/s/sql9d62h280u8eq/IFE%20connection%20importance%20to%2025%20NM%20search%20width.docx?dl=0
@DennisW
“There is a high likelihood the terminus is at ~30S near the 7th arc. My geometric algebra work (which no one else has commented on) supports that location. Integrated Doppler calculations support that location (again weakly). Richard’s drift studies support that location. You heard it here.”
There you go Dennis, evangelising again. “You heard it here”. Agree with that at least. 😇
@DennisW
I am in full agreement with your statement. The most likely MH370 End Point is 30S +/- 1 degree. There are even more reasons that have already been given on this website.
@Rob
We are not seeking to convert anyone to Christianity! We are sharing the results of our work in analysing the most likely MH370 End Point. I appreciate you have a different view. However, because Dennis or I disagree with you, that does automatically qualify us as an evangelist. Billy Graham, who died earlier this week, was an Evangelist. This is a calling, which goes beyond being a commentator on Victor’s excellent website.
@David
Thank you…that IFE was on my mind but I did not want to ask. I got lost on whether or not we were expecting further sat com signals after 00:19:37.
@DennisW
Wake me up when we get to 32.5S or thereabouts.
@Richard Godfrey
I agree with you at least on one thing. Yes this is an excellent website. A unique and vauyabe resource. I very much hope OI realise that as well.
@Rob
There you go Dennis, evangelising again. “You heard it here”. Agree with that at least. 😇
Yes, sorry about that. Tax time always disrupts my temperment. I’ll be OK again in a month or so.
Hello all, as I am not permanently logged in, I have some delay in answering 🙂
@TBill
1) “Where does said glide end? on Arc7? In the described “flip-over” type”
yes on Arc7.
2) “00:19 BFO’s may be iffy, which is my simple interpretation of what Jean-Luc is suggesting”
In our current analysis report, the last BTO/BFO are not deeply analysed as we considered them to be unreliable.
3) “I am assuming Jean-Luc is having difficulty fitting the -2 BFO into their scenario, if so logic dictates the MH370-Captio.net may need to discount the final BFO. That is the nature of developing an intentional flight path, one has to fit the data or else come up with an alternate explanation. The alternate explanation can be considered a new vision (only if correct), or unlikely, in which case the scenario probably needs rework.”
Indeed, in our analysis we are “reading” the data as they are and derive configurations that fit. (cf also my answer to Victor below). As we suppose that there was human intervention the set of “possibles” is large and we needed time to look at all of them.
@Don Thompson
“Do please explain this ‘flip over’ in more detail. Over what period does this occur, is this a longitudinal or lateral ‘flip’, a flip to a complete inversion?” :
This is currently under analysis. In fact all depends on the heading and if the people in command (PIC) flew headwind, sidewind etc… this conditions the type of flip-over offering a direct line of sight to the satellite. For example, if the PIC followed the recommended procedure then it must have been on heading ~130°/135°, and thus the flip-over is probably trigered by the right wing or engine touching first thus making the a/c pivoting on the sphere centered around that touch point followed by a nosing in the water. The duration is between 1 and few seconds depending on what touches first.
@Victor Iannello
@DennisW
“1) A nose-down flip during a ditching, where a nose-up attitude certainly would be attempted.” :
Definitely, but as I was born a sailer man especially on Hobby Cat, which is a boat designed for flat Hawaian lagoons, I have so many time experienced wave encounters in the Channel (in Britanny) which instantenously led to a fast flip-over.
An aircraft is similar to my boat, designed for calm air but when it touches the water, it becomes a boat constrained by the physics of the water out of any control.
All of this to say that when encounting wave and swell everything is possible due to the cinetic energy of both the a/c and the water. If the a/c does touch down nicely, the simple rebound is dependent of local physionomy of the wave/swell that will unbalance the a/c and it will plunge or break. The middle of the ocean is completely different from the nice river conditions as for Sully. A simple wave/swell of 1 meter high is “moving” tons of water. Under those local meteo conditions, 1 to 1.5meter height was most likely in that part of the ocean (current search area).
“2) The final log-on transmission coincided almost exactly with the moment of impact.”
Yes. But, when reading the larger BTO, I am not far from thinking that it is linked to the increased constraint to maintain the beam pointing so … (but this is a guess only).
“3) There was no loss in the required navigational data and antenna pointing capability after impact.”
Yes, it depends on the impact type and its violence and whether the electronic bay is not instantaneously damaged or not.
@Jean-Luc Marchand said: But, when reading the larger BTO, I am not far from thinking that it is linked to the increased constraint to maintain the beam pointing so … (but this is a guess only).
The second BTO value, after adjustment for missed timeslots, is very close to the first. It should be adjusted to 18,387 μs, as explained previously.
@Jean-Luc Marchand. At an antenna 13,800 fpm ROD (the Holland minimum) the aircraft CG would be travelling at about 9,200 fpm. If the original speed was 200 knots say, the aircraft angular momentum would be somewhat less than half the previous linear. That angular speed, averaged with nil at the beginning yields a time to complete that first half a revolution; in this case 4 secs. From that, impulse from the sea to reduce momentum that amount in 4 secs would require a force of 500,000 lbf (a round figure, but just by chance).
That equates to 1.42 negative ‘g’ horizontally for an aircraft of 350,000 lbs mass, no fuel. The peak total ‘g’ including the initial vertical element would be more but disregarding that, the mountings, wiring, APU and sensors might survive. However I think the black boxes would be rendered insensible as @Victor has said.
Still even putting that aside, a horizontal force of 500,000 lbf on the nose would not be resisted by that surely, even disregarding the vertical. If much of that horizontal force came initially from the engines and wings, with the average deceleration being 1.42 ‘g’ the initial peak would be higher of course, probably markedly. That would of itself then worry the mountings etc.
In any case I doubt the SDU and its companions sensibly would be operative.
One other thought is that the ROD 8 secs before was about 48 fps (Holland again). Sullenberger’s was 12.5 fps at ditching. It used to be that carrier undercarriages were designed for 15 fps, their landings being controlled crashes, with no rounding to speak of. (Maybe they are still). He would have had to round out in four seconds or the aircraft would have hit the water very hard, the fuselage breaking most likely.
@Mick Gilbert
RE: “@Ge Rijn
‘The issue was you all agreed the item was the monitor-fixture …‘
I most assuredly did nothing of the sort, I referred to item 11 by the description used in the debris report, viz ‘part of the seat back trim panel for encasing the IFE monitor’.”
Yes, you did use this refference.
I’m still awaiting a reaction from @Kenyon, @Don Thompson and @VictorI on the issue.
@Ge Rijn: My reaction is that @Kenyon, @Don Thompson, and @Mick Gilbert have all told you the same thing using different words. I have no idea how you can look at the evidence that was assembled and presented to you and come to the conclusion you did. Others can decide for themselves.
@VictorI
Let me repeat:
@Don Thompson:
“The recovered MH370 item is the seatback display LCD fixture, you are mistakenly confusing/misattributing the seperate ‘bezel’ part to be the same item.” Then: ” It’s certainly a mild trolling behaviour, it is provocative and anti-social.”
And then I won’t have to feel attacked?
@Kenyon:
” (me)The piece is definitely an IFE exterior cover. (WRONG.) (me)The cloth hook proves this too. (WRONG, ACTUALLY PROVES MH370 DEBRIS IS NOT THE EXTERIOR DECORATIVE BEZEL.) (me)It’s clipped on the seat and not much force is needed to remove it. (HUH?).
The recovered MH370 IFE debris is NOT an IFE exterior decorative bezel.”
And you have no idea what I’m talking about?
I expect @Don Thompson and @Kenyon to respond also.
@Ge Rijn: Perhaps your choice of words is causing confusion. You relate Item 11 from the debris report to damage to the passenger compartment of Asiana 214. Exactly which seat back suffered damage to the IFE assembly in a way that it would produce a part resembling Item 11? An image with a circle around the specific seat back would be helpful.
@VictorI
I pointed this out to @Don. He named seat 41D. The left seat in the middle second row on the image shows his monitor-encasing has seperated. The fourth middle row shows a partially seperated bazel/encasing with the monitor still in place also.
Then on the right fourth row another encasing has seperated completely also with the monitor left in place. The exterior-cover has a different layout in Asiana 214 but the same principal. I cann’t put circles but take a look, it’s not that hard to see the similariy:
http://pic.carnoc.com/file/130717/tj_1307170220564.jpg
@Ge Rijn: And which seat do you believe shows the equivalent to Item 11 either missing or damaged?
@David
In your recent IFE paper, where did you get your end-of-flight speed estimates? If you are thinking the final descent could be pilot input, the implication is that there could be pilot input (slow down/descent) before that also. Not to mention a possible implication of Broken Ridge target (if the descent was pilot-initiated).
@VictorI
The left seat in the middle second row as a starter. The others as proof too those encasings can seperate in a low speed, low AoA nose-up, crash/ditch landing.
@Ge Rijn: In the Asiana 214 photo, I see evidence that some of the plastic cover plates to the monitors popped off. That’s not the same piece as Item 11 from the debris report, if that’s your claim.
@VictorI
It’s not the same lay-out but it’s essentially the same kind of piece I’m sure. I await @Don and @Kenyon’s about it.
@Ge Rijn: You have already heard from @Don Thompson, @Kenyon, and @Mick Gilbert. Independent of the Asiana 214 photo (which I think you will not reach agreement on because the styles of monitors is different, and you refuse to acknowledge that the missing plates are decorative and simply popped off), the MH17 photos show how much imparted energy is required to damage and dislodge Item 11.
@VictorI
At first this was not the issue debated. It was about item 11 being the monitor-fixture or the encasing/cover. It’s the encasing attached with only spring-clips. Easily seperated with little force needed.
Seats and other parts from MH17 fell down from high altitude swirling to the ground. This was not a high speed nose down impact by any means.
Those seats impacted the ground with probably no more than ~120mph vertical impact. No wonder some monitor-fixtures got seperated.
But in this case we only have the MH370 encasing-piece and that’s much easier to seperate with much lesser force. Like in Asiana 214.
@Ge Rijn,
Are you for real: you can’t download an image from a website, annotate it, in order to illustrate your argument? Yet, you expect others to expend effort to explain what you’re missing.
@Ge Rijn: If you think the MH17 evidence shows that Item 11 will separate with little force, I give up.
@TBill. “In your recent IFE paper, where did you get your end-of-flight speed estimates?”
The speeds came from an earlier post, below. The key assumption is the initial 350 KTAS (220 KIAS at 30,000ft). The speeds and descents I arrived at will vary with that.
https://www.dropbox.com/s/69717uzs7dfihum/Piloted%20final%20descent.docx?dl=0
I note their independence from whether piloted or not. I think it likely that an equation for exact answers could be developed but thought iterating would do, output feeding back into input to arrive at coherence.
“…..there could be pilot input (slow down/descent) before that also”.
Just the entry TAS would affect the speed/descent angle combination. If a slow down was not immediately before the descent I gather from others he may not reach the 7th arc in time. There is no height dependence except in IAS/TAS conversion.
“Not to mention a possible implication of Broken Ridge target (if the descent was pilot-initiated)”. Yes, well I cannot join in with the notion of him having a specific target, which would necessarily have him active at the end, (leaving aside a zero altitude waypoint!) Maybe a year ago I participated with others in reviewing whether his fuel load was reasonable for a flight to Beijing, given the weather predicted at his destination and at diversion alternatives. To me it was entirely consistent.
If it was the captain who did it I take to @Donald’s description of the events of that day precipitating something already in his mind. Maybe it had been a personal protest and maybe he had gone over it to some degree but the selection of this flight looks spontaneous (ie that day, maybe that moment) rather than him having selected and planned around it, IMO. In other words there was no fuel planning to arrive at a specific site.
Had he spontaneously chosen a reachable site or some other objective like sunrise there would have been the uncertainty of what actual weather he would encounter and other eventualities so he would have arrived with reserves. In that case he would have shut down the engines (APU coming on) and ended with fuel remaining, or jettisoned. Neither is supported by fuel consumption estimates to the 7th arc, though I do understand these are assumption dependent. In general though he would have planned and ended up further north east. That is yet another implication of there having been an active pilot, IMO, and important one.
My narratives have too much fuel, or maybe wind, I know.
‘‘This blog is informative.” But also becoming a bit goofy. Their is additional radar data out there.
@VictorI
No. A vertical impact in the ground (or water) with ~120mph would generate big forces. MH17 pictures show also a complete monitor fixture seperated (probably) on impact with the ground. But MH17 cannot be compared with Asiana 214 in this regard. MH17 was a mid-air break-up and pieces like seats fell from high altititude impacting with high vertical speeds.
I don’t believe MH370 had a mid-air break-up. I only argue that item 11 which is the clipped-on monitor encasing needs not much force to remove or seperate. Altough the Asiana 214 monitor-encasings have a different lay-out I think those are clipped-on too and the pictures show those encasings can seperate in a relatively low speed, nose up crash-landing.
@David
OK thank you, I see the speed is during the descent so picking up speed. In my path vision, as for initial conditions, the pilot has come down to FL100-FL150 and lower speed (perhaps 280-300 TAS or less I suppose at 00:19 per below).
Arc6-7 Crossing Speed @ 94E
180S=441 avg TAS
135S=295-315 avg TAS
Therefore in order to match Arc7 timing, the heading has also been changed to closer to about 135 deg, which is the minimum distance between Arc6 and Arc7 (whereas an unpiloted nominal 180S path does need relatively high speed to cut across the Arcs6-7).
I am not sure where your study stands on active pilot to make the descent happen per the BFO’s, but if the pilot is active, then maybe some slower speeds are possible.
Basically, although Jean-Luc and I have different end points (by half a continent or so), we are both envisioning a different style of ending – slower speed and lower altitude approach via active pilot.
@TBill
Basically, although Jean-Luc and I have different end points (by half a continent or so), we are both envisioning a different style of ending – slower speed and lower altitude approach via active pilot.
A terminus farther North (say around CI) does result in less satellite induced Doppler at 00:19:XX, but not enough less to have a major impact on the need for a rapid descent to reconcile the BFO data. The difference is maybe 35-40 Hz or so. Jean-Luc either needs to adopt the rapid descent, or discard those BFO’s somehow. I am not at all in love with his rotation explanation.
@Andrew @GuardedDon
“Following on from Don’s comments above, the ADIRU has two separate functions: air data and inertial reference data. The SDU only uses the ADIRU’s inertial reference data. The inertial reference data is not affected if the air data is invalid because of a pitot heat failure. As Don mentioned, the ADIRU is initialised with a known position while the aircraft is stationary on the ground. Once it is initialised, the inertial side of the ADIRU is essentially a self-contained unit that does not require data from other sources.”
How is the ADIRU affected by a complete power loss? Does its intertial side initialization from on ground hold? Does it auto restore like a crashed MS Word? Or does it re-initialize or default somehow? Not my wheelhouse (caveat) but seems critical when we are assuming full power loss and reboot of some or all systems.
@Ge Rijn
Re: ‘But MH17 cannot be compared with Asiana 214 in this regard.‘
Again, you’re missing the salient point; impact damage from the crash of Asiana flight 214 is irrelevant. The crash airplane, HL7742, was fitted with different seats with different seat-back IFE components to the Malaysia Airlines B777-200ER fleet. It is a false analogy.
@ErikN
RE: “How is the ADIRU affected by a complete power loss?”
That depends on what you mean by ‘complete’ power loss. The ADIRU gets 28V DC power from the L & R DC buses and the hot battery bus, so in the worst case the ADIRU will be powered by the aircraft’s main battery until the battery runs flat. In the short term, the loss of all AC power does not affect the inertial side of the ADIRU. If ALL power is lost, including the battery, then the ADIRU is inoperative and will not ‘reboot’ if power is subsequently restored. It can only be realigned when the aircraft is stationary on the ground. However, that is not the scenario that was being discussed.
@TBill. “… if the pilot is active, then maybe some slower speeds are possible.”
At 290 KTAS, initial descent angle would be over 5˚ for minimum ROD, over 30˚ for the maximum of 14,800 fpm.
8 secs later for minimum ROD the descent angle would be 25˚, speed 320 KTAS; for the maximum ROD, the descent angle would be over 42˚, speed 372 KTAS.
@Mick Gilbert
Asiana flight 214 was fitted with the same monitors as MH370. So I think it’s save to assume the monitor-fixtures were of the same type too and the decorative encasings differ only in lay-out.
But I propose to leave it here. The issue started with @Don Thompson and @Kenyon stating firmly item 11 was the monitor-fixture. I showed it was only the encasing. Now we disagree on the forces needed to seperate this encasing. I believe analogy with Asiana flight 214 can be relevant regarding item 11 and the forces needed.
I only conclude (for now) item 11 can not serve as conclusive evidence of a high speed nose-down impact.
to add.. but also not as evidence for a ditching.
@Ge Rijn wrote “Take a look at the Asiana 214 interior pictures after the crash landing. They speak for themselves (easy to find on the net). Several screen mountings seperated and numerous interior panels.”
That comment didn’t age well (when contradicted by the “I showed it was only the encasing.“)
@Ge Rijn: It’s interesting that I had not decided whether or not Item 11 was evidence of a high-energy impact. I simply had not done enough investigation to have a firm opinion. However, after looking at what you, @Kenyon, @Don Thompson, and @Mick Gilbert have presented in the last couple of days, I am firmly in the camp of Item 11 being consistent with a high-energy impact. Nobody is going to change your mind, and it is futile to try. However, the data has been neatly assembled for anybody to decide for themselves.
@David @DennisW
“8 secs later for minimum ROD the descent angle would be 25˚, speed 320 KTAS; for the maximum ROD, the descent angle would be over 42˚, speed 372 KTAS.”
OK? I see no immediate math problem with an intentional dive at the end. The initial descent (to cloud level in my vision) does not apparently lead to a soft crash, you are saying.
@Andrew
Understood (I believe) that you were specifically addressing whether power loss to heaters on pitot tubes leading to bad air data would affect SDU. Excellent information. My f/u question was (I think) for similar reasons as to how the SDU might be affected by bad data. It sounds like a loss of ALL power would impact its function. Perhaps someone could analyze a bit if ALL power loss occurred long enough to knock out the ADIRU (the inertial data side of which cannot ‘reboot’ in air) how the SDU would be affected and whether this might affect our end of flight data. Maybe I am way off my rocker (not my area by a stretch) but this whole situation caught my eye. Could this have occurred? And how might it have affected the 0019 BFO?
Cf. @Paul
@ErikN
The ADiRU was functioning normally when the 00:19 Log On took place: AES identified itself as Class 3 capable. To do so required that the AES received ‘good’ ADIRU sourced data.
@Don Thompson
I still did not see your: “I’m gradually compiling a detailed description, …”
I understand.
It’s a decided non-issue now handily avoided by you like a man of character by putting me down without thinking first.
Let’s move on.
@Ge Rijn
‘Asiana flight 214 was fitted with the same monitors as MH370.‘
That is pure speculation on your part.
‘So I think it’s save to assume the monitor-fixtures were of the same type too and the decorative encasings differ only in lay-out.‘
It is not safe to assume that at all. If anything, the assumption should be that the equivalent part to debris item 11 on the HL7742 seats is the upholstered upper seat back case, none of which show signs of displacement.
The properly safe assumption is that impact damage from the crash of Asiana flight 214 is irrelevant. The crash airplane was fitted with different seats with different seat-back IFE components to the Malaysia Airlines B777-200ER fleet. It is therefore a false analogy and that’s where I propose to leave it.
@Ge Rijn
I am gradually compiling a detailed description, it is well underway, and I will complete during the week. You can be assured I am giving it cogent thought, so as nothing could be construed as unsubstantiated opinion.
Greetings all.
Dr Iannello can you go back to your PMDG flight simulations where you:
“The two fuel cut-off switches are set to OFF. Although it may not seem necessary to set the fuel level to zero AND employ the cut-off switches, in the PMDG model, the windmilling action of the engine shafts continues to supply electrical and hydraulic power if the cut-off switches are not employed.”
Do you know if a fuel starved engine on a B-777 will continue generating hydraulic and electrical power as it windmills? Some jet engines will generate one or both while windmilling, but others will not. If answer is yes, did MH370 have an operating auto pilot and hydraulically operated flight controls as it drifted down?
If there is not high speed impact damage on the items recovered, does that mean it made some sort of a ‘soft ditching’?
Apology if this has already been answered. I have lots of relevant questions.
Dave
@Dave Montgomery: Welcome to the discussion.
We now know that that the rotational windmilling of the high pressure spool of the engines is sufficient to maintain hydraulic pressure but is not sufficient to supply electrical power to the main or transfer AC busses. This was confirmed in a Level D simulator run performed by @Andrew. As a result, the turn to the left caused by the left flaperon “floating up” in “bypass” mode under RAT power does not occur.
Many people here believe the recovered parts show damage consistent with a high speed impact, with some wing control surfaces separating before impact. That opinion appears to the consensus, but not held by all.
I am watching this debate for a long time on Jeffs or Victors page, but what I see, is that only DennisW is keeping common sense, and when he is calling for Sagan, maybe not a cyclist,he is calling for himself. Dennis, I would like to thank you for bigger picture
As the engines flamed out, and the systems lost electrical power, does Boeing autopilot architecture operate from standby/ emergency batteries? If yes, how long?
Is there a Boeing expert on this blog?
@Sultan
Dennis, I would like to thank you for bigger picture
Thank you, but I am standing on the shoulders of giants. Most of them in the IG.
@Dave Montgomery
RE: As the engines flamed out, and the systems lost electrical power, does Boeing autopilot architecture operate from standby/ emergency batteries? If yes, how long?”
No. AC power is lost when both engines flame out. When that happens, the primary flight control system (PFCS) reverts to secondary mode because the loss of pitot heat potentially causes the air data to become unreliable. The autopilot is not available when the PFCS is in secondary mode, so it disengages when AC power is lost.
When AC power is lost, the RAT automatically deploys to power the standby electrical system. The APU also starts automatically if there is fuel available. AC power will be restored if the APU autostart is successful. However, the PFCS will remain in secondary mode until the primary flight computers disconnect switch on the overhead panel is cycled to DISC and back to AUTO.
@Victor
“Many people here believe the recovered parts show damage consistent with a high speed impact, with some wing control surfaces separating before impact. That opinion appears to the consensus, but not held by all.”
Do you have any feel for velocity needed to cause flight surfaces to peel off and cause flutter on the flaperon? Presumably that damage would happen at lower altitude with the higher density air below about FL150. Would water erosion make sense for the flaperon say a more nose-in trajectory? I suppose the water could still make a flowing force over the wing.
@TBill: It might not have been flutter that caused the damage. It’s possible that high speeds combined with pilot input overloaded flight control surfaces and caused damage and separation, as we see in SilkAir 185 and China Airlines 006.
@Victor, TBill
Actually the Silk Air 185 accident report rules out that parts of the airplane (Boeing 737) could come off due to pilot induced nose down vertical dive. In the case of Silk Air 185 only parts of the empennage were affected, which are generally more exposed to flutter damage, and this was only because engine power caused excessive descent speed at low altitude (p. 21). In the case of MH370, on the other hand, it is generally believed that the engines were no longer running.
https://reports.aviation-safety.net/1997/19971219-0_B733_9V-TRF.pdf
No doubt uncontrolled (spiral) descent would stress the airplane structure far more than straight vertical descent, and this could explain why parts of the wings and landing gear assembly came off at that point, similar to China Airlines 006, where the same kind of damage was observed, in which case the pilot was able to recover from a steep descent, something that is similar to the spiral descent scenario for MH370 (p. 12, section 1.12: “aerodynamic overload forces”).
https://www.ntsb.gov/investigations/AccidentReports/Reports/AAR8603.pdf
@Nederland: I am hypothesizing that the flaperon separated due to high speed AND pilot input, i.e., turning the wheel or applying aileron trim, and not necessarily from flutter.
@Victor
OK, maybe you could elaborate more on this, including rationale for that particular pilot input.
It was not just the flaperon that apparently came off (due to flutter or otherwise), but also the outboard flap and landing gear door (and probably some other wing panels).
@Nederland
“It was not just the flaperon that apparently came off (due to flutter or otherwise), but also the outboard flap and landing gear door (and probably some other wing panels).”
Dont forget the RH engine cowl fragments, the two Flap hinge 7 (RH) fairing fragments, both missing their outboard edges, the RH empenage skin panels, the barely recognisable RH aileron fragments.
Leave some parts to be detached on impact
@Rob
Yes, there are lots of parts that show pull rather than crush damage.
The left outboard flap (the only part from the left wing) to me indicates that this piece did not come down along with other parts, like the screen frame or vertical stabiliser, i.e. separated earlier, like the flaperon.
All other wing parts are from the right wing. Could it mean the right wing was better protected, i.e. the plane came down left wing first, and therefore only parts from the left wing that separated before impact can be found?
(Just to say I also interpret the early flight data as intentional/hijacking, but there are lots of reasons to hypothesise the final few hours or so could have been unpiloted).
@Nederland: I am proposing that flight control surfaces that show damage to the trailing edges but not leading edges might have separated before impact due to high loading from high speeds and pilot input. The right outboard flap was adjacent to the right flaperon. Considering the case where the flaperon was damaged and eventually separated, it is not hard to imagine how the proximate flap would be subjected to high structural and aerodynamic forces during the descent.
Rationale for aileron input? One possibility is a spiral descent would be more destructive.
If MH370 followed the path BEDAX-SouthPole, it would have crossed the 7th arc near -34.25,93.79, and likely impacted the sea within a 25-NM radius of this point. The area within this 25-NM circle that is inside the 7th arc is being scanned now by Seabed Constructor. (The area to the inside of the 7th arc was scanned by GO Phoenix only to a distance of 5 NM from the 7th arc at this latitude, but out to about 18 NM to the outside of the 7th arc.)
@Nederland
It is proving extremely difficult to interpret the aircraft’s last moments, from the debris. Possibly only going to have to wait What we want now more than anything is to see the wreckage on the
@Nederland
Continued: see the wreckage lying on the seabed. Currently we have different opinions; I see the right wing as more damaged than the left. The only certainty is that on one can be sure. Trite I know, but true.
“If MH370 followed the path BEDAX-SouthPole…the 7th arc (inside) is being scanned now by Seabed Constructor.”
…thanks for sharing that thought Victor
This 180s is my favourite warm spot. If nothing there I don’t hold out much hope for the rest.
@Mick Gilbert
RE:
‘Asiana flight 214 was fitted with the same monitors as MH370.‘
That is pure speculation on your part.”
Please compare the Asiana214 and MH17 monitor pictures closely. They have the same dimensions and the same lay-out. It’s not pure speculation on my part.
Bigger problem to me now is how to explain the outer shell of the encasing (item 11) seperated from the inner shell of the encasing while both also must have been covered by the one piece of fabric that encloses both when the complete encasing seperated.
It depends on how both shells are fitted together. Are they glued together, or clipped together? How (and) did the fabric cover seperated first?
It’s strange imo we only have the outer halve of the encasing found fairly intact (item 11).
I hypotesise the fabric deteriorated while drifting in the ocean for so long and then the two shells seperated one way or another (glue disolved, wave action?). A complicated explanation.
But I think it’s very hard to explain this sequence of seperation by a high speed nose-down impact too. But maybe better, I don’t have another clue yet. Maybe you or others have?
Another explanation I can imagine is the fabric of the encasing was ripped off by water forces running through the cabin with high speed during a high speed nose-down impact. Then also seperating the front halve of the encasing from its other halve during impact.
Very complicated explanation too imo.
But at least the MH17 pictures indicate it was not a mid-air high altitude break-up. All the pictures show the fabric of the encasings did not seperate during this high speed mid-air break-up or during their falling to the earth.
The found pieces as a whole still just leave conflicting evidence. I still see no conclusive draw one way or the other; high speed nose-down impact, or ditch-like ‘slow’ vertical speed nose-up impact.
To me most signs clearly point to the latter but lack of proof and conflicting indications make it impossible to draw final conclusions now imo.
The French analytics on the flaperon would be very welcome. But they still are not pubicly disclosed.
Hopefully OI has them.
JS Posted February 26, 2018 at 11:04 AM:
“Further, the fact that the initial distance rings were presented as elevation angles instead of distances also continues to bother me. What was the purpose of the elevation angle rings?”
Are you referring to this chart: https://ogleearth.com/wp-content/uploads/2014/03/inmarsatarc.jpg ?
Similar charts exist for each satellite in the Inmarsat system:
http://www.stratosglobal.com/~/media/Documents/skyWave/Public/skywave_coverage_map.pdf
I think it is plausible they just used a chart that was available at the time. As Jeff suggests, they had not at that time produced a chart that showed BTO-rings.
Sorry, wrong blog! Ignore my post.
@Gysbreght
Welcome back.
@Gysbreght: Elevation angle is not an obscure parameter. Not only is it used for pointing, but it also relates to the observability of the satellite and whether the signal will get lost in the noise at the horizon. In fact, elevation angle is used as part of the logic in the SATCOM to determine when to log-off one satellite and log-on to another. At a high level, it has a lot more relevance in understanding the performance of the system than BTOs, ping arc radii, and line-of-sight distances. It is not surprising to me at all that it was a parameter that was used in the early days to describe the position of the aircraft relative to the satellite, because at a glance it shows the feasibility of communication between the AES and the satellite.
@Victor @PaulS
I used to think MH370 was inside Arc7@180S path. Still could be true but I am less strong on it, now thinking the data suggests it was 180S with a heading adjustment at the end to say 30-33S….not sure if I would favor inside or outside Arc7 – maybe square on it.
@TBill: I don’t know whether or not the debris field will be found near (-34.25,93.79), but the BEDAX-SouthPole satisfies the satellite data very well, with no need for changes in autopilot or autothrottle after 19:41.
@Gysbreght
Indeed, welcome back! What a nightmare it must have been for you.
@TBill
If piloted the end of a glided flight could/would have required a slow down in landing/ditching speed. Like TransAir flight 236 who made a 360 turn to loose speed before landing. Or the Gimly glider who lost speed by adversing the plane to the slip-stream extremely.
All kind of things could have happened during and after those final BFO’s before the plane impacted the surface.
@Ge Rijn
I agree active pilot flight ending by definition has many possibilities that may not have been fully explored by the modeling community, apparently due to preference for or appearance that passive flight might better explain the flight path. My perception is active flight to the end is gaining some more support now from the BFO data and possibly debris data, but at the moment glide to ditch is not favored. But it is somewhat early idea to model active pilot to the end (eg; David and Gysbreght final descent work). Jean Luc’s path is one example (“foreign” hijacker in the MH370-Captio scenario) and he is trying to reconcile a gentler ending with the “not-so-gentle” final BFO data.
@Andrew, Dr Bobby Ulich. There is uncertainty as to whether IFE lack of connection after the final log-on decisively rules out piloted powered/glide flight beyond the search width. Chances are that it would but that remains uncertain and most likely will remain so it seems.
However as I have alluded to in a post to TBill and developed further with Gysbreght on the JW blog, there is a more likely serious consequence of a professionally piloted end. It supposes the reason for him flying that far would be that he had a destination in mind, obviously within the limitations of the Peking-destination fuel he had onboard. Surely he would not have supposed that fuel exhaustion would be coincident with that, so he would have allowed a reserve margin for unexpected conditions and eventualities more generally.
Searches past and present, in assuming the aircraft was unpiloted, suppose fuel exhaustion. However had there been a professional pilot he might have jettisoned the reserves or crashed with them at his destination, leading to a search based (in part at least) on the current full fuel usage possibility being ill-founded. The aircraft would not have got that far, yet that need not be apparent since the 7th arc log-on would have the same cause if he jettisoned or shut the engines down at the destination.
As to the 7th arc timing, obviously that is a given so his fuel consumption would have been at a lesser rate than currently supposed. I have touted step climbs and a descent from around the sixth arc as possibilities for a reduction but his average speed would have been less, covering less distance in the same time.
How much fuel he would have set aside and the distance that would suggest the aircraft might be further up the arc is a nice question. @Andrew, I do not suppose he would treat this as a no-diversion reserve since such as go-rounds would not need to be included. Can you think of a logic he would follow? @Dr B, in case you care to comment.
@Ge Rijn
Re: ‘Please compare the Asiana214 and MH17 monitor pictures closely. They have the same dimensions …’
Do they? I have compared the monitor assemblies and while the Malaysia Airlines monitor assemblies as evidenced in the MH17 photos have a 3:2 width:height ratio the assemblies in the OZ214 photo appear to be wider, something approaching 2:1.
Re: ‘Another explanation I can imagine is the fabric of the encasing was ripped off by water forces running through the cabin with high speed during a high speed nose-down impact. Then also seperating the front halve of the encasing from its other halve during impact.‘
Yes, I was thinking of a not dissimilar scenario. If you look at the photo of the frame in the Debris Report where the item is being held up against a seat back you will see evidence of the frame distorting on the left hand side and cracks on the right hand side. I’m not a structural engineer but to me that looks like a violent twisting force has been applied to the assembly which, amongst other things, has led to the frame splitting into its two constituent components.
@David: It supposes the reason for him flying that far would be that he had a destination in mind, obviously within the limitations of the Peking-destination fuel he had onboard.
That’s one possibility. Another is he went as far as possible along a path.
Surely he would not have supposed that fuel exhaustion would be coincident with that, so he would have allowed a reserve margin for unexpected conditions and eventualities more generally.
Maybe, although a large reserve margin might not have been necessary. If the desired end point was entered as a waypoint, the FMC would calculate the remaining fuel when this waypoint was reached. He could have even entered a holding pattern before turning south, and remained in that pattern until the FMC indicated he would have little fuel remaining at the waypoint. Along the way to the waypoint, he could perform step climbs as recommended by the FMC, and the CI could be adjusted to increase or decrease the remaining fuel at that waypoint. As he probably did not download a weather forecast for the route, there is a real possibility of some adjustment along the way. Basically, fly as fast as possible and still get to the waypoint.
@Mick Gilbert. “….but to me that looks like a violent twisting force has been applied to the assembly which, amongst other things, has led to the frame splitting into its two constituent components”.
Sticking in my ‘hear hear’, that distortion, exemplified in a post by BuyerNinety (below), is the key to the force level. It looks to have been from impact with or by the seat, quite possibly separated. On the aircraft right side I gather.
http://i2-prod.mirror.co.uk/incoming/article8154452.ece/ALTERNATES/s615/New-MH370-debris-found-on-Madagascar-beach.jpg
@David
RE: “How much fuel he would have set aside and the distance that would suggest the aircraft might be further up the arc is a nice question. @Andrew, I do not suppose he would treat this as a no-diversion reserve since such as go-rounds would not need to be included. Can you think of a logic he would follow?”
I guess if he had a specific end point in mind he might have allowed about a tonne of fuel for the lack of accurate forecast winds, etc and possibly another tonne or so for manoeuvring, assuming a powered ditching was part of the plan. If that was part of the plan, he probably wouldn’t want to let the fuel state get too low to avoid the risk of engine flameout while manoeuvring.
By the way, it’s not possible to jettison ALL the fuel. The system can only jettison down to about 5,200 kg remaining in each main tank.
@Andrew. Thanks for that judgement. I did not mention just circling to burn it. The the longer he did that the earlier he got there, pushing transit speed back up. Still, there might be a case there too.
I had forgotten re jetttison thanks.
This would not be a powered ditching though since there would have been no log-on if so unless there was IDG selection off. I was assuming a dive, engines off,accounting for the BFOs.
Hence from what you say he might have planned for a tonne to make sure he got there, though what he would have ended up with would have varied from nil to 2 tons (with favourable winds) assuming his assessment was plus or minus.
@Victor. “That’s one possibility”. Yes not the only, though to do it by fuel gauges and flow rates he would need great faith in his instruments to be confident that, adjusting as he went, he would not end up under or over (eg residual fuel etc.)
@David: When I said “a large reserve margin might not have been necessary”, I meant greater than 4 MTs or so. Especially if the FMC calculation was done using an intermediate value of CI, it should have been possible to link reaching the waypoint with fuel exhaustion by starting with a reserve margin of a MT or two. I will add that flying as far as possible would imply setting CI=0, but the calculated speeds for great circle paths don’t seem to align with this value, so maybe this was not the goal.
@all
Still on moderator approved status at JW. Not sure why I am even posting there other than to toss cold water on far out commenters that truly pollute a responsible interpretation of the data we have. I give up. I have no idea why JW seems to fear my comments. I am not hostile or attacking in any way. I am done with that site. People deserve the “unfiltered” opportunity to decide for themselves what is worth considering. How can this man (JW) call himself a journalist when he filters sincere and, IMO opinion, valid commentary. That is truly incredible to me.
@paul smithson @victor
You mentioned 180S routes and I have to ask you a question regarding one flight path. Does the route 180S from Anoko satisfies BFO/BTO data and no inputs to the autopilot?
Thank you.
@ Dennis,
Ah ! you have seen the light. There are a few recent contributors to that blog who cannot, or choose not to understand how simple the BTO concept really is. They are raising so many red herrings, and then criticising reasonable attempts by others to clarify the issues. F***wits really. I don’t know what their motivation is, unless it is to try to demonstrate how knowledgeable they are and to bolster their own egos.
As far as I am concerned they can go figure it out for themselves. Most of the answers are in publicly available papers and documents, including the ICAO protocol descriptions and the Inmarsat paper to the Journal of Navigation. There is another paper that briefly talks about the “virtual terminal” but that is yet another red herring.
If they want and expect more detail from Inmarsat, let them go ask for themselves. Good luck with that !
@David
If I may, a few thoughts on the “fuel remaining at 17:30” question:
It used to concern me too, until Dr Bobby provided us with an accurate fuel model. If the pilot wanted to travel as far as possibly which incidentally, I believe he did (quelle surprise) then constant M0.82 at FL400 would be a prime candidate for setting of choice. This cruise setting gives a range of 2795nm from 18:29 and MEFE of 00:18:30*. My proposed path is 2788nm from 18:29 to 00:17:30, the 7nm shortfall in range being due to the additional fuel used between 18:29 and 18:44 when climbing from from FL350 to FL400. *Dr Bobby’s MEFE for M0.82 at FL400 is actually given as 00:19, rounded to nearest minute, which I think is probably due to the fuel model calc using a slightly lower assumed value for average M1 than my assumed value.
In other words, I get a 7nm shortfall in range and a corresponding 1 minute shortfall in endurance. 7nm per minute (in round figures) ties in comfortably with the ground speed at that point. The result is more copacetic than I could hoped, in fact it’s in uncanny agreement. Conclusion: the plane ran out of fuel at 00:17:30 if it had been following my proposed path.
This result is of course only valid for the S38 terminus.
Good work BTW on the IFE logon issue.
@DennisW said: People deserve the “unfiltered” opportunity to decide for themselves what is worth considering.
Deciding how best to moderate is not easy. With no moderation, a blog on this subject would quickly get overrun by trolls and those with agendas, and the value would quickly diminish. As proof, just look at the crap on Twitter that most here choose to ignore. I have rarely chosen to moderate comments of contributors, but I do when I feel that somebody is being dishonest or deliberately derailing the discussion. That is subjective, and as you know, I was accused of doing just that on another blog. Rather than complain, I decided to avoid that blog and start my own blog, which are options that anybody can pursue. If you haven’t noticed, the other blog is run by somebody that is single-minded in gathering evidence to support a particular scenario. Anybody that pushes back in a strong and persuasive manner is not going to be treated well. People contributing there need to accept that.
@David @Mick Gilbert
RE:@Mick Gilbert. “….but to me that looks like a violent twisting force has been applied to the assembly which, amongst other things, has led to the frame splitting into its two constituent components”.
Yes, to me the damage to this front piece of the encasing indicates it endured a violent impact blow caused by another object that smashed into the monitor encasing. I assume this could have happened on impact with the water.
It’s gruesome to imagine but it could have been the head of a passenger sitting behind it.
I still don’t see how the fabric could have seperated at the same time though, unless ripped off by water-forces during the water impact too.
I think it’s still possible the fabric detoriorated and seperated while drifting in the ocean over time holding the two parts together till the fabric was gone sufficiently allowing the two parts to seperate by wave-action.
Anyway item 11, the Rodrigues-piece and the found interior panel prove the fuselage/cabin must have been breached/open at least at one point (probably at least the tail section regarding the V.stabiliser piece).
I see also item 11 makes indications of a high speed/high energy impact stronger.
@Marijan
DrB’s path from ANOKO is approximately 180S (except it is more like 183S or so).
A 180S path needs to start heading south at about Arc2 (93.7E), which is approx. the line south of BEDAX or ISBIX. My latest path looks a lot like DrB’s but instead of IGOGU/ANOKO the FMT is at 0894E which is just a little further west. I am using CTH mode which catches the winds.
@Victor
Yes, I agree.
@Ge Rijn said: I see also item 11 makes indications of a high speed/high energy impact stronger.
Yes.
@VictorI @Don Thompson
After all I think it is a good thing we dived deeper into this item.
I like to see @Don’s analises too when he is ready.
SC seems to be motoring through the NW side of the secondary search area. They are recycling the AUVs quite fast. As soon as an AUV is finished its cycle, it gets picked up, data downloaded, battery refurbished and reused further up the 7th Arc.
I count 13 AUV launches since 22-FEB-2018 20:06:19 UTC in the current search area. OI appears to be only using 6 AUVs at the moment for this purpose.
Below is a link with an overview map:
https://www.dropbox.com/s/7gyfj47t9xj8znt/SC%20Track%2028022018.pdf?dl=0
@Richard Godfrey: Thanks for the update. It seems that OI is constantly improving its operational efficiency.
@Victor
Each AUV goes out 25 NM, then turns around and comes back 25 NM.
The out and back cycle is repeated 4 times, so that each AUV covers 200 NM in total.
This takes around 55 hours at 3.6 knots, which is comfortably within the 60 hours battery life.
The spacing between each line is around 1800 m.
Each AUV covers 1.8 km x 8 x 25 NM (46.3 km) = 666.72 sq km.
With 6 AUVs deployed and assuming a 10 hours turn around time on top of the 55 hours in the water, this means a coverage of 666.72 x 6 / 65 hours = 1477 sq km per day.
@Richard Godfrey: That’s quite impressive if it can be maintained. With 8 AUVs, the rate would increase to almost 2000 km2/day.
@Richard Godfrey
Thank you for the update. OI really mean business. 1400sq km per day is really impressive. Very encouraging.
Impressive indeed. But their biggest challenge is still ahead if the plane gets not found before.. Broken Ridge.
For those interested in the 7th arc bathymetry, the geological feature that lies to the north of the Tertiary Search area is the Diamantina Escparment, and to the north of that escarpment is the Broken Ridge plateau.
See GEBCO Gazeteer of Undersea Feature Names, page 439.
“Diamantina, Fracture Zone: Not a Fracture Zone in customary sense of accepted terminology. This feature marks the break between “Broken Ridge” (actually a Plateau) and the northeast flank of Kerguelen Plateau. Much of this “zone” is represented by “Diamantina Escarpment” a more appropriate term.“
@Don Thompson: Page 93?
@Richard,
Your estimate, above, suggests that Constructor is realising 23% more coverage with only 6 AUVs than has been expected with 8.
@Victor,
Thank you, yes, page 93 of 439.
@Don Thompson @all
Part of the reason I was slow to decide on Broken Ridge as a final “target”, is that I could not see obvious “targets” around 32.5S. On the other hand there does appear to be a deep pocket to 17,300-ft by Google Earth almost exactly on Arc7 right between Arc7_0Alt and Arc7_35kAlt at about 32.5S and 95.75E.
In hindsight though I would think a craggy slope would be the hardest to locate the aircraft, but I am not sure the best “target” areas for hiding debris could be judged without the new bathymetry data.
@Don
“Your estimate, above, suggests that Constructor is realising 23% more coverage with only 6 AUVs than has been expected with 8.”
This estimate is not the overall average, which is much lower. It is only a snapshot of recent days.
@TBill (&others): I’ve never been suicidal [let alone murder/homicidal], but I have experienced moderate depression. So this is personalizing a bit — I think it’s more likely than not that s suicidal PIC decided to leave his final resting place to fate.
But even for the purposes of never being found, randomization would probably be a better strategy — endless, anonymous, deep. So I doubt that BR/Diamantina, or some feature that looks like his back or his initials, are more likely than any other place on the arc.
A long-haul veteran could only be impressed by the loneliness of the SIO, and FS pre-runs only emphasize that.
@ikr
So I doubt that BR/Diamantina, or some feature that looks like his back or his initials, are more likely than any other place on the arc.
I agree. What’s more your post triggered my negative attitude on using Bayes for a problem like MH370. There is absolutely no statistical basis for doing so. It (MH370) was a singularity. Any probabilities you assign to various aspects are simply a guess not based on any previous data whatever.
Quote below from Stanford Prof:
Bradley Efron, a Stanford University statistician, said the complications make Bayes a bad fit for the Malaysia Airlines hunt. “Bayes’ Rule is good for refining reasonable (or at least not unreasonable) prior experience on the basis of new evidence,” Efron, who also expressed skepticism to Al Jazeera America, wrote in an email. “It is not good when new evidence changes the situation drastically.”
Quote below from MIT Prof:
Arnold I. Barnett, a statistician at the MIT Sloan School of Management, worries that people who use the tools without fully understanding them may be led astray, “that the very act of quantifying a probability obscures the point that the numerical estimate is itself subject to uncertainty,” he said. “Thus the estimate might be taken more literally than is warranted.”
Bayes works great with events that have well-characterized components relative to a probability distribution. It is garbage and guessing otherwise. Similar to the CSIRO probability map associated with the failed aerial search. Where in the world did they get that stuff?
Likewise with the DSTG use of Gaussian stats to characterize a process which is clearly not Gaussian.
Sorry for the rant, but it is time for people to say enough is enough.
@DennisW: No need to preach regarding using Bayes rule to find MH370. One or more assumptions or prior distributions were obviously wrong. If they were all correct, the plane would have been found. Unfortunately, the failure of the Bayesian analysis is used by some to justify ignoring some or all of the Inmarsat data set, which like it or not, is the best data we have. Without that data, there is little justification to search ANYWHERE. There have been NO other scenarios proposed that are nearly as likely as our current interpretation of the BTO and BFO data.
@ikr
Well I am more thinking BR last couple weeks, because I realized my “unfinished” 180S flight path would probably have to turn in that direction to meet Arc7 timing. Before that @sk999 had also mentioned his least error analysis method would say BTO/BFO point to 32.4S. Then we also had @David and @Gysbreght suggesting the final BFO’s may suggest the final descent was due to active pilot at the controls.
Note that there is no consensus on “who” was at the controls, or even if anyone was at the controls, but I am just putting it all together and saying what if Broken Ridge was the target, is there a sea bottom feature to line up with? I have trouble seeing a certain target in that area (based on what was known before the recent bathymetry).
@Victor
One or more assumptions or prior distributions were obviously wrong.
Yes, but my comment is deeper than that. IMO the decision to apply Bayes to the MH370 problem is what was wrong. None of the inputs were based on any sort of appropriate prior sampling. They were in the educated guess domain. I think Andrew Gelman, Columbia University statistician, summarized it well below.
“People tend to believe results that support their preconceptions and disbelieve results that surprise them. Bayesian methods encourage this undisciplined mode of thinking. I’m sure that many individual Bayesian statisticians are acting in good faith, but they’re providing encouragement to sloppy and unethical scientists everywhere. And, probably worse, Bayesian techniques motivate even the best-intentioned researchers to get stuck in the rut of prior beliefs.”
I think the original work done by the team here as well as the more recent work is far more appropriate. Anyway, I am not suggesting any of this sort of discussion is helpful. We are where we are. Just reflecting on how we got here.
@all
Well, I made a test post over at JW to see if I was still on a moderator filter. I am. Wise reminds me of a person who thinks that roosters crowing cause the sun to rise. Despite Victor’s comments on being a moderator, my Wise situation in no way resembles that. It is simply a matter of embarrassment to JW’s beliefs. (Note that I did not say misguided beliefs).
@Rob. Thank you re IFE.
My comments on your scenario, not gratuitous I hope, other than the seeming conflict with drift modelling, are:
– If he was trying to get as far as he could why not see to starting with more fuel, particularly with flight choice, or was it that he felt compelled to go that day and where he got to was in fact secondary?
– Once he climbed to 40,000 ft and set course was there anything more to be done? If not, would he choose to sit there and watch?
– About that, previously I think you have had a maximum glide in mind, following the same course. Maybe that no longer applies but if it does I go back to the first point above.
@TBill. “….we also had @David and @Gysbreght suggesting the final BFO’s may suggest the final descent was due to active pilot”.
My concern has been over the possibility there was a pilot: what the consequences would be and whether there is some deduction we could make somewhere, or further research, which would make the situation clearer. There is as yet no strong indication there was a pilot that I can think of, or not.
To me the descent rates could have resulted from a relight. That has not been simulated IMO and the duration and yaw rate resulting might prove too little. There is a longer and stronger yaw which could come from the left engine still running, without A/P or TAC or right engine. That was simulated by Boeing in 2016. Gysbreght and I share the view that in those 2016 simulations, as depicted by the ATSB, the tightness of turns is inconsistent with the timing of the final BFOs, that is if the descents depicted resulted from bank.
I would agree that the descents were most likely in a pilot induced nose drop were there confirmation that the necessary bank and timings of it could not be realised in the scenarios above, supposing there will be no further elucidation by the ATSB on the interpretation of the Boeing simulations.
Unfortunately neither is likely to be forthcoming so I plug on with looking for other clues and at how important this all is to the search.
I am looking still at the effect on along-the-arc since the IFE connection timing seems unlikely to help with across it.
@TBill
RE: “..what if Broken Ridge was the target, is there a sea bottom feature to line up with? I have trouble seeing a certain target in that area (based on what was known before the recent bathymetry).”
If there was an intend to crash the plane above a specific sea bottom feature around the 7th arc at Broken Ridge I think it would have been easy for a PIC to choose one on forehand (before the flight) only using Google Earth, then just noting or remembering the coordinates to enter at the right time (FMT?).
The ~4500m till ~5200m trenches and holes there are, and were also before the recent bathymetry, well enough indicated to be usefull and easy to find using Google Earth imo.
What is your opinion on this?
@David
I never consider your comments as gratuitous. on the contrary, they are never less than considered and constructive.
“If he was trying to get as far as he could why not see to starting with more fuel, particularly with flight choice, or was it that he felt compelled to go that day and where he got to was in fact secondary.”
The way I see it is that an undertaking like this would need a good deal of advanced planning if success is to be guaranteed. So many things can could potentially go wrong. He would need to be completely confident in his risk assessment. The following would be mandatory: A late night takeoff, a two man flight crew (preferably with a relatively inexperienced co-pilot), sufficient fuel uplift to get deep into the SIO, and most importantly, a flight date that permits fuel exhaustion to be synchronised with sunrise (acceptable times for this are in practice restricted to a few days in early October and early March)
Yes, he would have to sit it could out once he was established in the cruise. That was unavoidable. An essential part of his plan was to remain in control throughout, to glide the aircraft after fuel exhaustion, and finally to hit the water in a way that caused the aircraft to sink quickly with minimum surface debris.
For the first part of the search on the NW side of the Secondary Search Area there were 6 AUVs deployed, for the second phase 7 AUVs were deployed. There have been 13 AUV launches in total in this phase. One AUV was used to finish off a small part of the Primary Search Area.
There are 3 AUVs still underwater at this time. The last AUV in this phase is expected to be picked up on 2nd March 2018 around 09:30 UTC.
Ocean Infinity managed to reduce the turnaround time for an AUV down to below 8 hours.
I expect Seabed Constructor to move to SE side of the Secondary Search Area tomorrow.
@VictorI
“@DennisW: No need to preach regarding using Bayes rule to find MH370. One or more assumptions or prior distributions were obviously wrong. If they were all correct, the plane would have been found”.
Victor, sorry but this is simply not the case. Just because the plane hasn’t been found doesn’t necessarily mean the assumptions about prior distributions have to be wrong.
What was wrong was the assumption that the plane was unpiloted at fuel exhaustion, as was the failure to seriously consider the possibility of an extended, piloted glide.
Granted it may be true that the ISAT data taken in isolation, is not amenable to Bayesian analysis. Instead, the ISAT data should have been seen as just one element in the overall picture. If the ATSB had been permitted from the outset, to give serious consideration to the rogue pilot scenario, we would not still be here, arguing over irrelevances. If politics had not had such a dominating influence, the ATSB might have been better able to entertain the pre-planned diversion scenario, and accordingly been free to give more weight to other elements in the picture, elements like aircraft performance ie fuel range, autopilot modes, manual waypoint located beyond fuel range, terminal area lighting conditions.
Far too much weight has been given to the final BFOs. The various drift modelling results are nothing more than a distraction. (CSIRO have hardly covered themselves in glory, have they? Even Dennis will will now concur with that, I suggest)
When OI have finally finished their pointless searches northwards, they will have no option but to go downrange of the DSTG hotspot, if they are not to go home empty-handed.
@Rob said: Far too much weight has been given to the final BFOs. The various drift modelling results are nothing more than a distraction.
And don’t forget the debris which shows evidence of a high-speed impact and flaps retracted. Should we just ignore all evidence which doesn’t conform to your theory?
In the end, I fully admit to not knowing where the plane is. My best estimates are based on our best interpretation of the evidence at hand. How anybody at this point can be so sure of the end point continues to amaze me. Do you think you are suffering at least a tad from confirmation bias?
Contrary to public opinion, the ATSB never dismissed the possibility of a “rogue pilot”. By that, I think you mean scenarios which include pilot input after 19:41. (You’ve heard me say that I dislike imprecise terms like “rogue pilot” and “death dive”, which reporters like Ean Higgins use in sensationalized stories.) In fact, the ATSB even tried to incorporate the recovered simulator data into their thinking.
The failure of CSIRO to pinpoint the plane is not a failure of drift modeling. The CSIRO points were linked to satellite images of objects which probably were not parts from MH370. The drift models predict end points anywhere from around 35S to 29S.
@Ge Rijn
This presentation “The data behind MH370 Search” has 32.5S (Broken Ridge) sea floor maps before and after the new bathymetry.
MH370_data
The more fuzzy “before” data appears to be Google Maps, but they have colorized the elevations (is that something we can do on Google Maps with a setting?).
So I am thinking with the new bathymetry details, one can pick out spots to hide debris. But it does not seem as easy to pick out spots with the older data.
Ocean Infinity has searched the areas defined by David Griffin of CSIRO and shown that MH370 is not there.
Victor Iannello summarised the situation as follows: “Unless positive news is being withheld, the confidence expressed by CSIRO was unfounded. This is not a total surprise: The objects captured by the satellite images had too much surface area to likely be from MH370, and the location of the potential impact sites were not consistent with the high speed descent suggested by the final BFO values.”
There is possibly an interesting flaw in the BRAN 2015 and BRAN 2016 models, used by CSIRO. Salinity is included in the BRAN 2015 model, but David Griffin makes no statements concerning salinity in his various papers. This is surprising with regard to the buoyancy of the Flaperon and the differences between the tests in the Derwent Estuary in Tasmania with low or zero salinity versus the real life Flaperon trajectory to Reunion with the much higher salinity of the Indian Ocean.
The surface water salinity in the Derwent River Estuary in Tasmania is zero. An old study put the salinity at 0 parts per 1000 in the river estuary. Only near Cadbury, within the influence of the open sea, is the bottom water 30 parts per 1000, but the bottom water is irrelevant for trials with a floating Flaperon.
Eric Guiler in his paper Hydrology of the River Derwent states “This paper records a series of hydrological observations made on the River Derwent over a twenty month period. The salinity, pH and temperature of the river are shown. The salinity of the water at the bottom of the river at Millbrook Rise (Station 47) is 0gms/oo• The surface salinity is zero at Boyer (Station 45). At Cadbury’s (Station 5) the salinity of the bottom water is 30 gms/“.
The surface salinity in Indian ocean ranges from 32 to 37 parts per 1000.
The difference between trials in the Derwent Estuary and the Indian Ocean is between 0 parts per 1000 and 37 parts per 1000. Was this difference fed into the BRAN 2015 and BRAN 2016 models used by CSIRO?
The Flaperon would sit higher in the water in the Indian Ocean because of its higher salinity and exhibit a higher windage, which in turn changes the trajectory, which in turn changes the encountered ocean drift speed and direction as well as the wind patterns encountered.
David Griffin should go for a swim in the Dead Sea, where the salinity is 342 parts per 1000.
COORECTION:
full LINK:
https://geoscience-au.maps.arcgis.com/apps/Cascade/index.html?appid=038a72439bfa4d28b3dde81cc6ff3214
@TBill
Thanks for the link.
Indeed with the Google Earth data it would not be possible to pick a small specific feature (which data were not known at the time too) but this would be not a problem I think. The (old data) trenches and holes are easily seen in Google Earth and they are miles wide with their depths easily found zooming in and scrolling over them. Anyone could do it.
Just find the deepest spots like 32.6S/97E or32.25S/96.25E or 32.5S/97E.
There are more possibilities in the area.
A PIC would only have to know and enter one of those coördinates to be sure to end up in very deep water in rough terrain.
@analytic folks with aircraft dynamics knowledge.
I’m wondering whether we might have another candidate test of the BFO=steep descent hypothesis. If hypothesis is correct (and without pilot input), then we must to be some way into a spiral descent to get to that rate of descent (cf David & Gysbrecht). And that involves quite a bit of turn completed. Could lower bounds but put on how much turning (and how much time doing it) is needed to get to minimum ROD inferred from BFO? That also puts a limit on how much as-the-crow-flies distance can be travelled (because some of it is curve or spiral)
So, my question to test is whether you could get from Arc6 to Arc7, including that “minimum” amount of time & distance lost turning/spiralling that is necessary for BFO=steep descent? Or does the time and distance between the two preclude spiralling?
I realise that there’s quite a spectrum on the horizontal distance between the arcs depending both on the uncertainty intervals on each and change in altitude between the two, as well as the angle that you make [start?] the transit. Nonetheless, it is pretty unlikely (one-tailed p on each, compounded) that you are more than one sigma over on arc6 AND one sigma under on arc7.
It is also unlikely that you have a flight path that is BTO compliant (ie pretty straight) from arc 1 to 6, only to do an immediate hockey-stick when arc6 is crossed to assume a path perpendicular to the arcs for the whole 8.5 minutes.
More reasonable, I think, would be to assume that the track angle crossing arc6 was the same as that required to fit BTOs up to that point – which can be derived from the respective path fits by latitude on 7th arc.
So, with the above constraints, is it even possible to have a spiral descent sufficiently-developed to produce the BFO AND cover the distance between 6 and 7, if we assume initial track angle at arc 6 was BTO compliant?
Sorry for the long post – tried to make it as concise as I could while still making sense 😉
@Richard Godfrey wrote,”… David Griffin makes no statements concerning salinity in his various papers.”
I suspect the reasons the changes to freeboard/draft of the Flaperon were not taken into account were:-
(1) The Flaperon was demonstrated to have high windage and a center of gravity well forward, irrespective of whether it floated bottom or topside up.
(2) With a little work the displacement weight could be deduced from tank test photographs.
(3) Surface tension, i.e. meniscus effects can be problematic when dealing with relatively small/light objects floating with a relatively large wetted area.
(4) If measurable, the displacement weight will be in the 50 – 60 kg range, but parts of the Flaperon interior are also flooded/permeated with water, and the volume of that water also needs to be calculated to arrive at the correct weight, leading to a corrected intact draft.
(5) The change in maximum draft due to floating in salt water S.G 1.025+ as opposed to freshwater S.G. 1.000 is likely to be no more than 2 or 3mm.
(6) Surface wind will have had the most notable affect on the Flaperon, i.e. directly and indirectly through sea state amplitude.
A good point to raise, but when placed into the overall context of the event being dealt with, the affect is minuscule.
That the CSIRO backtracking of “unknown” objects was not successful, is a direct result of the whole scenario surrounding the satellite photographs and the assumptions made, which flew right across the bows of the Occam’s Razor principle.
@TBill
Thank you. We will know soon if that was the case.
@VictorI
“And don’t forget the debris which shows evidence of a high-speed impact and flaps retracted. Should we just ignore all evidence which doesn’t conform to your theory”
So you are saying that retracted flaps conflict with my theory? Surely, you can’t mean that? For the n’th time, no rogue pilot would plan a soft ditching after deliberately flying into the middle of nowhere. The ATSBs presumption that retracted flaps means unpiloted plane is fatally flawed. If you actually, genuinely support that assumption, you have given up thinking for yourself. Let me tell you now that if that assumption continues to inform the search, OI might as well go home now. Sorry Victor, but you are clutching at straws. The debris does not support a soft ditching, but neither does it support a high speed, uncontrolled impact. It does however support an impact specifically intended to make the plane sink very quickly, while at the same time calculated to leave behind as little floating debris as possible.
Winter will soon be arriving in the SIO. The window of opportunity is beginning to close.
For the record, CSIRO always said that S40 was possible, though unaccountably, the degree of possibility seemed to suffer a revision as soon as it was realized that the search was about to end as an acutely embarrassing failure.
@Narijan
Yes we will know soon re: 180S, except I now envision it might have been 180S with some changes at the end to maybe 32.5S. My personal 180S path assumes active pilot making some changes starting about Arc5.
@Rob said: The ATSBs presumption that retracted flaps means unpiloted plane is fatally flawed.
I never said that, and the ATSB never said that. However, I will say that if the intent was to ditch the plane, it would be done with engines running and flaps extended, not after fuel exhaustion with flaps up, and not preceded by a 0.7g descent reaching 15,000 fpm.
It does however support an impact specifically intended to make the plane sink very quickly, while at the same time calculated to leave behind as little floating debris as possible.
The theory that the pilot landed in a way to leave the fuselage intact except to cause just enough damage to cause a leak in the hull and sink the plane is pure fantasy. As @Andrew, our resident B777 pilot and aviation expert said: I think this notion of a ‘flat stall’ ditching that breaks the aircraft sufficiently to sink it quickly and avoid setting of the ELT, but not so much that is leaves a lot of debris, is a little bit silly (to put it mildly).
@Rob
I also think it’s a complete illusion a pilot could have any control on how a plane would break-up on impact with the ocean the way you imagine it.
Even a perfectly executed ditch with engines and all flaps available would have a quite uncertain outcome regarding break-up on the ocean surface. Depending on swell and wave-height for instance. That’s allready one thing no pilot can control but can have huge impact on the outcome of a break-up.
When in an otherwise perfecty executed ditch the tail section would have hit a swell or wave head-on on initial impact it could have seperated the tail section to name one possibility.
But recovering from a steep descent without engines would be possible I still think. The more if the APU was running during the descent (and after) which is assumed it did at least before the expected IFE log-on.
While there still seem to be uncertainties about why the IFE did not log-on.
Gliding and ditching with retracted flaps would result in a much higher impact speed and much higher risk of breaking-up the plane. Although Air Transat flight 236 and the Gimly-glider did it safely but this was on a runway.
Though if the APU was still running shortly before a ditch the flaps could(would) have been deployed if there was an active skilled pilot.
And the ATSB did not make a conclusive statement about the outboard flap being retracted or deployed. Only indicating that it was most probably retracted when it seperated.
Item 11, the Rodrigues-closetpanel and the other found interior panel though prove the fuselage at least breached/broke in one place indicating a high-energy impact,I think a ditch-impact cannot be ruled out completely yet.
But not in the controlled way you imagine it.
@Victor, re “debris show evidence of high speed impact” . This conclusion is still premature with the data in hands (Yet simple to know with a proper structural anlaysis still pending from the investigation). It is well possible that a high g led to separation of control surfaces (yet torsional failure of hinges appear inconsistent). It does not necessarily mean the impact was at high speed if (as) some one was in command (as data suggests). No much speed is needed to break the hull depending on impact conditions. High impact energy is a better representation. This basically rules out a soft successful ditch but not necessarily a hard ditch or unsuccessful ditch. The later two could also generate sufficient energy to lead to the observed debris.
@HB said: It is well possible that a high g led to separation of control surfaces (yet torsional failure of hinges appear inconsistent).
I don’t think the hinge failure is at all inconsistent with high aerodynamic loading. If the flaperon was overloaded, likely the trailing edge would first break off, producing the jagged edge we see in the recovered flaperon. But after that occurred, the flaperon is no longer streamlined and uniform along the length of the remaining trailing edge. The remaining part of the flaperon would most likely be subjected to large, unsteady forces that would produce lateral and torsional loads that could easily fatigue the hinge and lead to separation before impact.
@VictorI
To shime in. I can see high aerodynamic forces possibly could have seperated the flaperon and outboard flap. But I think those high aerodynamic forces needed could only occure in a pull-out from a high speed dive. Not during a near vertical (spiral) descent.
ChinaAir 006 and SilkAir show no wing-flaps seperated during their high speed descents (and ChinaAir recovery) only control surfaces like horizontal stabiliser and elevator parts.
If as you now propose the flaperon (and outboard flap section) seperated due to high aerodynamic forces it must have been during some kind of pull-out I think.
But then the hinges would fail (snap) before the trailing edge imo.
The brake-surfaces of those hinges (and PCU’s) on the flaperons attachement points is much smaller and weaker then the complete surface of the flaperons trailing edge.
Therefore I think your sequence of trailing edge seperating before the hinges and PCU attacment points is not likely.
And if it happened it could only have happened during some kind of pull-out imo. Conflicting a straight vertical high speed impact.
@Ge Rijn: As I have now said several times, I am proposing the flaperon could have been damaged and separated by high speed AND pilot input consisting of turning the control wheel or applying aileron trim. For China Airlines 006, there was high speed and elevator input. MH370 and China Air 006 are analogous in that both involve damage to flight control surfaces, but they differ in the kind of pilot input that might have led to the damage.
@VictorI
“The theory that the pilot landed in a way to leave the fuselage intact except to cause just enough damage to cause a leak in the hull and sink the plane is pure fantasy. As @Andrew, our resident B777 pilot and aviation expert said: I think this notion of a ‘flat stall’ ditching that breaks the aircraft sufficiently to sink it quickly and avoid setting of the ELT, but not so much that is leaves a lot of debris, is a little bit silly (to put it mildly).”
Of course I well remember Andrew’s post. I replied that I didn’t in the least mind being given the implied epithet of “silly”.
I fully appreciate that Andrew is our resident B777 pilot and aviation expert, and respect him for being so, but what we’re discussing here is a unique situation. Accordingly, I would not expect any pilot, however experienced and knowledgeable, to have the advantage in anticipating how Shah had planned to terminate this particular flight. Pilots are not trained in the skills required to make a plane disappear in the middle of the ocean. It is not something a professional pilot would ever need or want to consider.
Victor, the words “head”, “brick” and “wall” have imprinted themselves on my consciousness during our discussions. In an effort to save my sanity, I will bow out. I’m sure you wont mind.
@All: The 4th Annual MH370 Remembrance event will occur on Saturday, March 3, at 2-5 pm MYT (6-9 am GMT, 1-4 am EST) in Kuala Lumpur. Details and live streaming of the event can be found here:
https://www.facebook.com/events/151645345542413/
@VictorI
Yes sorry, I forgot you put it that way earlier. It maybe could make a difference in trailing edges breaking off first before seperating of the whole flap(s).
I don’t think so but I cann’t be sure ofcourse. I understand the forces and dynamics involved can be quite complicated leading to different outcomes.
@Ge Rijn
I think it is safe to say that the latitude of a terminus along the 7th arc is nowhere near consensus status. The choice of +/- 25nm seems to be more in the deterministic category at least relative to the OI activity.
I offer the following explanations:
1> The preponderance of qualified opinion believes +/- 25nm is a reasonable choice.
2> OI has other information (that we do not have) that reinforces that choice.
At the end of the day, intelligent people make informed choices. Note, an informed choice is not synonymous with a correct choice. Said another way, I think you are beating on a dead horse. At this moment the search strategy has been defined, and my assumption is that it will be executed. When/if that effort fails is the time to consider alternatives. Second guessing now has little value unless some new information is forthcoming. My advice is to sit back with a bag of popcorn, and watch the show.
@Ge Rijn @all
…I am finally realizing on Google Earth the funny stitchmarks on the ocean floor are the areas where the bathymetry was already known in very good detail. So you can zoom down close and see the sea floor features. Therefore the intersection of Arc7 and the previously known bathymetry track potentially represents where MH370 might be found, assuming of course intent to land there, and if so, further assuming the aim was good.
@Paul Smithson. Your http://mh370.radiantphysics.com/2018/02/16/mh370-search-update-feb-16-2018/#comment-12844
Here is one view. In the currently accepted unpiloted scenario the bank leading to the final BFOs would start after autopilot/TAC failure at loss of AC to the transfer buses, two minutes before the final log-on request (LOR). Prior to that the aircraft would have maintained the course at the 6th arc. Its speed would have been slowed after right engine failure had the left IDG been operating. If the right engine had been carrying all electrical load when it failed speed would have been maintained to that point.
The bank needed for the average acceleration during the next 8 secs, between LOR and log-on, would be over 70 degrees by my earlier estimation. Supposing that was caused by the left engine relighting from beneath idle, or more likely being powered for cruise still, the turn would be to the right. Suppose it reached 70 deg at a bank rate which increased linearly with time, that rate would be 0.58 deg/sec.
Here I aim to give an indication only based on some approximations. At half that bank, the associated ‘average’ turn circumference during those 2 minutes would be approaching 120,000 ft, a 3.11 NM radius. With speed at that point the 350 KTAS I used earlier the aircraft horizontal speed in a descent of about 10 deg would turn the aircraft 210 deg in the 2 minutes to LOR.
@Brian Anderson might have views on a turn of around that. By my understanding the aircraft would reach the 7th arc too late, given the distance from the 6th arc. A left turn of that sort of rate would get it there too early I think, again assuming the figuring is roughly right. Goldilocks would be a slow left turn but slowness would not match the BFOs.
@victor
The photos show plastic defirmation of the hinges in torsional form (as it appear). Fatigue as you know is a failure mode before plastic deformation occurs. It is well possible that one hinge failed under fatigue and the other one under torsional force but it is not clear from the photo. The two hinges seemed to have plastically deformed which in turn may indicate deformation of the wing structure and no relationship with the trailing edge damage. The fatigue crack if it occurred would leave a flat surface with initial marks of crack propagation. A close examination should be able to confirm easily what has happened and allows to rule out quite a number of scenarios.
@TBill
Yes, then you probably also noticed this detailed track is neatly covering the deep trenches around 32.3S between (and beyond) 96E and 97E.
And it’s the only detailed ocean floor track in the wide surroundings north or south. It clearly stands out.
@DennisW
In general I agree with your view and your two points.
But I don’t agree with your view I (and we) are beating on a dead horse and better sit back and watch the show.
The search is not nearly halfway. This horse is only beaten to death when this search is over and the plane is not found.
I believe brainstorming here can still lead to new insights and information that maybe can influence OI’s search strategy along the way.
I think flexibility is key to OI’s succes.
Maybe they have conclusive evidence we don’t know of regarding the search width. But it seems the crash latitude is as unsure to them as it is to us.
@TBill
To elaborate on this detailed ocean feature in Google Earth along Broken Ridge. When a specific end-destination was also planned (which I think it was) this area around 32.3S east and west of the 7th arc would also have been easy to find in 2014 by any PIC using Google Earth.
And it would serve all criteria rather well:
-the area can be consistent with the Inmarsat-data
-the area can be consistent with the drift-data
-the area can be consistent with the fuel data
-the area can be consistent with motivation; hiding the plane (and evidence) in a very remote, very deep and very rough area.
With some fantasy (which I have plenty of as you all know by now) this detailed ocean floor track along Broken Ridge combined with the 7th arc could almost serve as a crossmark on a map.
@All: At the MH370 event, the head of Malaysian civil aviation, Azharuddin Abdul Rahman, said that he expects the search to end by mid-June. Assuming each swing lasts 6 weeks, and the first swing began when Seabed Constructor left Durban on January 2, four swings would be completed around June 19. It therefore appears the plan would include four swings, of which the second is ongoing.
@Ge Rijn
It is not fantasy, it is a solution to the Broken Ridge target hypothesis that some have long held as their hypothesis. Whether or not that hypothesis is correct, is the difference between a theory and a discarded hypothesis.
Aside from the points you made, it something that can be planned on a flight simulator with the flight path linked to Google Earth. The sea floor track of prior detailed bathy data is about 5 to 7 nm wide so that’s a big enough runway.
@VictorI
This is far beyond a 90 days search limit. Positive. But how would you explain this?
@Ge Rijn: Azharuddin explained that the 90 days only narrowly includes search days, and not days Seabed Constructor spends outside of the search area.
@TBill
I was the only one I know of that for very long held this hypothesis the plane ended in the trenches of Broken Ridge around 32.3S between ~95.5E and ~97.5E. You can all read that back a long time.
I got ridiculed about these assumptions many times. Not to put myself upfront, just to mention. Nothing has been settled yet. The plane could still be anywere.
Your comment:
“The sea floor track of prior detailed bathy data is about 5 to 7 nm wide so that’s a big enough runway.”
Yes exactly, there would have been no need to pic a precise spot.
Just crash in this 5 to 7nm area and you will reach your goal of ending a plane in very deep and difficult to search terrain.
@VictorI
Clear.
@TBill,
Your thoughts for a deep and, potentially, inaccessbile ‘target’ site is an enticing theory for the final resting place. However, apart from the seamounts on Broken Ridge Plateau and Kerguelen Plateau, the ocean depth exceeds a barely accessible 3000m. Google’s bathymetry is based on a dataset produced by Scripps Institute using GEBCO as the primary source & that has remained at about 1km resolution from the SRTM work over deep ocean areas.
If one is looking for clues about an impact location, a vessel traffic density map might be more revealing.
@Don Thompson
Blah, blah. The depths south along Broken Ridge are well know for a long time also before 2014. Still await your comment on item 11..
First answer that before talking nonsens about 1km resolutions and serving a link that leads to nowhere.
The mentioned bathymetry along Broken Ridge has a resolution of 100m.
Do your homework first.
@Don Thompson
Thank you…does your link work for you?
I do not have a good feel for what 1-km resolution looks like on the ocean floor, but according to Wikipedia, in 2011, Google Earth added some higher def zones of 100-m resolution. Clearly we have some high resolution zones you can zoom in to and see the detail, exactly on Arc7 at around 32.5-S as it turns out.
@TBill,
I may have omitted to paste the URL into the markup, above. Try this.
@Ge Rijn & @TBill, concerning seabed resolution.
Only 5% of the world’s ocean floor has been mapped at 100m or better. Google’s DEM includes this data but it typically comprises strips acquired beneath the tracks of research vessels (just as the Fugro ships collected MBES data during their transits to & from Perth/Fremantle). Some of that does cover the Indian Ocean as acquired during research cruises by Scripps’ RV Melville and Revelle. This reference, related to pre-2014 Google Earth, states that only 15% of ocean data was vessel mounted echosounder acquired while the remainder is derived from satellite earth observation missions.
With hindsight, including the knowledge of the MBES surveys carried out during 2014 along the 7th arc, it’s easy to speculate that bathymetry data contributed as an element of planning for the disappearance of -MRO. I suggest that’s over-thinking the problem, even looking at vessel density maps might also be over-thinking the problem. A big empty, deep, ocean may have been sufficient ‘target’. Not overlooking that any prior’planning’ made no consideration for the operation of an idle satcom system.
@Ge Rijn,
I have no concern for your impatience. Family matters have taken priority this past week or more.
@Don Thompson
It’s clear this specific track along Broken ridge has been scanned sufficiently at (100m) high resolution before 8-3-2014. And it was the only track scanned that way in the whole area hundreds of miles north and south. It must have standed out clearly on Google Earth to any PIC who planned a disappearence of the plane from the get go.
This information was available for anyone interested early 2014 and only needed scrolling Google Earth and zooming in.
And I’m patient @Don Thompson. But I won’t hold my breath as you noticed.
@Ge Rijn
I believe brainstorming here can still lead to new insights and information that maybe can influence OI’s search strategy along the way.
Yes, you are right to keep at it. I get a bit crotchety from time to time. Sorry.
@DennisW
Thanks. I see your pragmatic ‘realistic’ approuch can turn out quite pessimistic at times. But this is a good counterweight to keep the ‘positives’ (like me) balanced.
@Ge Rijn: Why do you think that the pilot believed that investigators would suspect the flight ended in the SIO?
@VictotI
Thats a hard question. I at least sleep about it.
It’s 10 pm here in Holland. I’m off to bed thinking about it for a while.
@Ge Rijn Goede nacht. En neem eerst een klein glaasje jenever!
@Ge Rijn.
I didn’t expect you to ‘hold your breath’… if only I’d thought.
NOAA’s Global Bathymetry Data Viewer will show individual MBES surveys in shaded relief, an example.
As you might note, the N-S line follows the E95º line of longitude. The E-W line diverges from the Diamantina Escarpment at E97º.
The resolution could be sub-100m, that would require the research cruise, vessel & instrument to be identified.
So, assuming a perpetrator in control, we should entertain the notion that some oceanographic knowledge had been acquired so as to understand these ‘stripes’ in Google Earth imagery?
@Don Thompson
OK thank you I was able get the shipping lanes map.
The “BR hypothesis” does assume pre-planning looking at some bathy charts, or apparently just Google Earth (but one would have realize Google Earth had those details).
On your “an example” link above, by gosh yes “x” marks the spot. Those volcanos(?) on the East track are 25nm off Arc7 and would be just the place to start. Very interesting sea floor for sure.
@TBill
Didn’t someone mention Shah had an interest in submarines? I cann’t remember who said this. Maybe the one who did can repeat it here with a referrence?
If true at least he would have understood this features in Google Earth I presume.
The volcano’s at 32.4S/95.1E are indeed interesting but also here towards 96E there are plenty +5000m holes/trenches well suited to hide a plane.
@VictorI
RE: ” Why do you think that the pilot believed that investigators would suspect the flight ended in the SIO?”
I don’t know what the pilot believed but I think he would have realized he could not crash(land) the plane without leaving debris floating.
He could have realized sooner or later some of this debris would land on shores and then drift-analysis would lead investigators to the SIO.
But the SIO is a big place. It would not have been a great concern to him.
Only with drift-analysis it would be impossible to find the crash-site.
Only combined with the Inmarsat-data and the 7th arc investigators were able to come up with the areas they are now searching.
And no pilot at the time could have been aware of the Inmarsat-data being send during the flight.
to add.. or used to find out a general direction and track of the plane.
@Ge Rijn: So if you agree that he believed that it would be nearly impossible to estimate where in the SIO the plane crashed, why would he be concerned with hiding the wreckage in challenging terrain?
@VictorI
Found debris would only indicate a very large area in the SIO (like the Geomar reverse drift-studies). But then the pilot/planner could have expected satelite images of this area (at the vanishing-date) would be studied possibly finding floating debris and further closing in on the crash area.
Then he also could have calculated with unforseeable factors turning up like the recovered sim-data or scientific factors not known to him like the Inmarsat data turned out to be.
I think to be as sure as possible no one would ever find the plane and discover who was responsable, the pilot/planner would have had reasons enough to make it as difficult as possible for investigators to find the plane. In this respect hiding the plane in one of Broken Ridge deep trenches/holes makes sence imo.
@Ge Rijn
Re: submarines interest- I believe that is well documented, but I don’t know if that infers anything, as submarines don’t go that deep excepting for special research vessels.
@TBill: There is a well-known scandal involving bribes and murder surrounding the Malaysian purchase of French Scorpene submarines. Some believe that the captain’s T-shirt with the submarine was a reference to this scandal.
@Victor @all
Wow OI is getting really close to a BR area that I might rank highly as a hot spot (near the high def focus area on Google Earth). They are getting close to the volcanos on Google Earth.
The out-of-focus volcano just north of 33S on Arc7 (almost where the ship seems to be sitting now) is one spot that I would tentatively assign more interest, because even out-of-focus you can see that is a “high” spot on Arc7, and I assume a sloping rugged mountainside area is hard to find debris.
@Shadynuk
Had ik al gedaan dus kon al niet zo helder meer denken..
Zoals je inmiddels weet is dit blog een echte uitdaging voor je verstand.
Zeker als Hollander die de nuances van de Engelse taal niet goed beheerst.
Bedankt voor je berichtje!
All: nothing special above. Just a Dutch in between of appreciation.
@TBill
Someone with a special interest in submarines would have known the importance of scanning sea and ocean floors with sonar at any depth.
Someone with this knowledge would have recognised those detailed tracks in Google Earth immediately.
I ofcourse agree with you SC is closing in on a very crucial area.
I’m only concerned they will take the width too small here east and west of the 7th arc. ~15Nm more east and west (beyond +/-25Nm) could make the difference. But lets wait and see.
@Ge Rijn, @TBill: What evidence is there of a special interest in submarines?
@VictorI
I don’t know. Someone here once mentioned Shah had a special interest in submarines. I asked for a repeat and a reference.
Anyway for someone looking at specific features and educated in geographics (like a pilot) it would have been not difficult to understand the meaning of those ocean floor tracks imo.
Even we figured it out.
@Victor
Hearsay, but I think someone mentioned it on JW within last couple days.
@Ge Rijn writes “Someone with a special interest in submarines would have known the importance of scanning sea and ocean floors with sonar at any depth.”
Please, do elaborate.
@TBill: Based on that, I’m even less persuaded than previously.
@all
I Googled the submarine question pretty hard, and came up empty.
I think this is the photo that links the captain to submarines.
On Twitter, @JoeNemo3 found some interesting features on the seabed from the bathymetry data using the Geoscience viewer. What appears to be hard objects are located at -32.53333, 95.89355, about 5 NM beyond the 7th arc. The debris field would be quite long, if from MH370. Perhaps somebody with a trained eye can take a look and provide an opinion.
@TBill. On the JW site I made the following reference to submarines, responding to Joe Nemo on Feb 17th 6:46 PM:
“….A point about his frame of mind. Corruption in the purchase of 2 submarines by Malaysia from the French may have contributed to Zaharie’s apparent exasperation with Malaysian government corruption – and hence the T shirt.
There are numerous articles on the web about this of which here are two. They make for disquieting reading.
https://sites.tufts.edu/corruptarmsdeals/2017/05/05/the-malaysia-scorpene-submarine-affair/
https://www.channelnewsasia.com/news/business/associate-of-malaysian-pm-charged-over-french-submarine-deal-9083744
@Victor
That is interesting Joe Nemo observation. Note the planned 25000 km2 search area (on your graphics) seems to end just a small distance short of that, I’ll say maybe a couple hundred yards, it would nice if they could just take it a small tiny distance further while they are up there.
@VictorI
Re Joe Memo’s bathy images: If you enlarge the second (mid scale) image you will notice the objects appear to be aligned with a fine, image-wide linear grain pattern. This grain pattern is probably an artifact of the scanning process. Therefore, it could be a boulder field which the grain pattern has served to make look artificial. So unfortunately, to this “expert eye”, not likely to be the target.
@Rob
Good observation, IMO.
@Don Thompson has downloaded the bathymetry data and says he believes the feature is a ridge extending south from an escarpment.
@Don Thompson @David @VictorI @DennisW
Thanks. It seems clear there is no indication Shah had a special ‘technical’ interest in submarines only a ‘political’ one.
The feature highlighted by @JoeNemo3 is coincident with a sharp ridge that extends into the floor of the Diamantina area (vertical profile shown over 4km span, >200m elevation).
The bathymetry, and ‘processed’ backscatter at 30m resolution, is reproduced from Geoscience Australia’s geoTIFF distribution. The backscatter image exhibits many artifacts from the absence of data.
It may be premature to interpret much from the MBES backscatter data.
@Victor @all
Even if Joe Nemo’s observation is an artifact, the current search border cuts off right in the middle of that trench. Depending on the schedule, it could be some time before they get back there to finish off that artifact area. Right now they seem to be at the high def area on Google Earth and Arc7. It’s OK though to take their time and do it right.
It’s hard to imagine how the surveys can accurately see the debris in the volcanic areas/rough terrain. I now see the one big volcano at 32.85S that I was assigning importance to, is apparently in the prior search zone.
@Victor @David
“…a fact the Pilot was very likely aware of given his apparent interest in submarines.”
It was Joe Nemo who had referenced this above a few days ago on JW
@TBill
RE: “That is interesting Joe Nemo observation. Note the planned 25000 km2 search area (on your graphics) seems to end just a small distance short of that, I’ll say maybe a couple hundred yards, it would nice if they could just take it a small tiny distance further while they are up there.”
OI allready showed they don’t take the given widths as fixed borders.
In the CSIRO/Griffin hotspots they also took several Nm outside this defined search areas.
I trust they will take your couple of hundred yards and even quite a bit further.
I think it’s all about the best prospects, wheather conditions expected over time (which will deteriorate towards June) and money-wise thinking ofcourse.
At least now (imo) they have the best opportunity to search this Broken Ridge area between ~95E and ~97E while the weather still allows it and time is still plenty.
@TBill
The schedule defining the Primary, Secondary, and Tertiary Search Areas was set out in the agreement with Malaysia. It was surprising that Constructor did not complete the Secondary Search Area on both sides of the 7th arc before moving into the Tertiary Search Area. However, assuming the present rate of coverage can be maintained, these areas should be complete before Constructor begins its transit to Fremantle on/around March 17th.
I have previously questioned how effective the survey can be over all prevailing seabed conditions, I’ve talked to a number of experts & have been assured a debris field is distinct.
@Don Thompson
In other words; if it’s there they’ll find it. If not we’ll know it’s not there also.
@Don @all
Don – thank you
I find this search area graphic by Kevin Rupp to be very useful:
https://pbs.twimg.com/media/DXfep6tVAAIlDRI.jpg
The Pink track is the prior search, the Green line is Arc7 (not sure by what definition, but looks like 0-ft Altitude) and Red is current SC/OI.
It shows how little of the area Inside Arc7 was previously searched. So I guess they got it now. It also shows the big Volcano at 32.85S was apparently searched previously. When viewing, I try to keep in mind Arc7 could be +- 5nm from where we think it is or thereabouts.
As Sartre once wrote, “Les jeux sont faits” – OI has committed to its search strategy. We can provide no more input. All we can do is watch and wait.
Which is an excellent time to revist flight MH21 from KL to Amsterdam on the night of March 7, a flight that departed KL at about the same time as MH370. The BFOs for this flight were published as Fig. 15 in the JON article. The predicted BFOs were shown as well, but the data behind the figure were not released. However, the DSTG used this flight as one of its “calibration flights”, and Fig 9.9 provided graphical representations of flight path, altitude, heading, and Mach.
Curiously, the JON article only shows BFOs up to 21:30Z, even though the DSTG figure shows data out to 24:50Z. I do not know why the last 3 hrs of BFOs are now shown. In the following, I focus only on the flgiht path up to 21:30Z.
Because the flight path does not include timing information (and is too coarse to capture the nuances of small but noticeable turns), we need to do a little reverse-engineering to get the path v. time. For the most part the path follows airway P628, but it is clear from the heading information that there were deviations/shortcuts along the way, or else the GDAS model has serious errors. Here is my best reconstruction of the flight plan using waypoints or (in a few cases), long/lat pairs that best match the heading/Mach/altitude information up to 21:41Z.
KL GIVAL MINAT IGREX PPB VATLA URKOK LARIK DORIL OPASA IKINA PUKES 73.368,27.35 71.442,28.258 68.690,29.525 AMBER
Mach fluctuates but is ~0.82 on average.
So what does this mean for BFOs? Short story – not much different from MH370 or MH371. RMS is 2.2 hz, using a fixed frequency bias of 178 hz. Peak-peak width of 9 hz. No Monster of Mumbai effects.
I still do not know how the DSTG obtained a distribution of BFOs with an rms of 4.3 hz, much less was able to justify a distribution with an rms of 7 hz. Irrespective of Gaussian or non-Gaussian behavior.
@sk999: Interesting result. If there was little drift of the fixed frequency bias, it would be hard to justify an endpoint latitude beyond the priority search area. I wonder whether flights other than the Mumbai flight experienced this drift. Puzzling.
@sk999
still do not know how the DSTG obtained a distribution of BFOs with an rms of 4.3 hz, much less was able to justify a distribution with an rms of 7 hz. Irrespective of Gaussian or non-Gaussian behavior.
The BFO error (drift in the AES oscillator) is non-Gaussian. Why even talk about RMS? It is ridiculous. It is non-ergodic and non-stationary. I don’t know how many times I have made this point. Take a look at Figure 5.4 of “Bayesian Methods…”. I know you are a bright guy, but why you continue to fight this is a mystery to me. BFO is essentially useless except for determining the aircraft went South and came down rapidly at the end.
@Dennis: If everyone would just agree “it went south and came down rapidly at the end”, about half the spy novel/film scripts on these blogs would disappear in a big puff of smoke..
@Victor, Andrew. Below, extending previous discussion about a professional pilot having fuel reserves, is a line of argument that the presence of an active professional pilot at the end is unlikely, IFE connection or not. It is based on his likely choices approaching MEFE.
For any comment please.
https://www.dropbox.com/s/h9hvlm1ev89761l/The%20likelihood%20of%20a%20professional%20pilot%20being%20active%20in%20MH%20370%27s%20final%20descent.docx?dl=0
@David
For you a pilot setting a course to Antarctica and waiting for fuel exhaustion is too passive.
For me a pilot using elevator to ensure a high speed dive is maintained and then maybe trying to pull out into a glide is too active.
Andreas Lubitz, the murder suicide pilot of German Wings, once set on his course to crash into the Alps, did not deviate from this course.
I do not expect that Zaharie Shah changed his mind after he had set his course to the Southern Indian Ocean.
@David said: In that case the only purpose that I can think of for a professional pilot to remain active, yet waiting, would be to crash the aircraft at a spot he had in mind, along a course he had set.
Or, if the plan was a high speed dive far into the SIO, he wanted to ensure the plan was completed, rather than take his own life before that.
Because there was a log-on that renders a powered dive unlikely therefore.
Maybe. But what people fail to appreciate (or don’t want to acknowledge) is that the pilot was fully aware of the state of the SATCOM based on EICAS messages that are displayed. When the left bus was first isolated, these would appear:
SATCOM
SATVOICE LOST
SATCOM DATALINK
After the reboot at 18:25 with the ACARS disabled, there would be this message:
SATCOM DATALINK
So he would know that the SATCOM was operational, and voice calls would work, although the ACARS link was not working, which means ACARS position reports would not be sent. It’s possible he would also be aware of the log-on sequence to the Inmarsat network.
It may be that with some fuel remaining, the pilot chose to again cycle the power, again generating a log-on sequence. He might have chosen to put the plane in a rapid, powered descent after the SATCOM and SATVOICE LOST messages again disappeared, which explains the downward acceleration caught by the last two transmissions.
Left engine failure followed by APU auto-start is the only sequence currently seen as consistent with a log-on, yet that sequence is unlikely with the aircraft being in the hands of a professional pilot.
With an active pilot, there are many possibilities, especially if you consider an awareness of the SATCOM, which he might have had based on the previous power cycling of the bus and the disabling of ACARS.
@Victor, thank you for pointing this out in detail.
The fact that Shah must’ve been fully aware of both sat com reboots hasn’t been widely acknowledged because of the long held notion that at least the final leg into the SIO was an unpiloted ghost flight. However, if an active pilot pointed down the nose of the plane after the second reboot was completed, we have to consider an awareness of the sat com. And this opens up many acting possibilities before the final crash and specific motives for staying alive and conscious until the very end.
@Sabine said: The fact that Shah must’ve been fully aware of both sat com reboots hasn’t been widely acknowledged because of the long held notion that at least the final leg into the SIO was an unpiloted ghost flight.
There’s also the very mistaken belief that he would need detailed knowledge of electrical schematics. That’s patently false. The EICAS messages make it abundantly clear what the state of the SATCOM is.
@David @VictorI
Another possibility that could make sence maybe; he isolated the Left IDG again in a (late) reaction to the second SAT-phone-call when nearing his destination after 0:11. Maybe thinking the SATCOM connection could give away his crash position. Or to just avoid the burden of a possible phone call again in the final minutes.
Then when the right engine flamed out he set-up for his final descent. In between the APU started and triggered the log-on without IFE log-on due to load-shedding during this descent.
He would have had full control for the left engine was still sufficiently running.
He would have had the choice to dive high speed or to recover and ditch the plane.
@Victor
“If MH370 followed the path BEDAX-SouthPole, it would have crossed the 7th arc near -34.25,93.79, and likely impacted the sea within a 25-NM radius of this point. The area within this 25-NM circle that is inside the 7th arc is being scanned now by Seabed Constructor.”
Victor the new weekly report says no hits as of 4-March. So sadly 180S *unpiloted* is looking like your second chance is not panning out. I am still thinking *piloted* 180S makes some changes after Arc5, so I am glad I moved my pin over to 32.5_S. But I am shocked they are already there….but I think more work to be done in BR for a while.
@Ge Rijn
I agree 2314 sat call could have been an event that an active pilot reacted to, in addition to other events at the time (sunrise and high winds at FL350). Speed, altitude, heading changes are all possible I feel.
@Ge Rijn: If he isolated the left bus after 23:14, the flameout of the right engine and start of the APU would not have powered the SATCOM, as the tie breaker for the left bus was open.
However, an awareness of the state of the SATCOM leads to many possibilities.
@TBill: Any single identified “warm spot” has a relatively low probability. This thing is only going to be found by progressively scanning a large area, and even then it’s questionable.
@VictorI
RE: “If he isolated the left bus after 23:14, the flameout of the right engine and start of the APU would not have powered the SATCOM, as the tie breaker for the left bus was open.”
Are you sure on this? Wouldn’t the APU have direct wireing to the not load-shedded or turned off systems when one or both engines are flamed-out or isolated another way?
What would be the use of an APU in this regard?
@VictorI
I’m not sure on this but I think the APU will override any isolation/flame-out of the left (or right) engine, restoring all essential instruments and controls which were not selected off or load-shedded.
Probably @Don Thompson can elaborate more on this?
And then essentially you are now only left with your YWKS, 45S/104E and NZPG warm spots. All others turned out to be not valid.
And the more towards the north, the data get less reliable. Both the Inmarsat-data and the drift-data.
IMO the only reasonably chance OI has, to find the plane, is along the Broken Ridge trenches. They should search it wide and thoroughly.
Up north (of Broken Ridge) there’s nothing special.
And those scenarios assume an unpiloted/unplanned end after FMT.
Which has been the most basic misconception from the start and still lingers on.
@Ge Rijn: Yes, if you think that when the left tie bus switch is manually set to ISLN, that it is possible that the APU can override this setting, I’ll let you continue that discussion with @Don Thompson.
@Ge Rijn
1. If you look at Victor’s graphic at the head of this post, you will see that YWKS was a warm spot first published (over a year ago) by me and that NZPG is the result of joint paper with Victor (over 1.5 years ago).
2. Victor’s warm spot resulting from a BEDAX – South Pole 180 degT south flight path, has not been searched on the SE side of the 7th Arc.
3. The Inmarsat and Drift data are just as reliable further north, as they are further south. The key is to make the correct interpretation and justifiable assumptions, which DSTG and CSIRO failed to make.
4. I have personally communicated to Ocean Infinity not to overly waste time with the Diamantina Escarpment.
5. The case for a piloted or un-piloted flight until the end has not been conclusively proven.
@Richard
“2. Victor’s warm spot resulting from a BEDAX – South Pole 180 degT south flight path, has not been searched on the SE side of the 7th Arc.”
You are correct Richard. Victor still gets a third chance on that one. Fingers crossed!
@Don Thompson
“It was surprising that Constructor did not complete the Secondary Search Area on both sides of the 7th arc before moving into the Tertiary Search Area.”
I agree Don.
Malaysia says it will pay Ocean Infinity $US20 million for 5000 square kilometres of a successful search, $US30 million for 15,000 square kilometres, $US50 million for 25,000 square kilometres and $US70 million if the plane or flight recorders are found beyond the identified area.
There is a contractual step difference of $10M between the primary and the secondary search areas. There is a further contractual step difference of $20M between the secondary and the tertiary search areas. Moving to the tertiary search area before completing the secondary would imply that either the money is not important or that the secondary search area does not hold much promise.
So although Malaysia agrees 16,000 square kilometres has been searched up until 4th March 2018, Ocean Infinity is still in the $30M mode, not the $50M mode.
@Richard
The OI payment schedule you posted is the same as other reports I have seen. It is very odd that the 10000 km^2 between 5000 and 15000 is paid at $1M per km^2, and the 10000 km^2 between 15000 and 25000 is paid at $2M per km^2. I think the original report was wrong, never corrected, and simply repeated by other reporters.
@Richard
$1M per 1000 km^2 and $2M per 1000 km^2 above.
@Ge Rijn
RE: “Wouldn’t the APU have direct wireing to the not load-shedded or turned off systems when one or both engines are flamed-out or isolated another way? What would be the use of an APU in this regard?”
“I’m not sure on this but I think the APU will override any isolation/flame-out of the left (or right) engine, restoring all essential instruments and controls which were not selected off or load-shedded.”
NO! There is no ‘direct wiring’ between the APU generator and the aircraft’s electrical components. The APU generator powers the tie bus, which then powers the L & R Main AC buses and any components connected to those buses or their various sub-buses. However, the APU generator can only power those buses if the respective bus tie breaker is closed. If the L Main AC bus has been isolated by manually selecting the L bus tie switch to ISLN, then the bus tie breaker will be open and the APU generator WILL NOT power the L Main AC bus.
The system is designed to allow the APU to automatically power the aircraft if AC power is lost, but that can only occur if the bus tie switches are in AUTO. If one or both switches are selected to ISLN, then the system CANNOT override that selection.
@Richard, Victor, Ge Rijn. Thanks for your comments. I do not think that the influence of a professional pilot’s choices on whether there was one at the end has come up before. Anything to try to get at a better idea of a pilot (professional and otherwise) vs none’s implications.
@Richard. “For me a pilot using elevator to ensure a high speed dive is maintained and then maybe trying to pull out into a glide is too active.”
Yes I agree. The need I had in mind for good elevator control was “to continue with that descent”. Silk Air: forward trim plus considerable pilot forward force in that instance.
About your post to @Ge Rijn, “5. The case for a piloted or un-piloted flight until the end has not been conclusively proven”.
I for one do not rule out an unprofessional pilot at the end. The case that there wasn’t a professional is hardly conclusive, particularly with my limited flying experience trying to guess what a professional would do, in an aircraft I do not recall ever seeing, much less know. However, with this event commonly we deal in probabilities. Not even the final limits of the final BFOs are entirely certain; and certainly not how they would have been realised, if unmanned.
The pilot-or-not question offers some scope for confirming or otherwise refining this search and to look at as a what next should it fail and yet other data and assumptions still stand.
@Victor. “It may be that with some fuel remaining, the pilot chose to again cycle the power, again generating a log-on sequence. He might have chosen to put the plane in a rapid, powered descent after the SATCOM and SATVOICE LOST messages again disappeared, which explains the downward acceleration caught by the last two transmissions.”
What would his motive would be in recycling the SATCOM for its own sake? If for another purpose what could that be? Also, do you mean this would be a coincidence in timing or deliberate?
I note that depending on the fuel remaining his destination would shift up the arc, all else being equal.
@Ge Rijn. “….he isolated the Left IDG again in a (late) reaction to the second SAT-phone-call when nearing his destination after 0:11. Maybe thinking the SATCOM connection could give away his crash position.” If sensitive to that wouldn’t he have done that before the first phone call, which surely he could have expected? If not, immediately after?
“In between the APU started and triggered the log-on…” Supposing you mean the APU auto-start commenced at right engine failure there was no AC power at that point. The autopilot and TAC would have dropped off line, left engine still running.
I would be surprised if a professional pilot would plan on countering that thrust manually pending the APU coming on line, then restoring the FCS from secondary, then re-engaging the A/P and TAC.
BTW one other facet is that with loss of AC the left engine would rely just on the APU DC fuel pump for pressurised supply pending the APU coming on line. That would deliver less than cruise fuel consumption and it would be supplying the APU start also. It may be that the engine would be able to supplement that with suction fuel, depending on aircraft accelerations around any axis and the fuel remaining. Maybe not, and it might flame out. For that reason also I do not think he would countenance an engine running like that, even temporarily.
As to the APU connection to the left and right main buses I see that @Andrew has answered that. I hope I am right in adding though that if the pilot had chosen to isolate the left bus by manually selecting the right bus tie switch to ISLN, and only that, the APU then would power the left bus when it came on line and only that.
@David
“What would his motive would be in recycling the SATCOM for its own sake?”
Sending a “message”? We do not believe ZS realized about BTO/BFO signals but perhaps he realized there was a system logon recorded?
@TBill. A taunt do you think, not to his knowledge disclosing anything about position and to be spotted some time after the event?
But there would be a risk that it would not be or treated as an aberration?
A non-professional fiddling in desperation?
Whichever, why the 18:25 log-on, with its IFE connection besides?
Just musing.
@Andrew
RE: “The system is designed to allow the APU to automatically power the aircraft if AC power is lost, but that can only occur if the bus tie switches are in AUTO. If one or both switches are selected to ISLN, then the system CANNOT override that selection.”
Thank you. This then effectively rules out the suggestion I made.
@Richard
@Richard @don
I wonder does the language in the payment agreement actually specify primary, secondary area etc? because by the payment schedule you’ve posted it doesn’t matter which area any find is in, simply how much area is covered before the find. in fact covering a theoretically less likely area first makes sense, as it increases the likelihood of covering more than 5000km before discovery and hence moving into the 30m bracket.
SC has started on the Secondary and Tertiary Search Areas on the S.E. side of the 7th Arc.
https://www.dropbox.com/s/b0n52bw1834eej1/SC%20Track%2007032018.pdf?dl=0
Hi,
http://www.bbc.com/news/world-australia-43299283
132 years from 32°49’105°25′ to an Australian beach ?
Probably not: The bottle have to be buried for 130 years or more…
😉 PL
All,
I have a new paper on the identification of the MH370 route into the SIO using a matched filter detection process.
You can get it HERE .
That link doesn’t seem to work. Try this:
https://drive.google.com/file/d/1NM7c9Vs6dJgOczqnU0682Jqw5Mjp84-J/view
Found the problem with the link. It lost a bit on the end. The third time is a charm:
https://drive.google.com/file/d/1NM7c9Vs6dJgOczqnU0682Jqw5Mjp84-J/view?usp=sharing
Some of you have asked what may have been Shah’s motive to (possibly) initiate another power cycle at the end of the flight and to wait with pointing the plane’s nose down until the second and final reboot of the sat com was completed.
Personally I think it’s possible that Shah didn’t know about the hourly handshakes but that he was well aware of the fact that the two log-on requests would be registered. Therefore they would provide an exact time frame of the plane’s flight from 18:25 onwards. This time frame correlates well with a fuel-exhaustive flight. Shah might’ve provided the investigators deliberately with a strong clue that the plane crashed because of fuel exhaustion. And since there would be no radar sightings whatsoever in any direction after 18:21, and no crash reports on land, the investigators would quickly conclude that the plane must’ve headed towards the open Indian Ocean, where it eventually crashed. Shah might’ve even de-selected the IFE system, so that the missing response after the sat com’s second reboot would lead the investigators to the conclusion that the IFE system’s response was missing because the plane crashed right after the final reboot, or because the APU had run out of fuel, which would also announce a subsequent crash.
But why would Shah have wanted the investigators to know that the plane crashed somewhere into the Indian Ocean? I don’t believe that taunting was his main motive. If there hadn’t been any sat data whatsoever it would’ve been totally unclear what happened to the plane – until the first debris was found. As we know, it took one and a half years until the flaperon was found. Until then the NOK and ALSO Shah’s own family would’ve been in a terrible limbo. They might even believe that Shah absconded with the plane to a country like Pakistan. I really don’t think that Shah would’ve wanted this. He loved his children and even visited his oldest daughter in Australia before he abducted the plane. And while they might’ve been estranged there’s no evidence that he hated his wife.
I believe that Shah didn’t want the plane to be found for reasons which are still not quite clear and still need to be discussed. But he might’ve wanted to send a conclusive message to the investigators that the plane had definitely crashed and everybody onboard was dead.
Shah might not have been aware that the sat data would eventually tell the investigators much more than that, and would allow to predict far more precisely than he had intended where the plane probably has gone down. But even with this additional information the plane is still elusive exactly four years after it vanished. So, Shah’s strategy has been pretty successful, even if he didn’t intend to provide the investigators with all the additional information which could be extracted from the sat data.
This theory raises the question why Shah didn’t provide the investigators with even stronger evidence that the plane went towards the SIO – for example by executing the fmt while still being within radar range. I don’t have any conclusive answers. But Shah might’ve wanted to keep it as vague as possible into which direction he was going. Also, a radar capture which would show a deliberate turn southwards would clearly indicate that the plane was still actively piloted, and that might not have been within Shah’s interests. He might’ve wanted to keep open the possibility of a runaway plane. Another possibility is of course that Shah did execute a final southern turn within Indonesia’s radar range – but he didn’t know that Indonesia shut off his radar facilities after midnight.
Maybe, others can come up with better ideas, why Shah might’ve wanted to provide deliberately an exact time frame for his flight.
@Sabine said: Another possibility is of course that Shah did execute a final southern turn within Indonesia’s radar range – but he didn’t know that Indonesia shut off his radar facilities after midnight.
That’s a good point that has not been discussed very much. It’s possible that the pilot deliberately flew a route that was non-threatening to Indonesia and would not provoke a military response but nonetheless provided evidence of a southerly path.
@DrB
You have come a long way from 40S to 31.5S. Nearly there to 30S +/- 1 degree, that I published over a year ago.
Since Ocean Infinity is already at 32.4S, your latest prediction of 31.5S will soon be covered.
I should add that it is by no means mutually exclusive to become a murderer for political motives and still care for the family. I don’t think that Shah was a crazed psychopath or a depressive mass murderer like Andreas Lubitz who obviously was in a state of mind where he wasn’t able anymore to give a damn about his or anyone’s family.
Shah probably had an exact plan which he tried to execute as precisely as possible. While his motives were most likely political, it’s still unclear what exactly he hoped to achieve. We don’t even know if he succeeded or if his operation was a failure. And that’s of course the weakest point of this theory. Important pieces of the puzzle may still be missing. But I think we are not far off the truth if we put Shah into the same category as a politically motivated terrorist who believes that killing is necessary and morally justified for the greater good. And as we all know – terrible crimes have been committed for The Greater Good. The highway to hell is paved with good intentions…
@Sabine L.
Interesting views but purposely making the FMT within Indonesian radar range with the intend to give investigators a clue seems unlikely to me.
If Indonesian radar was on-line that night (which is still unsure imo, the Indonesians only stated MH370 was not seen by radar in their airspace) he would not only have given a clue but he would have given his (deliberate) heading/track into the SIO.
And this together with the known fuel-data and the Inmarsat-data would have abled the investigators to calculate a crash-area soon after 8 march-2014.
All search strategies would have been focused immediately on this area including aerial searches.
@DrB: Thank you for that paper. It will take some time for contributors here to wade through the details. Although I disagree with your assessment of the uniqueness of your proposed endpoint of 31.57S (in much the same way I disagreed with the uniqueness of your previously proposed endpoints), I certainly think your endpoint is within the realm of possibility. The paper provides further justification to continue the search beyond the 25,000 sq km identified by the ATSB and CSIRO.
@DrB
Wow. As Victor stated, it will take some time to digest your paper. I also thank you for it. Your dedication and work ethic are impressive as are the efforts of many others here – a long list.
@Sabine
Yes, categorizing events as “deliberate”rather than as artifacts has a big effect on the solution space. Unfortunately it does not make it smaller. 🙂
@DrB @Richard Godfrey
I notice you’re both nearing 32S more closely. And @Richard Godfrey has even made a side-step lately beyond 32S to ~32.2S.
[VI: In a later comment, @Richard Godfrey states that he has NOT recently proposed 32.2S latitude.]
But you’re both still working with the premises the flight was not actively piloted after FMT and therefore the end-of-flight and crash were uncontrolled too, ending up close to the 7th arc in a high speed impact.
You both could be right but very wrong too imo.
All known evidence and indications show there is no reason/evidence at all the flight changed into an unpiloted flight after FMT.
Many debris finds show clear signs of a relatively low vertical/horizontal speed, nose-up AoA impact. Other finds prove the fuselage breached at least at one point indicating a quite high energy impact.
The case now seems to focus more on the possibilities of an all controlled flight by a pilot. And with this a possible recovery and glide after the steep descent. Which could affect the search width.
I see educated consensus is building close to 32S but still on the premises of an unpiloted flight and ending after FMT.
This could turn out to be a big mistake if OI holds on to this believe also.
We’ll see. The most crucial area gets searched right now imo.
After this only unpiloted scenarios after FMT could make sence like you propose.
@Ge Rijn
I don’t think anyone here would disagree with your view of the search width. The question is prioritizing the search area. My view is that it is better to use the current width and extend the search to the North rather than expanding it now at more Southern latitudes. I think Richard leans more to 30S (as do I). I think a search at +/-25nm up to 26S is the best course of action right now. If that proves to be unsuccessful, then it is time to consider expanding the search width as a possible next step.
@DrB
Welcome back thank you very much for the new path proposal. Took me a while to find AGEGA…I see it now!
Question:
What is the north/south military radar track along about 93.75E marked by the “x” on your flight path?
Comments:
My updated proposed 180S path (below) is quite similar to yours:
VAMPI MEKAR NILAM 0796E 0894E 180S_CTH
But I am thinking the following happened as you observed: “…it is not impossible for 9M-MRO to have descended to a much lower altitude between 18:29 and 19:10 and then climbed back to FL335 between 19:15 and 19:41.”
I am thinking said descent happened at 18:40, to allow passing under N571. I am thinking MH370 is in the process of gradual re-ascent at 19:41
@Ge Rijn
“And @Richard Godfrey has even made a side-step lately beyond 32S to ~32.2S.”
NO! I have never made such a statement!
I HAVE BEEN SAYING FOR OVER A YEAR NOW 30°S +/- 1°.
@DennisW
I generally agree with your view. The plane could well be within the +/-25Nm range also with a recovery and glide. I’m sure a long glide would not have been an objective. But it could well have ended several Nm outside the set search areas.
This Broken Ridge region is the only region that could explain a motivated, pilot controlled ending imo.
All latitudes north and south are compliant with no pilot imputs after FMT imo.
My only concern is OI will not have the time and/or funds to go back to this Broken Ridge area to search wider.
IMO they better do it now extensively while time is still plenty and the weather allows it.
With the pace they go, this area between ~95E/~97E and ~32.2S/33S is searched within a few weeks.
@Ge Rijn: I have edited your previous comment to reflect @Richard Godfrey’s correction.
@Richard Godfrey
Do you want me to find this statement from you and repeat it here? Did you forget?
It would cost me lot of scrolling through the comments. But if you insist I’ll come up with it. Unless this was a troll using your name when he suggested the scenario I referred to; he made this suggestion MH370 passed over ~32S Broken Ridge ending in a trench around 32.2.
You were not the person stating this?
@Ge Rijn
There had been an interesting discussion about use of Z’s birthday as a waypoint, leading to 32.2S. But I was the only one taking it seriously and I moved my pin to 32.5 and I am willing to embrace 32.2S. But I am not using Z’s birthday as a waypoint.
@TBill: Yes, here was the comment:
http://mh370.radiantphysics.com/2018/01/21/the-search-for-mh370-begins-again/#comment-12431
@Ge Rijn
If you look at the Graphic at the head of this post you will see my name attached to 2 points, one around 30°S another jointly with Victor around 27°S. Neither of these are 32°S or 32.2°S.
If you look at the comments you will see:
“@DennisW – I am in full agreement with your statement. The most likely MH370 End Point is 30S +/- 1 degree.”
http://mh370.radiantphysics.com/2018/02/16/mh370-search-update-feb-16-2018/#comment-12724
Or …
“@DrB – Nearly there to 30S +/- 1 degree, that I published over a year ago.”
http://mh370.radiantphysics.com/2018/02/16/mh370-search-update-feb-16-2018/#comment-12955
and on and on and on …
But you choose to ignore all these statements!
When some time ago, on a previous post, I played mind games with an ASSUMPTION that ZS used his birthday of 31/07/1962 as an MH370 End Point to mean 31.07°S 96.2°E, you take this COMPLETELY OUT OF CONTEXT as a statement that I have suddenly abandoned all my published work and no longer hold to a MH370 End Point of 30°S +/- 1 degree. This is not the case.
@Victor if you start down the road of correcting GR’s errors of fact or interpretation you face a truly herculean task.
@Richard Godfrey @TBill
This is not the point. Godfrey again accuses me of lying and utter incompetence. A truly self-absorbed trait.
This is not the way to move forward imo.
Copy/paste:
Richard Godfrey says:
February 14, 2018 at 4:23 pm
“@DennisW
Please bear with me for a moment.
What if, it was a case of “making it happen” rather then “taking what you get”.
Assume that MH370 was a carefully planned operation by ZS.
Assume the objective was that MH370 never be found.
Assume that the final waypoint was 31° 07′ S 96° 10′ E, which would be entered as Victor points out as S3107.0E9610.0 (31.11667°S 96.16667°E).
The final waypoint turns out to be directly on the 6th Arc. This is before fuel exhaustion and just before reaching the Diamantina Escarpment at the edge of Broken Ridge.
ZS continuously monitored the distance to go and the remaining fuel range to make sure he made it to his birth date final waypoint. However, he did not want to go much further because he wanted to hide the plane below the escarpment in a gully or canyon.
A few minutes after passing the final waypoint, MH370 runs out of fuel and ends up crashing just beyond the edge of the escarpment in deep water at around 32.2°S near the 7th Arc.”
And now I make up things?
@TBill:
“I am thinking said descent happened at 18:40, to allow passing under N571. I am thinking MH370 is in the process of gradual re-ascent at 19:41”
Yes I think that descending at 18h40 to pass under N571, is highly probable. I invite you to have a look at the video we just posted where the crossing of all the routes is clearly shown on her way. If you admit that she passed under N571 for safety reasons (which we do think) then the next routes should be considered too: P574, P627 and then P756 which Minimum Flight Altitude is 15 000ft. We have clearly sketched them.
https://www.youtube.com/watch?v=Jd_eJIINlBw&sns=em
Climbing at 19:41 would mean crossing P756 within its operationnal flight levels which is unsafe and risky. If you look at flight radar tracking sites right now, you will see that there are flights between FL350 and FL380 potentially timely crossing MH370 route.
Collision risks exist.
@Ge Rijn
NO, YOU ARE TAKING THE PLAY ON THE BIRTH DATE OF ZS OUT OF CONTEXT.
@Ge Rijn, @Richard Godfrey: So Richard did mention 32.2S, but only in the whimsical scenario that the captain’s birthdate was turned into a waypoint. Time to move on.
Few people here are in disagreement with progressively searching north after the primary area is completed. Other than claiming bragging rights if an endpoint is predicted, there’s no need to argue about possible endpoints, as long as all reasonable ones are searched. And the truth is, all bragging rights have long been lost, as few here have not abandoned a prior endpoint.
@Jean-Luc Marchand: It’s heartening to hear that in your scenario, the same hijackers that forcefully took control of the aircraft and crashed into the sea after fuel exhaustion were safety-minded enough to follow ATC directives and descend to the recommended altitude while crossing an airway.
@Bobby
All you need to do is move your decimal point between the 1st 2 digits (3.15N) & 96.77E & you are more or less bang on with my area of interest @ 4.6N & 90.7E. Which ties in with an NOAA drifter buoy that covers all debris find spots.
You may find that moving your decimal point ridiculous but I am for once excited about a location in the SIO. Hopefully they will search there. Fingers crossed & great work!!!
@Richard Godfrey
You stated: “NO! I have never made such a statement!”. Which you did clearly afterall.
And with your denial you pointed to me as a liar afterwards. And I don’t take this kind of mind games for granted as you should know by now.
Just leave it @Godfrey. Your comment to @Dennis actually made a lot of sence to me.
@Ge Rijn
Victor stated “So Richard did mention 32.2S, but only in the whimsical scenario that the captain’s birthdate was turned into a waypoint. Time to move on.”
@Victor
“crashed into the sea after fuel exhaustion”
If I may, I would raise the question: would anybody hijacking an aircraft want to crash ?
They (we don’t know who by the way: were they Hijackers or the pilot ?) wanted to land safely and thus fly safely. They miss-evaluated the required fuel for reaching safely their target. Actually, nobody – not even Boeing – would be able to predict the specific aircraft engines consumption before a flight.
All simulators we have used are optimistic by about 1ton of fuel. Sufficient to do a one go-around. We had to use the operational manual to correct the simulators and cross-checked with DR Ulich’s model for a more precise computation.
Keeping the aircraft safe and invisible was an important constraint in our study. But they missed by few… 600kg
@Richard Godfrey
Agreed.
“So Richard did mention 32.2S, but only in the whimsical scenario that the captain’s birthdate was turned into a waypoint. Time to move on.”
Oh, Richard you are naughty, playing mind games with Ge Rijn. You had me fooled as well. Please, next time you are going to be whimsical, warn us in advance!
@Jean-Luc Marchand: I’m sorry, but your scenario makes no sense to me in so many ways. Let’s talk fuel consumption. There was plenty of fuel to reach Christmas Island. You believe that in order to evade radar, they flew slow and burned excessive fuel. If their intent was to land on Christmas Island, why wouldn’t they simply choose a fuel-efficient speed and altitude? Invisibility wasn’t a concern because they were planning to land on Christmas Island. They certainly didn’t choose to fly low and slow over Malaysia and Thailand, where the threat of an intercept was greatest. An intercept from Indonesia wasn’t a concern because they did not fly over Indonesian territory when they skirted around Sumatra. Instead, in your scenario, they chose to fly in an unsafe manner, and ran out of fuel.
@Rob
The comment was to Dennis, prefaced by “bear with me” and started:
Assume …, Assume …, Assume …,
@Jean-Luc
I appreciate your work, but I am still thinking the current search area is correct…Re: Descent- of course Victor always had that idea, but I was proposing a variation (in this blog space) which was a descent to FL280 to cut over to DOTEN from IGOGU, consistent with a pilot following late nite flight path rules and perhaps faking a trip north per simulator studies. However, when you showed a descent to cross under N571, that made a lot of sense to me, and works better for my path timing. So I borrowed that from you for now.
My thinking (re: flight path crossings) is very few flights that time (except N571 is slightly busy) and why couldn’t the pilot use a 500-ft odd level for path crossings, not to mention TCAS as you suggested? We certainly do have the apparent truth that the aircraft must have cut through all of those flight paths to go South.
@Richard,
I do recall your original post on the subject, and I think I even commented at the time that I tried various similar combos myself. We all use tricks to remember things. Even passwords are a nightmare for me. I must have some 30 or more – everything from brokerage and bank accounts to Facebook, Google, and Yahoo. It is overwhelming, and I think it is a very logical thing to assume that someone picks something easy to recall as opposed to something written on a scrap of paper in their pocket.
I also found Richard’s idea intriguing that Shah could’ve chosen a specific crash location according to very personal associations. It would be his final resting place after all. I reconsidered when Dennis pointed out to me that a strictly target-oriented crash location isn’t really compatible with a fuel-exhaustive flight. It’s almost impossible to coordinate fuel exhaustion with the arrival at a specific location. In this case it would be far more sensible to reach the targeted crash location in the most fuel economic flight mode and then circle until the remaining fuel is exhausted.
Victor has added a new wrinkle, though, with his suggestion that Shah might’ve induced the second reboot of the sat com with another power cycle in order to signal the end of the flight, even before the fuel tanks were completely empty. This scenario would make a specifically targeted crash location a bit more plausible…
Honestly, theorizing was a lot less complicated when we all still assumed that the leg into the SIO was an unpiloted ghost flight 😉
@Sabine
As Victor pointed out, the 777 instrumentation could keep a pretty good estimate of fuel remaining at the arrival of a selected waypoint. It is possible that Shah could have loitered just about anywhere until the fuel remaining was compatible with his selected endpoint.
@Dennis, this might be true – but wouldn’t it be nevertheless more sensible to loiter and use up the remaining fuel after you have arrived at your chosen destination?
Occasional readers of this blog who aren’t well acquainted with the MH370 conundrum must think we have gone totally crazy, since we are seriously discussing these bizarre scenarios…
@ Dr B,
That’s an impressive piece of work, and truly impressive perseverance with 1/2 to 1 hour of computer processing time for each iteration.
As you know, I have a particular interest in the “Final 15 Minutes”, and you have but a few brief comments on this phase in 8.8 of your paper.
You have a couple of comments that allow some ambiguity:
“In addition, when the left engine also failed at 00:17:30, a bank developed and the course most likely changed by 00:19 in a way that is not predictable.”
and
“Thus, generally speaking, the 00:19:39 position is directly on the 7th Arc and is very close to a simple continuation of the current route.
I think that we agree that the aircraft speed [and altitude] will gradually reduce after 00:11, but it is uncertain what happens to the speed [ . or more particularly the ground speed . .] after the second engine failure.
If the ground speed continues to decrease after the second engine failure, and a bank to the left develops, then the likely end-point will move further NE along the 7th arc. Indeed, the 7th arc may not be reachable with a “simple continuation of the current route” [track]. And a bank to the right would not allow the 7th arc to be reachable either.
I might suggest that your end point could be moved a further 25NM to the NE before applying the error band of +/- 30 NM or so.
@DrB
I have read your paper a bit more carefully than a first pass. I am a bit disappointed in your characterization of BFO error. You might be interested to know that not a single manufacturer of oscillators uses an RMS metric to characterize performance. Not a single one. I would challenge you to find one. Yet you cling to this metric. Why?
The reality is that that BFO is not useful as a a quantifier of flight paths. When I read this in the DSTG paper (“book”) and others, I basically tune out.
@Brian Anderson, Dr B. I doubt the aircraft will go left at fuel exhaustion. That assumes no hydraulic power but that from the RAT. However a recent discussion here and a simulation by @Andrew indicate that the windmilling engines should provide both L & R hydraulic power. Also, the backup generator PMGs should power ACEs other than the L2. Bear in mind too that we are talking about 2 minutes to log-on and very low demand; and there are small flight control hydraulic acccumulators also.
As to a right bank and turn, the impetus for that most likely would be yaw from left engine asymmetric thrust, the aircraft being without A/P and TAC. That could be from a relight from below idle after it had suffered fuel exhaustion; and supposes the engine had been generating backup power, at least.
However that might not cause enough bank to realise the BFO descent and its timing. Had the right engine been providing all AC power when it had failed, the left would be at cruise power still, again without A/P and TAC. Pending the APU coming on line a minute later (that leading to SDU reboot) the left would draw suction fuel from its main tank. On restoration of AC power by the APU the boost pumps would be repowered. (In that scenario the left engine obviously would provide L hydraulic power irrespective of that from windmilling.)
At main tank fuel exhaustion the APU DC pump would provide a limited supply for a relight from below idle; the same if the suction fuel supply failed.
I did some back-of-the–envelope/indicative calcs to give an idea of turn radius and 7th arc timings, those being based on the bank needed for the BFO descent rates (IMO). If manned there need not be that bank.
http://mh370.radiantphysics.com/2018/02/16/mh370-search-update-feb-16-2018/#comment-12860
I don’t think there’s been comment here on Paul Allen’s Petrel locating the USS Lexington. In the past 2-3 years, they’ve worked through their list of WWII capital ships, also bagging the USS Indianapolis and the battleship Musashi.
No doubt that the likes of Allen or James Cameron will eventually check out any particularly tempting MH370 flagpins. So if nothing happens in the next month or two, it’s open season for whackos.
[JW excluded: the Petrel isn’t registered for duty in the Aral Sea.]
@ David,
I re-read your earlier comment about turn radius etc. If the left turn intercepted the 7th arc too early, then perhaps the radius was just a little greater and the intercept occurred a second time, just a little later, but in the other direction.
@Ikr, if the OI search will not be able to find the plane, I fully expect that people with the monetary means and the necessary technical equippment will try to find the plane for years – maybe even for decades to come. Solving the greatest aviation mystery of all time has already turned into a veritable quest four years after the plane vanished.
Both ships of 3 Franklin’s lost arctic expedition, the Erebus and the Terror* have been found near the shores of King-William Island in the last couple of years. The search has been going on for more than 150 years! Interestingly the ships could’ve been found long ago if the searchers had paid more attention to traditional Inuit eye witness reports. They have been located in shallow water and weren’t really that hard to find – if one were able to deviate from pre-conceived ideas. The recorded oral accounts of the Inuit have been fully validated, and the locations of the ships force Franklin scholars to re-evaluate their scenarios of the expedition’s last months. The examination of the well preserved wrecks could eventually answer more questions.
*I wonder who named these ships. They were doomed before they even set sail 😉
@Brian Anderson. Yes I follow. However I have heard no rationale for a left turn other than the flaperons of earlier, now discarded by me at least but also others I think.
If the right engine outlasted the left and the left carried all electrical load at its fuel exhaustion, the right would yaw the aircraft that way. However @Dr B has it failing first as would all I think unless both engines ran on the left tank for a while (ie it was piloted) and there would seem to be little point to that. I do not think the electrical load difference borne by left from the first log-on would be big enough.
I included the left only because a rationale might yet emerge.
@Sabine, re. your comments iro the Franklin Expedition ships:”The search has been going on for more than 150 years! Interestingly the ships could’ve been found long ago if the searchers had paid more attention to traditional Inuit eye witness reports. They have been located in shallow water and weren’t really that hard to find – if one were able to deviate from pre-conceived ideas.”
You touch on a profound subject. It’s more convenient (and in the short term, more resource efficient) to assume someone elses initial assumptions are correct, than to go against the flow and question them. The result is that the initial errors are compounded to the point that they become set in stone. History is replete with examples. Basically, you can put the “blame” for this on the evolutionary development of the human mind, but 150 years is too long have to wait for MH370 to merely provide yet another example of it, don’t you think?
@Rob, thank you.
The history of the search for the ships of the Franklin expedition teaches indeed a few valuable lessons: my main takeaway is, that Occam’s Razor can’t really be applied succesfully when deliberate human actions come into play, since the potential scenarios multiply with the number of human agents and their potential goals. And in unprecedented situations humans don’t necessarily act in a simple or easily predictable way. In the case of the Franklin ships it has been dogma until fairly recently that the ships have been abandoned only once after they were frozen in, and that they probably sank near their last recorded position while drift parameters have been taken into account. But the Inuit have always told a far messier story. They insisted that at least one ship had been re-manned, lived in, and was eventually taken down the coast of King William Island. And lo and behold – after the second ship was found just by chance and far off the beaten track (because someone had reported to have seen a mast several times at that location), it turned out that the Inuits were right all along. The ship couldn’t have drifted to it’s final location – no way. It must’ve been re-manned and moved deliberately. And that means: many long held notions of what happened in the final months of the doomed expedition must be wrong, and some survivors may have been alive and acting far longer than previously thought.
It’s difficult, though, to apply the lessons learned from the search for the Franklin ships to the search for MH370. It was of course a huge mistake to cling for so long to the notion that the plane was a victim of a bizarre disaster and that the last leg into the SIO must’ve been a ghost flight. That said – what could’ve been done differently? The main crux is of course that as far as we can tell, there has never been a thorough and impartial police investigation. But the only chance to actually find the plane is scanning the sea floor along the 7th arc at areas of special interest (the so-called hot spots) until the plane is found. If the plane won’t be found by the current searchers the width and the length of the search corridor needs to be expanded – if there’s still the will and the money for continuing the search.
Of course you have your specific scenario and you wish that your predicted crash location should be included in the search. But others are equally sure that the plane must be at their predicted location. The plausibility of the many pet scenarios varies greatly. Some can be weeded out immediately IMO. And it’s probably not possible to cover them all during the current search.
Personally I’m not all that optimistic right now. If Shah – or whoever piloted the plane – really was alive and consciously interacting with the plane, there are many possibilities how he could’ve defied our expectations, even if the plane did crash after a steep descent somewhere near the 7th arc.
If the sat data are valid, the plane could well be found eventually, since people will keep searching. The search for the Franklin ship shows how persistent humans can be when something has sparked their imagination. But it can take a very long time…
@Sabine
Iro “the notion that the plane was a victim of a bizarre disaster and that the last leg into the SIO must’ve been a ghost flight. That said – what could’ve been done differently?” What could have been done differently? Well he said (donning his rose tinted glasses) for a start the Malaysian government could have played the honest broker and simply admitted in public that one of their pilots had apparently gone rogue (a taboo subject among the airline business) and the evidence suggested that he had deliberately flown deep into a part of the ocean that was extremely unlikely to ever be identified as a possible resting place for him and his plane. Along With the much needed acceptance of malign intent would have come the ability to throw out the by now patently flawed uncontrolled descent theory, and replace it with something a bit more in keeping with the evidence: an extended, piloted glide.
Now removing rose tinted glasses…
An impartial police investigation is a contradiction in terms, pure and simple. The police are an arm of the government.
However, I remain more optimistic than you Sabine. OI are doing their job at a cracking pace. Let’s wait and see, popcorn in hand.
@All: The fourth interim statement was released by Malaysian officials. There is nothing of any significance, as expected. (The Malaysian server is slow, so have patience.)
@Rob, sorry, but there’s no hard evidence whatsoever for an extended glide, but there’s definitely evidence for an accelerated descent.
It’s of course not totally impossible that the pilot recovered the plane in order to enter into glide, but evidence? No.
I’m not even saying that a final glide is per se an unlikely scenario, but I don’t believe in a glide after the plane has been recovered. That doesn’t make any sense to me whatsoever – unless Shah/the pilot knew about BTOs,BFOs and the 7th arc. In this case he might’ve executed this unusual maneuver in order to mislead the investigators.
I think it’s time to revisit these kind of scenarios if the current search comes up empty.
@VictorI
Only one statement stood out to me:
“5. Aircraft debris possibly from MH370 are still being discovered around the southeastern coast of the African continent and the adjacent islands.”
I’ve seen no new reports/news on found debris for a ~year so I wonder what this is about.
@Ge Rijn: I don’t know what new discoveries of debris there might be. Considering the suspicious occurrences in Madagascar, I would be surprised if there are any more efforts to recover debris there. In fact, the base of the vortex generator was never handed over to Malaysia, as it is evidence in the investigation of the death of diplomat Zahid Raza.
@DennisW
It is interesting that DrB found a magnetic heading path, when the prior attempts could not find a magnetic path solution (Victor had a solution with continuous descent). I have not checked DrB’s BTO’s but BFO fit is fine.
@Sabine L
RE: “It’s of course not totally impossible that the pilot recovered the plane in order to enter into glide, but evidence? No.”
Conclusive evidence, no. But the flaperon, right wing outboard flap section, engine cowling parts (especially the big right-hand cowling piece with intact front edge), left wing trailing edge piece of the outboard flap, the left and right wing flap-fairing pieces, the left and right wing flaperon closing panels, the right wing aileron piece, the nose-gear door piece and the lack of crush-damage but (almost) all tension/pull related damage indicate a relatively low speed, nose-up AoA impact comparable with a ditch-impact.
The found interior pieces prove the cabin must have been breached at least at one location which is almost unavoidable during a ditch in the ocean I think. Even the Hudson-ditch had his bottom tail section severely breached during the ditch.
There is no conclusive evidence of a ditch-impact but as long the above is not conclusively proven to have seperated otherwise it’s still the best and only explanation imo.
I apologize if this has already been posted. The goal of this project is to map the entire ocean floor by 2030. Ocean mappers line up for XPRIZE final
https://www.yahoo.com/newsroom/vibes/science/ocean-mappers-line-xprize-final-56c2e39?soc_src=newsroom&soc_trk=com.apple.UIKit.activity.Mail&.tsrc=newsroom
@Trip: Here’s a better link for that article:
https://www.bbc.com/news/amp/science-environment-43317417
Readers here might recall a previous post that discussed the approach that the Virginia Tech team is pursuing using a team of small, inexpensive AUVs. I’m happy to say that this team has advanced to the final round of competition.
@Sabine
“@Rob, sorry, but there’s no hard evidence whatsoever for an extended glide, but there’s definitely evidence for an accelerated descent.”
I agree. There isn’t hard evidence. The best evidence for an extended, piloted glide, evidence that gets stronger by the day as the search progresses, is that the aircraft hasn’t come down near the 7th arc. This same evidence also calls the BFO accelerated descent = uncontrolled descent interpretation into doubt, or at least forces a reassessment.
I am not going to labour the point, but as Dennis points out, the handshake and phone call BFO’s are good for one thing and one thing only – showing that the aircraft went south instead of north. They are indeterminate when it comes to measuring horizontal velocity, and applying RMS as a useful statistical tool is pointless, and misleading. The BTO data, OTOH, is in another league. The BTO strongly suggests an aircraft flown in LNAV great circle mode from 1822 all the way to fuel exhaustion, with the absolute minimum of manoeuvring, a straight line from FMT to fuel exhaustion. The aircraft should, according to the BTO and the fuel range/endurance (not to mention the sunrise) be at S38. However, it isn’t there, and that’s the best evidence, albeit indirect, for an extended glide.
We can argue about the significance of the debris until hell freezes over, and not arrive at a conclusion acceptable to the majority, so what’s the point? The only thing that should be noted is that retracted flaps don’t rule out a piloted ditch. The ATSB have made this error, and OI are searching empty seabed as a direct consequence.
I’m going to leave it there. Let’s see how the search pans out
@Victor
I’ll tell you, this OI search is very exciting. I thought we’d have to wait forever for them to plod along but no, holy mackeral, they are in thick of volcanos and trenches way quicker than I can keep up.
Today’s update courtesy Kevin Rupp shows OI going “out-of-bounds” where I was hoping they would go, possibly close enough to send a AUV over to Joe Nemo’s back scatter spot. They seem to be thinking like I would checking that trench feature.
https://pbs.twimg.com/media/DXwiVaoXcAEtOLR.jpg
PS-
You can see the graphic above is just Google Earth, and you can clearly see there is a zone of very high sea bottom detail on Google Earth. For the BR hypothesis, that zone is a possible hot spot, although I’d claim anything from the BR cliff. If it’s on the BR plateau, the the BR hypothesis was wrong.
@TBill: I did inquire about that spot that Joe Nemo found. The experts believe it is likely part of a large rock slide associated with a ridge line. In any event, I am assured that it will be closely reviewed after it is scanned.
@TBill: It also appears that the search is not slowed down very much by Broken Ridge. That’s really great news if that can be sustained.
Bobby Ulich’s latest paper, shows a constant magnetic track southwards, following a deviation northwards to waypoint AGEGA, resulting in a MH370 End Point at 31.57°S. Bobby has come a long way north from his original MH370 End Point at 40°S. Bobby was stuck on 35°S, as recently as the start of the new search with Ocean Infinity. Despite changing his MH370 End Point several times, he now claims that his latest result is the only valid one and all other results from other people are wrong?!
Victor and I published a paper back in August 2016, which showed a MH370 End Point around 27°S. Bobby determines this is an unacceptable candidate according to his route fitting criteria. I published a paper back in February 2017, which showed a MH370 End Point around 30°S +/- 1 degree. Bobby has dubbed this as not flyable by a Boeing 777-200ER due to fuel considerations. If you can get to 31.57°S, why can’t you get to 30°S?
Victor and I have double checked our calculations and cannot find the errors that Bobby is eluding to. Just for the record, the criticism from Bobby in his paper, concerning meteorological data and TAS are nonsense. Bobby is aware, the YWKS flight model has been revised and published several times last year, most recently in September 2017. Bobby is also aware that since January 2018, I use 40S 100E as the ultimate waypoint and not YWKS, as it is not certain that YWKS was in the MH370 navigational database.
I agree with Victor’s comment to Bobby “I disagree with your assessment of the uniqueness of your proposed endpoint of 31.57S (in much the same way I disagreed with the uniqueness of your previously proposed endpoints)”.
Meanwhile, Ocean Infinity have covered up to 32.4°S inside the 7th Arc. I can’t help but feel that Bobby is jumping on the “Northern Bandwagon” at the last minute, now that the more southerly areas have been searched and MH370 has not been found.
@TBill
Nice graphics. And amazing work indeed by OI/SC. It seems they are searching every hole also outside the detailed GE-track.
Unbelievable the pace they are going and how well their technology turns out to work without setbacks as could be expected (with new technology).
I hope they take their chance now searching this trenches all the way to ~97E. But most of all ofcourse I hope they find the plane right now.
@TBill:
You said: “What is the north/south military radar track along about 93.75E marked by the “x” on your flight path?”
The larger “X” in my Figure 19 is just the point 15 NM west of AGEGA. It marks the northern end of the lateral offset line from AGEGA to BULVA. It is not the “military radar track”, which is shown by the small red X’s.
Regarding the possibility of a descent, that might have happened, but the 18:40 BFOs must be satisfied and the path needs to be 180.0 degrees true at some time before 19:41. Kate Tee’s low-flying aircraft must have been no higher than about 10,000 feet and much closer to her (very near NOPEK) than 40 NM. It may be possible to satisfy all those conditions, but I don’t see how you can get 180 true track far enough west except by AGEGA to BULVA. If you accept that, then there isn’t time to fly close to Kate Tee and then go to AGEGA.
@Brian Anderson,
You said: “I might suggest that your end point could be moved a further 25NM to the NE before applying the error band of +/- 30 NM or so.”
I was tempted to do that. However, the best-fit course is only about one degree to the left with a ground speed error of -3 knots. The CMT route hits the 7th Arc much closer to perpendicular than the straight routes, so you don’t need a significant turn for the speed to be reasonable.
@DennisW,
You said: “The reality is that that BFO is not useful as a quantifier of flight paths.”
I agree that the BFOs are of limited usefulness when it comes to precise route fitting. I said that in my paper. They tell us the plane went south, and they are useful in eliminating routes which have large BFO residuals. The route fits within a region of interest are totally driven by the BTO residuals, and the BFO errors don’t contribute anything significant over a degree or two in bearing.
I see the RMS statistic for the BFO residuals because it is a simple and useful fitting tool when one wants to reduce both systematic route errors and the dispersion of the residuals. The former is much more important, and it is certainly difficult, if not impossible, to discern between possible oscillator drift and systematic route errors when they are small.
@Richard Godfrey,
You said: “Meanwhile, Ocean Infinity have covered up to 32.4°S inside the 7th Arc. I can’t help but feel that Bobby is jumping on the “Northern Bandwagon” at the last minute, now that the more southerly areas have been searched and MH370 has not been found.”
That is nonsense. My result is unbiased and is independent of all search results. The calculations come out where they come out.
You said,” I published a paper back in February 2017, which showed a MH370 End Point around 30°S +/- 1 degree. Bobby has dubbed this as not flyable by a Boeing 777-200ER due to fuel considerations. If you can get to 31.57°S, why can’t you get to 30°S?”
It’s not that you can’t fly to 30S, it’s that you can’t fly to 30S at M0.848 in a B77, as you originally proposed (or even at M0.84). The fuel mileage is too low. That is not nonsense. It is a fact. In addition, the BTO errors are too large at that speed.
You also said:” Bobby is aware, the YWKS flight model has been revised and published several times last year, most recently in September 2017. Bobby is also aware that since January 2018, I use 40S 100E as the ultimate waypoint and not YWKS, as it is not certain that YWKS was in the MH370 navigational database.”
I’m amazed that you know what I am aware of, and not aware of. In any case, my comments were made to the specific papers you wrote that I have referenced in my paper. If you have revised those previous results, then would you please post a link to the specific newer results. I did look in Victor’s articles and didn’t see a link to anything other than those I referenced. I’ll have a look at your revisions and then post my results.
@Rob re “We can argue about the significance of the debris until hell freezes over, and not arrive at a conclusion acceptable to the majority”
Of course with the superficial analysis done so far. The investigators do not want to perform a detailed analysis which, without any doubt, will inform the end of flight scenario and part of what has actually happened even if the plane is not found. I personally dont believe this is to do with incompetence, it is an easy task. Or at least they do not want to make that information public like some other pieces of evidence. Who knows. Let us hope this appears in the next official report, but i am afraid it would be too late.
@DrB
Thx for the reply. I actually regretted my comment to you and a related comment to sk999 earlier relative to BFO. We all have “hot buttons”, and that is one of mine. I realize your interpretation is reasonable. I just feel historically that an over-constrained interpretation of BFO has been one of the largest contributors to modeling errors.
@DrB
The original paper was updated to 0.84M soon after Feb 2017 and published.
However, your main complaint was an endurance shortfall of 42 minutes.
As a frequent contributor you are aware what gets published here.
The BFO do not show too high an error.
Perhaps your error limits are wrong.
@DrB
OK. Maybe the key on Fig 19 could be clarified by putting the larger “x” on there next to red “x” so the reader can see the difference.
The other comment, is, you are sort of saying in there that the search below 32.85 was unsuccessful, which might be accurate in a few weeks, but the search is not completed yet. Victor still gets a 3rd chance at SPOLE waypoint, and you know what they say about 3rd chances.
@DrB, @Richard Godfrey: Surely great circle paths ending near 30S latitude satisfy the BTO by any measure and probably satisfy the BFO, depending on what we believe are the error limits. You can easily see this from Figure 2 in this post. For the path towards YWKS (or towards 40S,100E, doesn’t matter much), the RMS errors for the BTO and BFO are 31 μs and 5.7 Hz, respectively for LRC speed at FL350, with a position at 19:41 of (6.8127,93.3176), an initial track of 172.6°T, and a crossing of the 7th arc at 29.7S latitude.
@Ge Rijn
OI continues further focus on the high def focus zone of BR region in Google Earth
https://pbs.twimg.com/media/DXx-8oRX0AA8UKy.jpg
(courtesy Kevin Rupp)
@Rob, Sabine, HB
“We can argue about the significance of the debris until hell freezes over, and not arrive at a conclusion acceptable to the majority”
The subject is not up for popular vote.
@HB, specifically.
I reject your comments, above, entirely. I do expect that comprehensive examinations have been made of the flaperon and the flap, and that the findings have been presented to the lead investigation team. However, the flow to the lead investigation team is largely one-way. It seems their practice only to disseminate anything if it is absolutely irrevocable.
ATSB made public one finding after their inspection of the flap: that it was retracted; they made no reference to the fractures of the aft link, nor the flap carrier. I really can’t imagine that the ATSB team ignored those fractures.
@TBill,
I could have been more specific and said the search up to 32.85S has been unsuccessful within about 10 NM of the 7th Arc. It will remain unsuccessful even out to 25 or 100 NM. The aircraft is farther north.
@all
The fourth interim statement from the Malaysian ICAO Annex 13 Safety Investigation Team for MH370, released on 8 March, is available here. There is nothing new, however, the last paragraph clarifies their intentions with respect to the final report:
“The publication of the detailed investigation report, that is the Safety Report, based on currently available information has been suspended pending the outcome of the latest search effort, since any new evidence uncovered is likely to significantly affect the investigation. In the event that the aircraft is found, the Team will conduct further investigation. If the aircraft is not found and a decision is made to discontinue the search, the Team will resume the completion of the report and release it in the months ahead.”
@DennisW,
My results for the different lateral navigation methods would not be changed significantly if Iignored the BFOs altogether, except to pick the right hemisphere. I trust Inmarsat on the BFOs, since they are in the best position to know. They said:
“Figure 15 shows the measured BFO for flight MH21 that travelled from Kuala Lumpur to Amsterdam at the same time as the MH370’s final flight, overlaid with the upper and lower BFO predictions (± 7Hz) using the refined BFO model. This suggests that ±7Hz is a conservative estimate of the typical accuracy BFO calculation achieves, as well as illustrating the BFO versus time characteristics for a flight moving along a path close to the Northern route. While the validation demonstrates the general accuracy of the BFO technique, it is important to note that agreement is only achieved with ±7 Hz accuracy during this flight, and to assume better accuracy for the measurements taken on MH370 would be unrealistic.”
They believed, and rightly so in my opinion, that the BFOs for a single flight would have an error not larger than 7 Hz. That might be a better criterion for judging a particular route than using RMS or standard deviation or mean error. For my 181.2 degree CMT route, the largest BFO error is -4.8 Hz, so that is acceptable according to Inmarsat. The mean is -2.4 Hz, the standard deviation is 1.9 Hz, and the RMS is 3.0 Hz.
The LNAV routes from 166-173 degrees have peak BFO errors of 11 or 12 Hz. By Inmarsat’s criterion, all these routes, including the Antarctic airfields NZPG and YWKS, as well as the 40S/100E waypoint, are unacceptable. These are the routes on the left side of my Figure 11b.
As I showed on the right side of my Figure 11b, there are also LNAV routes near 182 degrees that have acceptable BFOs, but the BTO errors are too large in this case. The bottom line is that none of the LNAV routes are “correct.” The either have BFO errors that are too large, or they have BTO errors that are too large. Even if you decide to ignore Inmarsat’s criterion, which Victor and Richard Godfrey are doing, there is no central peak in the figure of merit curve versus bearing that indicates the correct route. In short, MH370 did not fly a great circle after 19:41.
@DrB said: Even if you decide to ignore Inmarsat’s criterion, which Victor and Richard Godfrey are doing, there is no central peak in the figure of merit curve versus bearing that indicates the correct route.
The fact that there is no peak doesn’t mean that there is no correct route. However, if the peak BFO error has a hard limit of 7 Hz, that would be a further constraint. For the BEDAX-SouthPole route, the peak BFO error is 7.7 Hz at 19:41. A small climb solves that. For the GC routes, most of the RMS error is due to average error and not deviation from the average. That would suggest a change in fixed frequency bias occurred.
You are again sure you found the unique path. You could be right this time. I am much less certain. We’ll know soon enough.
@Victor @DrB
“…the peak BFO error is 7.7 Hz at 19:41.”
DrB’s path also has the same error at 19:41 and the reason is probably that the aircraft was still maneuvering at 19:41, as far as I am concerned. There was probably an ascent in progress 19:41, but maybe a turn.
I am doing a short essay on this. I call it the BFO paradox. A straight level flight we should expect a curved BFO plot from 19:41 to 21:41. Instead we actually have a straight BFO plot, meaning, the flight path was *not* straight and level at 19:41.
@DrB
Yes, the Inmarsat 7Hz statement, for the KL to Amsterdam flight was available early on. I was never happy with the conclusion reached by the collective relative to that statement, and ALSM and I debated it at lenght. Later we were made aware of the BFO’s from the Mumbai to KL flight shown in Figure 5.4 of the DSTG book. This figure shows errors much larger than 7Hz. What this result indicates is that the BFO statistics are not stationary. You will get different statistics for every flight you sample. It is for this reason that oscillators are not characterized in terms of mean, peak error, or standard deviation. Inmarsat was wrong to make the 7Hz statement, and the DSTG was wrong to even calculate a BFO variance. Both are meaningless.
@TBill
I am doing a short essay on this. I call it the BFO paradox. A straight level flight we should expect a curved BFO plot from 19:41 to 21:41. Instead we actually have a straight BFO plot, meaning, the flight path was *not* straight and level at 19:41.
I don’t think this is going to pan out for you.
@TBill,
You said: “@Victor @DrB
“…the peak BFO error is 7.7 Hz at 19:41.”
DrB’s path also has the same error at 19:41 and the reason is probably that the aircraft was still maneuvering at 19:41, as far as I am concerned. There was probably an ascent in progress 19:41, but maybe a turn.”
My CMT route has a BFO error of -4.8 Hz as shown in Table 3, not 7.7 Hz. There is no need for an ascent then or a turn.
@VictorI,
You said: “The fact that there is no peak doesn’t mean that there is no correct route.”
Your statement is confusing, but I think you are trying to say that the correct route may not be on top of a narrow peak in the figure of merit. On point that I strongly disagree. There must be a peak at the correct route. Just because you have never found a narrow peak does not mean that it doesn’t exist. It certainly does exist, as I have demonstrated, but only when you fit the correct route or a very, very close facsimile. There is no other explanation for it besides the coherence of the 5 BTOs. That is really what defines the correct route.
In the MH370 case, it coincidentally happens that, because of the winds, a CTH route closely mimics a CMT route. This local peak is found in both cases, CTH and CMT, even though the lateral navigation calculations are totally different, and it turns out the peak is for the same physical path. The only common element in the two fits is the 5 BTOs. Embedded in them is the route information. In addition, it is very precise in position and in bearing because of the route curvature, which destroys the rotational symmetry of the problem. The large separation of the BTO arcs also contributes to a surprisingly narrow angular tolerance (a fraction of a degree).
Up to now, we were satisfied to simply require BTO and BFO error statistics that matched our expectations. As you (and I, and others) have demonstrated, many routes can do that. There is no unique solution if that is your only requirement. What I discovered is that those conditions are necessary, but not sufficient, to determine the correct route. There is another condition, that there must be a narrow peak in the figure of merit at the correct route. This peak must occur because, when the route is incorrect, the data noise affects the route fit so as to produce smaller residuals (smaller than expected). That’s what you are doing, minimizing the residuals. However, when you are exactly on the correct route, the residuals are increased to the expected value because only then are the route parameters and the noise completely independent of each other. At nearby bearings (or speeds, or . . .) the route parameters are unhindered to better fit the noise itself, leading to smaller residuals.
It takes a while for this concept to sink in. It took me almost four years to understand it. Just think about it abstractly for a while. What could possibly create this local peak which is a fraction of a degree wide and which is for a path that is fixed in location on the surface of the Earth, except the correct route information which is “buried” in the 5 BTOs. It’s like looking through the holes in a stack of Hollerith punch cards (I am showing my age here). Take a stack of 5 cards and slide them around one at a time until you can see a clear path through the stack. That alignment is the route information and the hole you see is the local peak in the figure of merit. All the cards have to be lined up to see it. Move one card and the hole disappears. That means that all the route parameters need to be correct in order to see the hole (i.e., to see the FOM peak). That’s why I had to search all possible routes consistent with the B777 capabilities and with the 00:17 MEFE. Quite frankly, if we didn’t know MEFE, I don’t think the correct route would ever be found analytically. That allows a great simplification in reducing the number of speed and altitude combinations that have to be searched. Even with this reduction, it took me about 5 weeks of calculations working 7 days a week to cover the remainder.
@VictorI,
Comparing a test case is a useful way to check calculations. I see that your 19:41 position is 100 NM south of Richard G’s YWKS position! That is a big difference, and the different 19:41 position and the lower speed help in reducing the fitting errors, but there is a catch in the PDA (which I discuss below).
For your 19:41 position (6.8127N,93.3176E), speed (LRC), and initial bearing (I used 172.600, but I am not sure if you used 172.600 or 172.55??), and assuming zero along-track wind error, I get the following:
From 19:41 – 00:11:
BTO Errors: 35.2 microseconds RMS (you get 31)
My BTO errors are -30, 27, 17, -56, 34.
BFO Errors: 6.3 Hz RMS (you get 5.7).
My BFO errors are -6.0, -3.1, -5.8, -8.8, -6.6.
BFO Errors: 8.8 Hz Peak (you didn’t say)
00:19:39 position (with no deviation in course and at 20,000 feet): 29.7998S 98.4010E (you get 29.7S)
SAT Deviation (C): 10.2, 10.8, 10.9, 11.2, 8.1
Along-track GDAS Wind Speed (kts): -8.9, -4.8, -0.5, 3.3, 10.0
So, our RMS BFOs are fairly close (6.3 Hz vs. 5.7 Hz). The BTOs are a bit different (35 Hz vs. 31 Hz). I suspect that difference is due to using different temperatures and along-track winds. I gave my wind and temperature values above at the handshake locations. If you posted your values, we could see what is causing the slight BTO discrepancy. I don’t think it is the BTO models; more likely it is environmental, causing slightly different average ground speeds. But the disagreement in the calculations is not serious, and it does not change the conclusions in a significant way. Inmarsat and I think that a 9 Hz BFO error (and a 6 Hz RMS BFO) is unacceptable. You do not. I also think that a 35 microsecond RMS BTO is too high, since it is 20% higher than 29. If your 31 microseconds is correct, that would be getting much closer to the expected value.
For MEFE at 00:17, I need a PDA = -1.7% (with no descent and with Air Packs On), which is 3.2% better than the nominal PDA = 1.5%. A descent/Hold/climb can save about 0.5% in fuel. Turning the Air Packs off early can save 2.5%. So, a descent/Hold/climb and Air Packs off together can save a total of 3.0%. That gets you close enough to MEFE at 00:17, but you must have the Air Packs turned off before 19:41, and they must stay off the remainder of the flight. That’s why my LNAV plot Figure 11b shows larger errors at 172.6 degrees than your test case, since I assumed in my paper the Air Packs were On. That produces a different solution, one with larger BTO and BFO errors.
@Richard Godfrey,
It doesn’t matter whether you use M0.84 or M0.848 to YWKS (or 40S,100E) because, as it turns out, the fuel is inadequate to fly until 00:17 in either case (as I said), even with Air Packs off. So your proposed YWKS route can be eliminated altogether.
None of the LNAV routes ending north of 32S, including all you have published, meet Inmarsat’s criterion of 7 Hz maximum BFO error. They are the most knowledgeable source of BFO behavior. None of the LNAV routes meet this criterion and also have acceptable BTO errors.
It’s not that my error limits are too stringent. It’s just that none of the LNAV routes are the correct route.
@DennisW,
You said: “Inmarsat was wrong to make the 7Hz statement, and the DSTG was wrong to even calculate a BFO variance. Both are meaningless.”
I suspect Inmarsat based their statement on a lot more experience and data than the single flight they showed as an example in their paper, but they did not expound on it. I wouldn’t go quite as far as you did in saying the statistics were meaningless. However, I will say that my results would not change if you used just the BTO residuals for the figure of merit in route fitting, ignoring the BFO residuals altogether. If you look at my Figure 11, the conclusions are unchanged. Only two lateral navigation methods have central peaks (in the BTO errors), CTH and CMT, and they are for the same path (the correct route). So, as I said previously, the BFOs tell you which hemisphere the solution is in, but the BTOs tell you exactly where the correct route is in that hemisphere. The difficulty is that you have to search for it using trial and error (at least I was unable to develop an automated method). You have to guess the correct route/speed/altitude to find the central peak within the region of interest. If you let the bearing vary in the fit, you are guaranteed not to land on the peak.
Hi everyone,
Yesterday it was 4 years ago that this atrocity took place. (Apologies Victor for posting an unscientific comment here) thoughts were with NOK, their life sentence in this nightmare. Like all of us, I had so hoped the fuselage would be found before yesterday.
@Donald said “He snuffed out the lives of the very people who were blindly trusting him to deliver them to safety. I say this only to impress upon readers of this blog that this is a man with significant psychopathy and that it is unknowable as to how he would choose to fly the aircraft”, is so well articulated and true to the core.
@DrB
I am not suggesting that your BFO assumptions have a material impact on your results. I simply do not know since I have not attempted to replicate your work. I will take your word for it for now.
The Inmarsat 7Hz statement flies in the face of figure 5.4 of the DSTG book. That data was obtained from 9M-MRO a few days before the diverted flight. The DSTG book also computes an RMS value of ~5Hz BFO error for some 20 flights of 9M-MRO. A 5Hz RMS value is not compatible with a peak value of 7Hz. Your 7Hz assertions basically ignore the data published by the DSTG. That is OK. You are free to do that, but you should say so, IMO.
The only points I am making are:
1> The published data regarding BFO accuracy is not at all consistent
2> The use of Gaussian stats (mean and variance) to characterize a non-ergodic and non-stationary process is fundamentally wrong.
It is well known that oscillator behavior cannot be characterized by mean and RMS metrics. No one in the oscillator industry uses these metrics. The Allan variance (or modified Allan variance) is the commonly used metric. The reason is that the variance in frequency error increases over time, similar to a random walk. It is also true that a zero mean is a rare occurrence. It is most common for the frequency to simply wander off in a unidirectional fashion related to how the oscillator and the oscillator structure ages. The Allan variance captures this behavior fairly well, and provides consistent results when characterizing the performance of large groups of similar oscillators i.e. it yield consistent results.
@Don, re “I really can’t imagine that the ATSB team ignored those fractures.”
For sure not all data have been shared. ATSB considered the data given to them only. It was not their responsibility to perform a detailed analysis and they made clear in the final report that they deferred the detailed structural analysis to malaysia. Also why is this not on the statutory interim reports? On that basis it is understood the analysis given was superficial ie based on visual examination only (no radiography etc), unless you have inside information not made public to indicate otherwise. Note the findings about the flap position was based on visual examination only.
Do you also disagree that such analysis will inform the end of flight scenario?
It should be known by now whether the pieces detached mid air or not and whether a fuel exhaustion scenario could lead to such mechanism without pilot input. I dont agree that this information cannot be deduced and it cannot be useful to understand what has happened.
@Don Thompson @HB
From the ATSB 2-11 debris report:
“Numerous other discrete areas of flap damage were analysed. Some of the damage was
consistent with the flaps in the retracted position, while other areas did not provide any useful
indication of the likely flap position. It was therefore concluded that:
The right outboard flap was most likely in the retracted position at the time it separated from
the wing.
The right flaperon was probably at, or close to, the neutral position at the time it separated from
the wing.”
Again, this makes clear the ATSB made no conlusive statement about the position of the flap(s).
@Dr B. To Victor, your, “Quite frankly, if we didn’t know MEFE, I don’t think the correct route would ever be found analytically. That allows a great simplification in reducing the number of speed and altitude combinations that have to be searched.”
You have the engine failures separated by 8 mins, the second, left, being 2 mins before the BFOs. However if the left engine generators were isolated when the right failed, MEFE would occur 6 mins after the BFOs.
There would be these effects:
• The speed from the 6th arc would be higher, both engines operating to within 2 mins of the 7th arc; just the left for those last 2 mins.
• 7th arc altitude will be higher.
• Fuel available would be less. The right engine would be operating for 8 more mins. The left engine would have over 8 mins (auto throttle operating vs not, in fact extending the 8) fuel “left over” to continue after the 8th arc in a right turn. That would spiral the aircraft into the sea with the left engine still lit, supposing it did not flame out during the minute it would be drawing from the left tank suction supply (prior to the APU coming on line). If not it would attempt relight(s).
This in not academic, being simulated by Boeing for the ATSB in 2016 when it seems that other scenarios did not realise the BFO descent criteria. It might well be this configuration met those.
If your model is sensitive to those 8 mins less fuel available you might need a plan B.
Also, in your post to Brian Anderson you said, “The CMT route hits the 7th Arc much closer to perpendicular than the straight routes, so you don’t need a significant turn for the speed to be reasonable”.
With bank over 2 mins supposedly prompting the final BFOs unmanned, a straight course, ie perpendicular to 7th arc, is not an option. The aircraft might well be heading back having overflown it. It would have turned right (IMO, at least). Given the small time from the 6th, that delay could make a noticeable difference to average speed, though the higher average speed to the 7th arc would tend to reduce that.
PS I suppose your fuel flow in your fig 4 is for the left engine?
! For “8th” arc read 7th please.
@Don Thompson @HB
From the 2-11 ATSB report:
“Numerous other discrete areas of flap damage were analysed. Some of the damage was consistent with the flaps in the retracted position, while other areas did not provide any usefull indication of the likely flap position.”
I read this as; some of the damage was consistent with the flaps in the retracted posistion but many (numerous) did not supported this view.
@DrB: Some short comments:
1. You are finding a sharp peak in the BTO because the effects of magnetic variation and crosswinds strongly distort the rotational symmetry for the CTH and CMT modes. As CTT and LNAV modes are not affected by magnetic variation and weakly affected by crosswind, the peaks are broad and difficult to use as a discriminator as over a range of paths, as the RMS BTO error is close to the standard deviation of the population. Tail/headwinds, temperature fields, and satellite inclination also distort the rotational symmetry, but less so.
2. You insist that the “Inmarsat BFO criteria” of peak error less than 7 Hz is correct. I am saying that I don’t know. What we know is that the DSTG presented data from 20 flights, not one, the DSTG study presented their results more than a year later after Inmarsat, and that study included the participation of Inmarsat. Significant drift of the fixed frequency bias was reported in the DSTG study, and a significant effort was made to model it. Your statement implies that the DSTG made a serious error in these calculations. That is certainly possible, but you should acknowledge that fact.
3. Even if there is no drift after power up of the SATCOM, we don’t know if the bias is exactly the same after a power down, cool down, and power up. You assume it is the same. I am not so certain. I doubt we have any data from an inflight power cycle to make that determination. I wish we did.
4. You make other assumptions, such as level flight and no pilot input after 19:41, wind variation, track variation. All could be reasonable, but they are still assumptions.
5. I see that you have abandoned the criterion to connect the path after 19:41 and before 18:28 with a simple path, as your “S-turn” is not simple. Whether there was a descent, hold, S-turn, or some other excursion, we can only speculate, but more northern endpoints require some kind of delay before the final turn south.
Again, you have convinced me of the viability of your solution, but less so on the uniqueness of the endpoint. As your proposed endpoint will be searched, having already been identified as viable, I don’t see much point in arguing over the correctness of the analysis. We’ll know soon enough.
The discussion with Ocean Infinity and the ATSB about whether to continue past the priority area of 25,000 km2 occurred months ago at a time when you were advocating an endpoint at a latitude of 34.8S, which fell within the 25,000 km2 priority area. Months ago, if some of us hadn’t encouraged searching beyond the 25,000 km2, it likely would not be occurring today. Your analysis provides further evidence that the decision to search beyond the priority area was correct. I hope it is found at your currently proposed endpoint, or even sooner.
@VictorI
RE: “Months ago, if some of us hadn’t encouraged searching beyond the 25,000 km2, it likely would not be occurring today.”
I think so too.
In fact OI/SC has started the search beyond the 25.000km2 allready as I saw in @TBill’s latest linked graphic.
I’ll stress they have to proceed at least till ~97E now the time and weather is right.
But we all have our priority assumptions I understand. And OI will have their own educated agenda too.
Still @DrB provides challenging data and views for the ones who (want to) understand them. I don’t quite but his effort is amazing to me.
@Ge Rijn: DrB’s report represents an astounding amount of effort, and should not be ignored.
@Ge Rijn
The conclusion about the retracted position is correct in my opinion. What this reads is that no other information can be extracted from the supposedly visual inspection. I agree with this part but i disagree that no useful information can be obtained from a detailed analysis. What i believed happened is that ATSB was only responsible for the search and determning the search location. They were not accountable to inform what had actually happened. Since the confidence was so high abut the search location, they must have thought that debris information would not add anything. The precise point i disagree with. Every piece of evidence must be analysed in details independently. Ie in this case, independently from the isat analysis. It is only when all information is available that scenarios naturally develop. @Don inferred this was done, maybe from inside information, but i very much doubt it was done to the level of details to inform what has happened as per icao annex 13. This intormation may come too late if it comes at all.
Then just another far fetched association like @Richard Godfrey suggested some time ago “with bearth date (hold your breath..).
Shah had his paragliding accident in 2007. He broke his second lumbal vertabra, FI:
“The factual information states, that there were no significant medical issues with both flight crew even though the captain, following a paragliding accident in 2007 causing the fracture of the 2nd lumbar vertebra, took “analgesics on an irregular basis” for his pain.”
Waypoint 32.27’S/97E is exactly in the middle of a +5100m deep hole and easy found in GE. If envisioned (like I did before) Broken Ridge as resembling a man lying on his belly this hole exactly represents a second lumbar vertebra in the sequence of the 5 lumbar vertebra.
Last digets of 32.27 could indicate; ‘my broken back in 2007’. 97E would be a very easy longitude to remind also.
I know this is in the realms of utter speculation. But who knows what was in the mans mind.
@Ge Rijn: All theories involving Broken Ridge will be tested, as OI gives all indications that this region will be scanned.
@IrtheTurner
Well said: “Yesterday it was 4 years ago that this atrocity took place. (Apologies Victor for posting an unscientific comment here) thoughts were with NOK, their life sentence in this nightmare. Like all of us, I had so hoped the fuselage would be found before yesterday.”
Thank you for the solemn thought. I somewhat regret making technical remarks yesterday. However, I am still hoping the fuselage was found before yesterday. If not, we have DrB’s new pin to give us hope on the Plateau, not
to mentions Richard’s “old” pin. I note DrB’s new pin is reasonably close to the birthday waypoint, which is my fallback if the BR hypothesis does not work out.
OI/SC is between a “rock and a hard pace” on sat communications for us tracking them. We now know OI/SC will turn off their satellite coms when they find something, right? Or do we? They have me guessing now.
@VictorI
I truly hope so. It’s just ~15/~20Nm outside the +/-25Nm zone.
With their pace they could search this aera probably within a week.
If nothing found it would be only a ~week delay with less risk of changing circumstances and time/fund-constraints later when nothing gets found more north and they have to go back to search wider.
@Ge Rijn: If you are suggesting that they search wider than +/-25 NM, that ain’t gonna happen at this point in the search.
@VictorI
I guess you know?
Anyway it cann’t stop me from stressing they go wider at this point in time and space now they are there.
OI has allready shown they won’t limit themselves to fixed +/-25Nm bounderies. Their flexibility is key to their succes I firmly believe.
@DennisW @Victor @DrB
Well BFO is a very subtle curve, but 19:41 BFO definitely seems off, should be 104 for straight level flight at cruise speed, versus 111 measured. Subtract 104 from 111 and you get 7-hz that Inmarsat said. Perhaps 7-hz is an approximate normalization factor for the 19:41 point, which is offset either due to data scatter or maneuvers were still in progress. Let’s face it, everyone struggles to meet that 19:41 BFO, except maybe me and Jean-Luc because we are both taking maneuvers to make it fit, which is probably correct.
Straight level flight from 1941 is probably a wrong assumption, although perhaps from just after 19:41 say 20:00, it was constant (but only until Arc5 or maybe a little after that).
Which brings me to my to last point:
@DrB said “…none of the LNAV routes are the correct route”
I know we are shooting DrB’s work from the hip (I greatly appreciate the work) but when we make pronouncements “like LNAV not working” I presume the end of that sentence is the assumption “for a passive pilot flight” scenario with no pilot inputs, which I am not restricting myself to.
Having said that, I am also using 180S_CTH because I have impression it fits the data better than LNAV with the winds. It struck me that CMH might also work, but I did not get that far along to try to make CMH work.
@DrB
You have every right to claim that the MH370 End Point is 31.57°S. You are quite possibly right. You have obviously spent a huge amount of time analysing the data, which I aplaud.
You have no right to claim that this is the only possible correct solution. There is a small chance that others might be right. I have also simulated over 700 flight paths and run a large number of floating debris track simulations. I came to a slightly different recommendation for Ocean Infinity at 30°S +/- 1 degree.
In any case, both solutions will be checked soon and it will probably turn out that MH370 is found somewhere else.
Anyone can be a “cherry picker”:
– BFO excludes certain routes on one hand, then the BFO only tells us which hemisphere on the other hand.
– Fuel gets me to 31.57°S, then fuel does not get me to 30°S.
– Speed of Mach 0.848 was a colossal error and not possible, then Mach 0.848 doesn’t matter.
– Peaks must be found, therefore CTT or LNAV which do not show peaks cannot be right.
Let us focus on the fact that we agree on one thing “the aircraft is further north”.
Contrary to your criticism, I have published many of my flight paths over the last years. Not all of the meanwhile over 700 flight paths, but all the major versions. The fuel analysis I use is from Barry Martin and is also published at aqqa.org. There are 2 summary papers that I have written, one on flight analysis and the other on floating debris analysis. Both have been published on this web site.
I would encourage you to look at the wider picture and consider:
– The transoceanic debris tracks for the 25 floating debris finds confirmed or suspected from MH370.
– The RNZAF photos of floating debris and the short term tracks as previously analysed by Victor.
I also started my computer career with punch cards and I like your analogy with the holes lining up, that is why I suggest you look where all the evidence lines up.
There is an even better fit for a CMT flight path via waypoint ISBIX, that I previously published in August 2017. Unfortunately, the end point in this case is shown as 30.3°S.
https://www.dropbox.com/s/w36m7ayzi41efw9/MH370%20Flight%20Path%20Model%20V16.0%20ISBIX%20CMT%20Flight%20Data.png?dl=0
@TBill said: Let’s face it, everyone struggles to meet that 19:41 BFO…
I think @DrB’s work shows that there is at least one scenario with low BFO error at 19:41 that requires no manual manoeuver or change in BFO bias.
Interestingly, for great circle paths, both the peak and RMS BFO error remain low if the fixed bias is allowed to be different before and after the power up at 18:25.
@TBill
My own analytics shows that a 180 degree track at constant altitude at 420 knots ground speed produces the correct BFO at 19:41 (not very sensitive to latitude). Not sure where you are encountering an issue.
You won’t have any trouble with bfo error at 1941 for an early FMT and no loitering. The difficulty is that this path takes you to 38S and your BFO errors grow as you go…
@DennisW @Paul Smithson
Dennis- yes slow speed works but then where do you go from there? Gimme a path that works after that. Slow speed is no problem for me as I am ascending so I am slower and ascending.
Paul S- show me…for example Yap’s IG path spreadsheet takes a 300-ft/min ascent. Some paths take a spontaneous slow down just before 19:41.
@TBill
It really depends on your definition of “works”. My paths are less constrained by BFO than yours. I could synthesize many paths from a Northerly 19:41 (including a CI path). Basically the ISAT data, in my view, has always been less prescriptive than for most other people.
@Paul Smithson “You won’t have any trouble with bfo error at 1941 for an early FMT and no loitering. The difficulty is that this path takes you to 38S and your BFO errors grow as you go…”
Or in other words, minimum BTO errors and minimum BTO errors are mutually exclusive?
Can’t help thinking that’s telling us something fundamental about the nature of BFO errors, something we’re failing to pick up on, something that could come back to bite us some day.
@Paul Smithson “You won’t have any trouble with bfo error at 1941 for an early FMT and no loitering. The difficulty is that this path takes you to 38S and your BFO errors grow as you go…”
Idiot (me, not you Paul) I meant to say of course:
Or in other words, minimum BTO errors and minimum BFO errors are mutually exclusive?
Can’t help thinking that’s telling us something fundamental about the nature of BFO errors, something we’re failing to pick up on, something that could come back to bite us some day.
@David,
You said: “PS I suppose your fuel flow in your fig 4 is for the left engine?”
Almost. It is actually for a “new” left engine with 0% PDA. It’s a bit confusing, and I may fix that if I put out a revised version.
Your comments regarding the possibility of the SDU reboot being caused by the right engine fuel exhaustion are interesting. There is actually some support for this idea in the PDAs needed to match the 00:17 MEFE. For instance, for my described scenario with R engine FE at 00:09 and L at 00:17, the PDA needed is +2.7%. That’s higher than the known value of +1.5%, but still within my estimated error in my fuel model. If R engine FE was at 00:17, and L engine FE was at 00:25, the PDA needed is +1.1%. That’s closer to the nominal +1.5% than the +2.7% for L engine FE at 00:17. I don’t know if that difference is meaningful, but, at the very least, one can say that the R engine FE at 00:17 is no worse a fit to the fuel model than is the L engine FE at 00:17. This does not change my route fits through 00:11 in a material may.
The distance from the 6th to the 7th arcs is not consistent with the higher average speed of the late FE, unless a fairly large turn occurred in the last 2 minutes so the aircraft spends a lot of that time flying parallel to the arc, or even circles back. In this case I would say the aircraft would have to be well off the extended prior course (by as much as 15 NM). That is one of the reasons I suggested searching +/- 30 NM along the arc.
@DennisW,
You said: “Your 7Hz assertions basically ignore the data published by the DSTG. That is OK. You are free to do that, but you should say so, IMO.”
Actually, I did not ignore the DSTG results in my paper. I said in Section 4.2: “The CBFOs have a standard deviation (1σ) of about 4.3 Hz when 20 flights are analyzed as one ensemble with outliers excluded [17]. “ I referenced the DSTG figure of BFO errors.
@DrB
OK.
I am still a bit uncertain about how the 20 flight data was calculated with respect to the “ensemble” numbers. I have asked for the equivalent of figure 5.4 for the 20 flights without a response a couple of times now.
Blaine undeterred (thankfully, good grief it takes guts to be a good Samaritan these days…in addition to exemplary performance in the service of humanity)
http://english.astroawani.com/malaysia-news/blaine-gibson-undeterred-neither-pain-nor-poolonium-will-continue-mh370-search-169930
@Dr B. I am glad the right-engine-caused log-on can be accommodated and note your +/- 30 NM reasoning thanks. However I hope what your earlier estimation of that can embrace this new possible ingredient without a squeeze.
A side effect of that electrical configuration is that it makes a tight spiral more likely and with that, proximity to the 7th arc, though that does not alter the wander of the 7th arc point up or down it.
@TBill: Blaine has provided a tremendous service in helping to solve this mystery. He continues to be unfairly attacked by many, mostly because he has refused to stop in his quest to collect more evidence. I have interacted with Blaine enough to be sure his intentions are sincere, and it pains me so see the mean and untrue things said about him. I’m glad that kind of behavior is never seen here.
@Victor. About the 4th anniversary, on JW BuyerNinety posted:
https://www.youtube.com/watch?v=leyjsTTWL1Y
Curiosity. This shows 8 AUVs in the ‘hangar’ at one point.
However I have never seen a photograph or video of SC with 8 USVs onboard, including in Durban.
With what appears to be a 6 deck-stowage limit, the last two would need to be accommodated on the davits and are in the water in all the photos and videos. The alternative would be that there are just the six and 2 AUVs are controlled directly from SC?
@David: Huh? What 6-AUV deck stowage are you talking about? At 2:05 you see all eight AUVs in their bay. And in the following time-lapse, you see rotation and launch from the bay.
Am I missing your point?
@Ikr. The 6 deck stowage is of USVs, the surface vessels. Launched by davit.
The 8 AUVs are launched through the 2 hangar stern doors as you imply.
Less than 8 have been operated recently. There could have been damage in moderate seas. I would not much like to be on an AUV steadying line in the hangar when repositioning AUVs with the ship rolling and little space for personal mobility. I supposed there could be a defect in, or damage to a couple, or to USVs, or to the apparatus for ship control of the 2 closest AUVs if there be such a thing, or data handling limits or simply no requirement.
However I did not mean this to be other than a question out of general curiosity.
@Ikr. Another casual observation even if beyond your query. The more AUVs and SUVs to recover the more time to do that. No problem if one can assume some long lasting holding pattern while waiting, though I have my doubts about the SUVs in extreme seas and wind. If a long spell of weather is forecast where no recovery is possible, there is a risk their energy reserves will deplete, meaning they need recovery before it arrives. It may be that SC’s AUVs and SUVs cannot be recovered in beyond quite moderate conditions.
I speculate that there if a spell exceeding those is in the offing, with recovery time which will be available being uncertain, best not have too many to recover.
USVs vice SUVs. Dyslexic.
@David: The USVs are not used for this mission. After launching an AUV, the host vessel is used to periodically re-calibrate the navigation system of the AUV. You can learn more about the launch, calibration, and recovery sequence of the AUVs at Richard Cole’s website: http://recole.org.uk/
@Victor. Thanks for that. I had been following @Richard from time to time but had not noticed the absence of evidence. Apparently evidence of absence.
But yet evident onboard.
@ALL
My thoughts and prayers go out to the NOK. It’s unimaginable that four years out the whereabouts of their loved ones are still a mystery, and they have not been afforded the decency and relative comfort of closure. Personally speaking, my anger and pain would be unbounded were I to experience what they have been through and endured.
IMO pinning any hopes on ZS’s birthday is pointless. Regardless of whether he believed he would be ‘tracked’ into the SIO or not, a birthdate is in every sense a commoners trick. IMO he would have chided and scolded himself for giving it even a moments consideration. It was beneath him and not sufficiently sophisticated, symbolic or cryptic to appease his ego. His narcissism (i.e. birthdate) being subordinate to the ego and diabolical genius hard at work.
I realize no one is taking the birthday possibility too seriously, which is for the best.
IMO the BR target would have appealed to Z for a few reasons (trenches, ‘masking’ and difficult terrain, metaphorically), already mentioned above.
Lastly, though Rob is companionless down at 38-40S, I must say that I find the drift analysis less than compelling. And I am plagued by the preponderance of snow and ice videos that ZS had on display. And of course his ‘Icemaker Complete’ DIY video with the Whirlpool appliance.
But you science men and women that get this stuff say that far south is nigh impossible. Let’s hope so!
@DrB @all
Thoughts on DrB’s pin at 31.54S
(1) +-30 nm gets not quite to the cliff edge of Broken Ridge. That is interesting. If the aircraft is found in BR, I suspect we’ll debate for years if that was intentional, or not.
(2) DrB’s pin is just 38nm from Victor’s birthday waypoint suggestion, which is just outside Arc6. If they do not find the aircraft by the time they get there, I am going to think about if a case can be made. I am thinking the case would have to be to assume the Arcs are displaced maybe 20nm too far east, which is a controversial stretch case I know.
Way too many variables to determine terminus.
Think this; what is maximum glide distance from highest altitude?
What is the most confident last known position.
Maximum glide distance Radius.
There is more radar data out there.
All the simulator “what if” is a distraction.
William Shea
@TBill
I am thinking the case would have to be to assume the Arcs are displaced maybe 20nm too far east, which is a controversial stretch case I know.
Just curious. Why do you think that?
@Donald “Lastly, though Rob is companionless down at 38-40S, I must say that I find the drift analysis less than compelling. And I am plagued by the preponderance of snow and ice videos that ZS had on display. And of course his ‘Icemaker Complete’ DIY video with the Whirlpool appliance.”
Donald, I’m actually very happy with the place I’m in at the moment believe it or not. I’m well used to being companionless (a certain neurological condition has seen to that). I am so confident in fact about my pin position, that I’ve ordered extra popcorn, and I don’t even like the stuff!)
Have you noticed how heated things have become as OI proceed to demolish one cherished pin position after another? Some uncharacteristically rash statements have been flying about, from people who should really know better. The scientific method is playing second fiddle to egotistical infighting and reputation protectionism.
Dont worry, in spite of all this, I have made sure to give OI a fighting chance of going home with the money, if you catch my drift. It’s up to them now 😉
Follow the conversation here with rapt attention. But, can’t seem to find an answer to a very mundane, but nonetheless vexing, question. So, I ask in all sincerity as it figures into AUV recovery, especially in weather. How, precisely, does the cable that pulls the AUV aboard SC get attached to the AUV? Does the AUV release a line to be grappled from SC? Is there some floating docking mechanism? Must a line from SC be fired out to the AUV, which then somehow captures it? Thanks.
@Rob,
Don’t worry about the knockers mate. I still read your stuff.
SC has recovered the 6 AUVs deployed at the northern end of the Tertiary Search Area and is heading back to the start of the Secondary Search Area on the S.E. side of the 7th Arc at around 34.8°S, marked by the yellow diamond in the attached graphic.
https://www.dropbox.com/s/p8unj0n012zeth7/SC%20Track%2010032018%20Wide.pdf?dl=0
@Richard Godfrey, @TBill: Looks like we’ll soon get our “third chance” at searching the BEDAX-South Pole path.
@Rob
Yes, don’t worry about it. The positive of your unshakeable focus and determination on ~38S to me is you are continually challenging/provoking the assumptions of the ‘neuro-typicals’.
Although I believe your scenario is out of the question for a long time now, it’s usefull to keep people sharp in a way.
Nothing has been quite settled yet.
It’s a proces now of reduction. CSIRO hot-spots have failed to comply.
@DrB’s former hot-spot also did. @Victor’s SPOLE-track also for it would require impact close to the 7th arc. The plane is not found at those areas.
What’s left now (regarding this topic) are the Godfrey/YWKS-spot, the Ianello/45s/104E-spot and the Ianello/Godfrey NZPG-spot.
All this remaining spots are based on unpiloted/uncontrolled impact close to the 7th arc imo. For there is nothing specific in those areas to attempt a controlled impact. It could well turn out to be the case but imo most evidence is still peaking against such an end-of-flight.
In this regard I might be on the spectrum too.
I firmly believe if the plane is not found along the trenches of Broken Ridge between ~95E and ~97E (or even beyond) it won’t be found further north too. I hope it will but we’ll see.
Time for some pop-corn..
@Victor
…and the third time is “charm”
ouch though it would tend to say not hits so far
@DennisW
If I were to propose a birthday waypoint case, it would probably be, that’s where the plane is. So that’s back on Arc6.
@Ge Rijn said What’s left now (regarding this topic) are the Godfrey/YWKS-spot, the Ianello/45s/104E-spot and the Ianello/Godfrey NZPG-spot.
I have said numerous times that the identified spots are barely “warm”. I have for a long time advocated a continuous, progressive search, without regard to particular assumptions leading to the exclusive searching of preferred spots. I include them in figures for reference only.
@TBill
Like @Donald suggested a birth day-waypoint would be too simple for someone with a big ego and a sophisticated, educated mind planning a disappearance.
If he used something like this as motivation for an end-point it rather would be something much more complicated and personal reflecting his state of mind and thinking at the time.
@Ge Rijn
Well if the descent was intentional, and if the SDU relogon was a message, then the question becomes was there a target? or was it just ending the flight anywhere in SIO.
@VictorI @TBill
My point is; if the plane not gets found along Broken Ridge then there was not a specific target also and the plane ended high speed close to the 7th arc piloted or not.
A message by deliberately cycling the IDG power again at ~0:19 would only make sence if the PIC was aware of the possible capability of the SATCOM connection to reveil his global position and used as a distraction to let everyone believe afterwards the plane ended in a high speed impact (close to the 7th) arc while after the steep descent he glided to a position which would be very hard to locate. This only makes sence along the Broken Ridge trences imo and not along any point north of ~32.2.
Those areas and points north along the 7th arc can only be based on an unpiloted- or maybe piloted fuel exhaust scenario close to the 7th arc with no intended end-destination at all, for there is nothing specific there and in those cases no specific end-waypoint would have been set at a specific location.
In those cases only an end-waypoint for a track towards NZPG, SPOLE, YWKS or 45S/104E (or anything in between) makes sence. Then waiting for fuel exhaust and diving the plane near the 7th arc piloted or not.
If the plane is north of ~32.2S it will be found well within the +/-25Nm zone. If not OI has to return to Broken Ridge and search well outside this +/-25Nm zone.
@Richard Godfrey,
You said: “You have no right to claim that this is the only possible correct solution. There is a small chance that others might be right.”
It’s not a question of rights. I said that, using the method and assumptions I made, there is only one answer. You can make whatever assumptions you want and use a different method of analysis, and you may get different results. So far, I have not seen any other method that produces just one answer. That doesn’t prove my answer is correct, just that the other methods are inadequate to discriminate the correct route, or that insufficient information is contained in the satellite data (actually just the 5 BTOs) to do so.
You said: “There is an even better fit for a CMT flight path via waypoint ISBIX, that I previously published in August 2017.”
The data sheet you referenced shows that the magnetic track of that route varies from 176 to 183 degrees! How can that be a “better fit”? Do you always allow the magnetic track to vary by 7 degrees in your fits?
You said: “Anyone can be a “cherry picker”:
– BFO excludes certain routes on one hand, then the BFO only tells us which hemisphere on the other hand.
– Fuel gets me to 31.57°S, then fuel does not get me to 30°S.
– Speed of Mach 0.848 was a colossal error and not possible, then Mach 0.848 doesn’t matter.
– Peaks must be found, therefore CTT or LNAV which do not show peaks cannot be right.”
I will address those points one by one.
1. Yes, the BFOs are useful in excluding routes with very large errors, so large that they are outside the regions of interest. Within the ROIs the BFO is not a very useful discriminator. In fact, the BTO is much more discriminating, both within and without the ROI. The BTOs will eliminate all those routes that could have been eliminated because of large BFO errors. The only significant contribution of the BFOs is to pick the correct hemisphere.
2. Concerning fuel, some routes to 30S are possible and some are not. It all depends on the speed/altitude and the 19:41 starting point. If that point is far to the NW, the fuel to reach 30S may be inadequate. The fuel adequacy is not as simple as the latitude of the 7th Arc intersection. It’s how you got there that counts.
3. Your route at M0.848 is unflyable by a B777 because the speed is too high. In addition, the fuel consumption at the fastest possible speed, M0.840, is too high to match the 00:17 MEFE. So that route can be excluded on two grounds – it is too fast and the fuel consumption is too high. How is that “cherry-picking”?
4. You obviously did not understand my discussion of peaks in the FOM curves. I said there can only be a peak for a single route – the correct one. There does not have to be, nor should there be, a peak for every lateral navigation method. Yes, if a lateral navigation method curve has no peaks, it cannot be the correct method. That includes CTT and LNAV and CMH. If a lateral navigation curve has a peak, it must be the correct route (but not necessarily the correct navigation method). In the case of MH370, there was more than one peak because the winds produced a CTH route that was nearly the same as the correct CMT route.
For French speakers, here is an excellent round table discussion on MH370 theories, including that proposed by Jean-Luc Marchand and his colleagues. Although there is no new information presented, it is interesting the hear the views (and passion) of the guests. All guests believed the disappearance was a voluntary act.
http://video.lefigaro.fr/figaro/video/mh370-connaitra-t-on-la-verite/5748205443001/
@DrB said: That doesn’t prove my answer is correct, just that the other methods are inadequate to discriminate the correct route, or that insufficient information is contained in the satellite data (actually just the 5 BTOs) to do so.
Agreed.
@Ge Rijn: You continue state your opinions as facts. You are exhausting many of us.
C’est une plus belle langue..
The mystery is alive still in France too.
But no word on the flaperon details which is in their possesion still.
If we only had those details a lot of discussing could be irrelevant instantly.
@DrB
Many thanks for your response.
@VictorI
I constantly try to ground my views with links and considered arguments over the years. I’m not interested in pleasing anyones thoughts or egos on the subject.
My goal is the same as yours and anyone else here to try contributing in solving this drama and mystery.
You and many here (like me) are also very persistent in their assumptions without knowing all the facts.
Victor Iannello says: “For French speakers, here is an excellent round table discussion on MH370 theories, including that proposed by Jean-Luc Marchand and his colleagues. Although there is no new information presented, it is interesting the hear the views (and passion) of the guests. All guests believed the disappearance was a voluntary act.”
The most interesting advice is, IMHO, the one of Jean Paul Troadec (ex BEA director). You can download this discussion there until 18/03 (~273Mo):
http://www.transfernow.net/124iy1q47dh5
Regards,
PL
@DrB said: That doesn’t prove my answer is correct, just that the other methods are inadequate to discriminate the correct route, or that insufficient information is contained in the satellite data (actually just the 5 BTOs) to do so.
@Victor said: Agreed.
I agree too that there is insufficient information contained in the satellite data. That is why I suggested the information from the the short term drift analysis matching the RNZAF aerial photos, which Victor has analysed and published and the information from the transoceanic drift of the 25 MH370 confirmed or likely floating debris items, which I have analysed and published. The 3 sources of information taken together provide a much clearer picture in my view.
It appears that 3 AUVs have been launched so far on the S.E. side of the Secondary Search Area in a rolling pattern transverse to the 7th Arc.
https://www.dropbox.com/s/2hx7betkevy3427/SC%20Track%2011032018.pdf?dl=0
@Richard Godfrey,
You said: “I agree too that there is insufficient information contained in the satellite data.”
That’s not what I said, nor is it what Victor agreed to. It is one option I stated. The other option is that all previous analysis methods are unable to discriminate the correct route. In my opinion, that is the most likely case. If Ocean Infinity continues to make good progress, and if they don’t get stalled trying to map the Diamantina Escarpment first, we shall know the answer relatively soon.
@Richard Godfrey @VictorI
All well and plausible but you still seem to ignore the ‘flaperon-like’ piece and debris field that was spotted at 28-3-2014 around 32.4S/97.8E.
Imo this debris field has at least the same significance as yours.
It should not be ignored by OI.
Speakers at the Figaro roundtable, including Feldzer and Troadec, are clear that this was not a fire or other technical issue, but that the aircraft was hijacked. However, they all believe that pilot suicide is unlikely. The reasons for why the aircraft was hijacked are unclear but there are many possibilities. This could have an impact on reconstructing the flight trajectory. Full automated flight is just a simple possibility, a full piloted flight another. The recommendation is to search in the current area. Should this be unsuccessful, the further recommendation would be to continue searching until Christmas Island.
@DrB
Many thanks for your response!
I was agreeing with the second part of your statement only, not the first part of your statement.
@DrB
Be my guest as far as checking your’s and Richard hot spots before BR, so they can do a thorough job at BR when the go to do it, and we would sort of know then it’s there in BR. Although it appears they may have checked off a lot of that Diamantina Escarpment box this week. I guess we do not know yet exactly what they did.
I hope you are correct about your hot spot. But I would say the biggest conflict is that many observers feel it may have been an intentional course, with some path changes. Paradoxically, most of the modelers only want to give us passive flights, for reasons I am yet trying to understand. At least we have Jean-Luc going to CI, I give him credit for thinking through what a pilot (hijacker in his case) might do along the way.
@TBill said: Paradoxically, most of the modelers only want to give us passive flights, for reasons I am yet trying to understand.
Once you allow for pilot inputs along the way, the range of possible paths gets very large, unless you further constrain the paths with other evidence. For instance, if we have reason to believe there was intent to hijack the plane to Christmas Island, then we can generate paths consistent with that (as CAPTIO-net have done). But without those pre-dispositions, how would you limit the search area to something manageable? Already, I believe the range of automated, great circle path is quite large.
@DrB
Mapping the Diamantina Escarpment between ~32.2/33S and ~96/~97E would cost OI probably not more than a week.
Imo far better to do it now when the weather is still right and time plenty than to have to return later when the opposite might be the case.
@Victor
Based on much study, I like the following assumptions for active pilot model cases:
(1) No maneuvers after 19:41 until Arc5
(2) Could be maneuvers before 19:41 that are still resolving at 19:41
(3) Could be maneuvers after Arc5
Note I reasonably happy the current search zone covers the above cases quite well, although we may need to go up 20S to capture all of these cases.
@TBill. If you like the very consistent shape of the BTO curve and similar on the BFO then it tends to suggest to manoeuvre all the way back to somewhere very near 1828. That is the chief reason that I am inclined towards a very early FMT.
The second is that a turn in that vicinity takes you very close to Banda Aceh. The third is that the initial approach vector of 190 for Banda Aceh (if CTH) just-about-works for both BTO and BFO (though I have to admit alternate path models offer superior fit for both). The fourth is that this fit is achieved with timing and speed consistent with an attempted approach to Banda Aceh (Econ descent).
What’s not to like about this model? (1) BTO fit is only just good enough (2) BFO fit not great – would have to have drifted (3) the terminus implied at 38S has already been searched if we believe steep descent at final BFO (4) why fly past Penang only to attempt approach to Banda Aceh.
Having said this, I still prefer this option to further up north up the 7th arc.
@TBill said: Based on much study, I like the following assumptions for active pilot model cases:
And others like other assumptions regarding pilot input, based on various criteria that are consistent with various scenarios.
OI plans to progressively search a rather large range of latitudes along the arc. That seems like our best shot right now.
@Paul
What’s not to like about this model?
Early FMT models are not compatible with drift analytics.
@Victor
It is possible Shah ended up ditching up near Christmas Island.
MH370 could well be found near the 7th arc to the East of Christmas Island.
@Victor @DrB
Modeling only out to Arc5 is rewarding I feel because there are probably near-exact solutions, which I tentatively think is 180S CTH from ISBIX (but perhaps my accuracy with FS9 is not good enough to be sure).
Then all you have to do is some trial and error to see what the pilot needed to do to meet Arc6.
I hope DrB approach works, but if not it would probably be the same reason the Bayesian approach did not work: namely trying to find a passive flight solution to an active pilot flight. And also the big issue is public criticism (eg; from the denialists) when the (next) exact 100% sure solution fails. This could undermine the search, assuming we may still need to search more.
>>Having said that, DrB is providing, in so many words, the nextgen MH370 path search methodology (better than Bayesian). I would ask what the solutions would be if we only ran it out to Arc5? How close are those solutions to the passive flight solutions out to Arc7?
@TimR: I’ve asked you many times to disclose whatever information you claim you have that would suggest a terminus near Christmas Island is likely. Now that we have a private organization running the search, we may have more flexibility to consider alternative locations. What do you have to share?
@Victor
As I have mentioned on a number of occasions I am not in a position to disclose many of the details due to security concerns.
What I can say is that the aircraft was under the control of Captain Zaharie Ahmad Shah who wanted to ensure as much time as possible for negotiations to be conducted on the ground in KL. He therefore intended to “loiter” as he proceeded around below Sumatra before turning up to Christmas Island intending to land, release the passengers and place himself at the mercy of the authorities.
Something went wrong towards the end of the flight.
@TimR
Do you know everything was going to plan until problem at CI, or do you only know that the plan did not happen as planned? Maybe the pilot had another alternate plan in mind.
SC appears to be returning to the start point of the Search at 36.3°S, which is also the start point of this phase following the return from Fremantle.
Before that OI launched 5 AUVs in the Secondary Search Area, which will be ready for pick up tomorrow.
There is possibly some unfinished business at 36.3°S.
https://www.dropbox.com/s/16lhijnyno2bvy5/SC%20Track%2012032018.pdf?dl=0
@TimR: In the past, you have claimed you have no knowledge of what occurred during the flight. You know only what was planned to occur. It seems it is possible that there was a loiter near Sumatra followed by a turn to the south for some unknown reason. Plans changed.
@TimR “As I have mentioned on a number of occasions I am not in a position to disclose many of the details due to security concerns.”
Tim, you are merely taking advantage of what is now a patentlyof confused and chaotic situation iro the search effort, and using it to pedal divisive crap. Come to think of it, that’s sadly typical of what’s been happening ever since the night the plane disappeared, and has in no small measure been one of the reasons why they are still looking for it now.
In stark contrast to that, some of us have actually been using something would wouldn’t know much about (ie brains) to try and solve the mystery.
@TimR
For some time now you come up with claims you have proof of a planned action by Shah towards CI. But you will not disclose this proof for security reasons.
Concerning your own security? Or others?
I expect you can/will answer at least this question.
@Rob
Look at the bright side. Tim has been articulating his position for a long time and is not claiming:
1> The ISAT data was spoofed
2> The debris was planted
3> The drift analytics are flawed
4> The plane was flown North into Kazakstan
5> That people who disagree with him should have their inputs filtered
There are more weird things out there to be annoyed about.
@TimR @Rob
TimR is now making his explanation consistent with Jean-Luc’s MH370-Captio.net scenario, which is fine. Previously I believe TimR said Jakarta was the target landing airport, and he had no idea how close it ever got.
But its frustrating. On the whole the scenario makes some possible sense. Not too far above in the prior thread, someone said HHH told 4 Corners interviewers that some were told to “keep an eye” on MH370 before take-off. Of course they did not seem to follow thru. Also MAS CEO Dunleavy got to KLIA promptly (by luck he had a 6AM flight scheduled that day). He said they always assumed MH370 was still in the air (which they should have realized from the sat calls).
So it all seems generally consistent with the idea that officials had some expectation of a diversion. Presumably if so they were expecting a safe landing which never happened. We are left to ponder if Z never intended to follow thru with the plan, or something went wrong, which gets us nowhere fast.
@TBill
I have been in communication with Tim R for a very long time – ever since publishing my own CI scenario. There has been a laundry list of landing places, not sure if Tim put this in a blog anywhere, which would be consistent with an ongoing negotiation. The Jean-Luc scenario has not impacted Tim’s story at all.
PS- DrB’s new flight path now starts the same as I am thinking, with MH370 turning around 94th parallel. DrB had no explanaton, I have suggested heading south would have allowed MH370 to go left at BULVA on 94E which is on vector M300 direct path back to KLIA. I got some criticism on that idea, but that’s was my explanation. But MH370 just kept going South. The only difference between DrB and me, is I am going 180S CTH.
@Rob
RE:”..what is now a patentlyof confused and chaotic situation iro the search effort..”
Actualy I don’t think it’s a chaoctic situation yet by far.
The CSIRO/Griffin hot-spots have been eliminated. But I think those were based on the best regional drift-analysis but quite unsure satelite images.
Their regional drift-analysis around ~35.5S the drift was to the west in a ~small band of current.
But this same band of current then makes a sharp turn to due east around ~32S bending north-east around ~99E and then more north, west again.
I mean half of the CSIRO/Griffin analitics (the drift-analysis) could still well be helpfull.
They could well fit with the ‘flaperon-like’-piece and debris field (of 13 other spotted pieces) associated with it which was spotted around 32.4S/97.8E. at 28-3-2014.
Then all the ‘warm-spots’ mentioned beyond ~32S to the north are not searched yet.
So, many opportunities are still left.
to add.. also @VictorI’s proposal based on the debris photographed around ~29S west of the 7th arc fits rather well with this CSIRO/Griffin current.
I’m sure all known (and accepted) facts considered we(they) are tightening in on the crash area.
Imo the only big uncertainty left now is the search width.
But this will show during the uncoming search if the plane not gets found within +/-25Nm.
Hi,
A new French radio round table discussion on MH370 (audio only) on RTL this evening:
http://www.rtl.fr/actu/debats-societe/le-mystere-du-vol-mh-370-7792591323 (about 40 minutes).
The show title is “L’heure du crime” (The Time of Crime”), so don’t expect a theory about a mechanical failure!
Participants: Ghyslain Wattrelos, Gilles Diharce and Pierre Julien.
Same stories from GW and Diharce who are selling their books. Interesting passages from Pierre Julien (RTL journalist) from 26:10 to 31:00, specially from 28:30.
Downloadable from the page above and I uploaded it there: http://www.transfernow.net/00fpb3f9yw8d
Regards,
PL
@TBill,
Your post of 9.49 March 9th: Everyone is doing an exceptional job in trying to determine the location M9-MRO. That is, IMHO, a true homage to the people who lost their lives. People have invested so much of their own time and dime, to trying to pinpoint the location of this aircraft and perhaps give NOK some form of closure (if that is at all possible). So, thank you!
@Ge Rijn
“I mean half of the CSIRO/Griffin analitics (the drift-analysis) could still well be helpfull.”
Out of sheer humanity, I feel the need to manage your expectations in some way: so far, CSIROs attempts at drift modelling have only served to distract and misleading. Your optimism is touching, but what’s needed is realism and pragmatism, if this plane is going to be found”. CSIRO were given their chance to shine, and they failed miserably – time to go back to the basics.
“They could well fit with the ‘flaperon-like’-piece and debris field (of 13 other spotted pieces) associated with it which was spotted around 32.4S/97.8E. at 28-3-2014. Then all the ‘warm-spots’ mentioned beyond ~32S to the north are not searched yet.”
That flaperon-like piece you mention is far too cleanly/sharply defined as a rectangle to have come from the plane. Compare it’s outline to the “Pemba” flap section, and you’ll see what I mean. The ocean is full of floating debris; it’s one of the biggest environmental problems we face today.
“So, many opportunities are still left.”
This is complete nonsense. You need to go back to first principles, and start afresh.
SC is on its way back to the S.E. side of the Secondary Search Area having spent 13.5 hours at the Calibration Point 36.3092°S 91.7987°E.
The first of the deployed AUVs in the Secondary Search Area are awaiting pick up.
https://www.dropbox.com/s/4tijpbc8vqbxyde/SC%20Track%2013032018.pdf?dl=0
@Rob
The ‘failure’ to find the plane at the CSIRO hot spots is not due to their drift-analysis imo but in connecting those analysis with the satelite images. The negative result only indicates further the ‘debris’ shown in those images were not MH370-debris.
Their drift-analysis though are still relevant and usefull.
The debris field detected around 32.4S/97.8E including the ‘flaperon-like’piece is as relevant as the debris-pictures west of the 7th arc used by @VictorI imo.
All images are without enough detail though to make conclusive identification possible. So the only way to find out is to search those areas. I see this as a pragmatic approuch.
The search is not over by far. So no need to go back to first principles and start afresh yet at all.
Yes, I’m still optimistic. Many good opportunities are lying still ahead to be searched.
Only when this search is over without result I’m prepared to go back to first principles.
Australia asked to take the lead in examining debris and human remains if the wreckage of missing airliner is located.
Note the headline includes the recovery also but that is unsupported by the text.
Final report supposedly within days of wreckage being found. Mmmm.
http://www.abc.net.au/news/2018-03-13/mh370-australia-asked-to-lead-recovery-effort-if-wreckage-found/9544284
Sorry, wreckage of course not being found
Maybe the arrangement was/will be that the Malaysians would fly a -200ER down if they have one unsold, plus engineers, to help with parts identification, location and reconstruction.
@David
For one thing the whole article points to the possibility that the plane did not disintegrated in a high speed impact and the investigators seriously consider the option the fuselage stayed more or less intact and the plane was actively piloted till the end.
I mean, if they count with the possibility of finding a locked cockpit door and preserved recognisable human bodies/remains (after 4 years) a high speed nose-down impact is not an option.
@David
If MY had an unsold B777 there is so much investigation we could do with it. Any chance of that at all?
As many of you know, the weekly search report was just released. From the report:
Earlier during the underwater search operations, an ROV was damaged and a decision was made to ‘wet store’ the ROV to minimize disruptions to search operations. Seabed Constructor would recover this ROV enroute to Fremantle port.
Based on search patterns and comments made by individuals close to the search operation, some of us were almost certain that an AUV (not an ROV) was lost in a previous swing. Why Malaysia chose to first hide and now cloud this issue is unknown.
Just an incident I guess. Failures can be expected to happen with such new technology. But it’s a bit strange they (Malaysia) not explained the time-out right away this way. No problem, many would have understood this things can happen making no deal about it.
Maybe it’s only about Eastern cultures are not as direct as the Western when it comes to bringing ‘bad’ news.
In India when I lost track on my bike at times, I learned to ask at least five people to show me the way to go. I knew by then many just told me a (wrong) direction for they disliked showing they didn’t know either.
Their shame and willing to please was too great.
With three out of five answers I chose my direction and even then I got lost again some times.
Point to make; also cultural differences can cause huge misinterpretations.
@Ge Rijn
Your country has some differences as well. I can walk into a doctor’s office in the US with a paper cut, and get a prescription for opiates and antibiotics in a matter of minutes. In the Netherlands you could walk in with a gangrenous limb, and get sent home with a Tylenol and a recommendation of bed rest.
I’ve been investigating whether it’s possible that the fixed frequency bias (FFB) of the SATCOM changed due to MH370’s inflight power cycle. That would mean that the bias of 150 Hz is not correct after 18:25. To study this, I looked at the BFO data for IOR traffic when 9M-MRO was on the ground at KLIA. There were four log-ons, of which two were Class 1 using the LGA and two were Class 3 using the HGA. Since navigation data needs to be available to use the HGA, most likely the open-loop pre-compensation algorithm was used for Class 3 service. The Class 1 service shows a different value of FFB, likely because the closed-loop pre-compensation algorithm was used.
In short, the FFB before the power down at 9:02 was around 171 Hz. After the power up at 12:50, the FFB dropped to 167 Hz, and then after the SDU calibration routine at 16:00, it dropped to around 150 Hz. (The calibration shifts the FFB in increments of 16 Hz, according to the ATSB.) So, it appears the FFB shifted by 4 Hz due to the power down – dwell – power up cycle.
It would not be surprising to see a similar shift in FFB due to the inflight power down at 17:21 and power up at 18:25.
@VictorI
That is definitely a concern, IMO. However, the BFO value at 18:25 matches my best estimate of the aircraft speed and heading at that time almost perfectly. Of course, this has been kicked around extensively relatively to the reboot with ALSM and DrB.
~ 296 track
~ 510 knots ground speed
@Victor
What direction would that be?
Lets say bias may have shifted to 146 (from 150)?
I am thinking that means at 22:41 we should have seen BFO=200 vs. 204 actual
You calc BFO for SPOLE was 199.7, so you were closer ?
I do think there is some flavor of that in the BFO data say 2-4 hz. I was thinking 2.
@DennisW
Haha, yes I think you’re right. But in Holland, if you really need those opiates or antibiotics (or any kind of care) you’ll get them too for (almost) free. We have this social insurance system we all contribute some money to, for we believe health-care is not a privilege only for the ones with enough money. For it benefits the general health, productivety and social wellbeing and security of all here. We don’t have the kind of slumbs, extreme poverty, public shootings, constantly luring social unrest you have in America where the only standard you are measured on is what you possess and earn. The ‘American Dream’. Purely materialistic.
And this makes the ideal American identity which you obviously are a child from.
Different from European countries thinking though and even more from far Eastern countries thinking.
I think most Americans are therefore not able to understand the sensetivities of those other countries which wisdom and moral is based on many centuries with the most thoughtfull scientists and filosofers.
While American society is only ~200 years old and based on refuges and gold-diggers without any culture generally.
And those older cultures are not primarily based on possesions, greed and selfish affirmation like the current American culture.
Kind of lecture maybe. But I’m sure those differences in culture play a part in the way information is handled between different parties regarding MH370.
@Victor. Holy cow. That’s rather an important finding. Where would that place the latitude of minimum bfo error on the 7th ARC?
@Ge Rijn
OK I accept your premise. I would suggest one minor good trait America has is good aircraft crash investigations. We are expecting a full-out search for the truth no matter where that leads. So that’s why folks like Blaine Gibson are out there trying to help, because that is the kind of initiative and effort we are expecting.
@TBill
And you are pretty good at making airplanes as well, though we still have the edge with engines (or am I just wishful thinking with that?)
@Victor, holy 🐮
@TBill: We don’t know the direction of the FFB shift due to the inflight power cycling, if that shift indeed occurred. Great circle (LNAV) paths north of 35S would require a shift in FFB. For instance, for the GC path ending near 25S latitude, the RMS error (assuming a bias of 150 Hz) is 8.7 Hz. But that RMS error is composed of a mean error of -8.1 Hz and an RMS error about that mean of 3.2 Hz. Previously, I would have said the RMS error of 8.7 Hz is too large. But it’s also possible that the FFB shifted up to 158.1 Hz, and then the RMS error (using the new FFB) is only 3.2 Hz, which is acceptable. So, the acceptability of a path (based on BFO) would depend on the how much FFB shift is possible.
@paul smithson: We could answer that question if we better understood the shift in FFB that might occur from a power up – dwell -power up cycle. I referenced only one data point from one occurrence.
It would be helpful if others checked my work so we could compare results. To calculate the statistics, I used a BFO correction of 0 Hz for channels R4, R10, R11, T10, and T11, and a correction of -4 Hz for channels R8, T8, and T12. The EAFC+SAT term was computed using a digitized version of the pilot frequency error graph from the JON paper, with the uplink Doppler shift (from Burum) and downlink Doppler shift (to Perth) subtracted from that. The agreement with the EAFC+SAT tabular values supplied by the ATSB is excellent.
@TBill. “If MY had an unsold B777 there is so much investigation we could do with it. Any chance of that at all?”
They have had hulls with engines they have placed recovered items beside in flotsam identification. I would have hoped that Malaysia would have requested their use if still available in investigating outstanding issues, and have, including those referred by the ATSB.
What specifics within safe flight do you have in mind if usable; or on-ground testing? If decompression rates and temperatures, for example, I suspect Boeing would be up with that already.
There looks to be scope for more SDU testing in labs but also in parallel with other onboard systems, including IFE.
Victor,
“The EAFC+SAT term was computed using a digitized version of the pilot frequency error graph from the JON paper, with the uplink Doppler shift (from Burum) and downlink Doppler shift (to Perth) subtracted from that. The agreement with the EAFC+SAT tabular values supplied by the ATSB is excellent.”
As you may have already discovered, the Inmarsat procedure from JON (and the prior ATSB report) has you adding in the downlink Doppler to Perth (term labeled “Satellite movement” in Table 8 of JON), but you then subtract it out from the pilot frequency error graph. I have eliminated both terms from my calculations, and guess what – it still works! A lot of cruft builds up in one’s code over time (as clearly happened with Inmarsat), and at some point it is worth the time & effort to just clean it up.
I’ll try to make my own calculation of the frequency shift.
@sk999: Yes, the two operations cancel, and a clean program would eliminate those steps, unless there is the desire to observe the calculated value of EAFC+SAT.
@VcitorI
I regard doing anything useful with the BFO data relative to a terminus as a waste of time. The AES (or SDU as the preferred name here) oscillator is simply not accurate enough to draw any reliable conclusions. This is oscillator physics 101 here.
@all
So from an OI perspective the decision to undertake the search is a matter of probability – search cost per unit area… probability of finding the aircraft per unit area. I estimated their contract values, and came out very close (along with the notion of a stepped payback).
What about Malaysia? What should they be willing to pay? That is a notion in decision theory, which none of you have taken a course in (don’t ask me how I know). A common paradigm in decision theory is the notion of a clairvoyant, a know it all. The question then becomes how much would you be willing to pay a clairvoyant to know an answer. In the case of Malaysia I cannot think of any reason why they would pay a clairvoyant anything at all. Why did Malaysia agree to any terms with OI?
@DennisW
“The question then becomes how much would you be willing to pay a clairvoyant to know an answer. In the case of Malaysia I cannot think of any reason why they would pay a clairvoyant anything at all. Why did Malaysia agree to any terms with OI ?”
An interesting set of questions Dennis, and I know nothing of “decision theory”, but from a “political perspective”, I see it more like this.
The only logical reason I can think of, given the early history of the first few days, is that they (Malaysia) know what happened, and why, and thus, where it is, or most likely is, and thus, no need for a clairvoyant at all. Further, since that is not where OI proposed to look, and are looking, they know that OI will not find it.
Thus, the logic is, Malaysia really needed a way to move from “search suspended” to “search ended”. In a way, OI stepping up to the plate has been the “perfect answer” for Malaysia. I think Malaysia finally agreed to allow OI to search where they wanted to, simply to appease the NOK (and others), as a PR exercise, but most importantly, they realised that it would be the perfect vehicle for them to get to “the end of the search”.
Thus, they can just sit back, and let the clock run out on the “90 days” (however actually defined in the contract).
When OI has finished the existing defined area(s), and the clock has run out, with OI having come up empty, there will be no further likely area to search (in the public domain) for the NOK to hang their hopes, and their demands on.
So, Malaysia can then finally, and oficially, end the search, issue the final report, close the book, and move on. Not only that, they will not have to pay a red cent.
The wreckage is on South Georgia Island. Coordinates: 54°34’44”S, 36°11’42”W.
I took measurements of the object which is embedded in the snow, measurements are roughly 206’ long and 22’wide. MH370 is a Boeing 777-200ER with dimensions of 206.6’ long(from nose to tail not including stablizers) and the fuselage is 20’4” wide not including wings. The Boeing 777-200ER has a maximum range of 8888.7 miles. This distance is calculated with maximum passengers. The total distance from south Georgia Island to last primary contact over Andaman Sea (8300miles) then add the route to last known sighting over South China Sea 536 miles, then distance back to Kuala Lumpur Int’l 248 miles gives a total distance of 9088 miles. The jetliner could have made that distance if it had been completely fueled and added with the fact that flight MH370 had only 239 passengers plus crew. Seating capacity for the Boeing 777-200ER is 440 passengers plus flights crew. The longest recorded flight of a Boeing 777-200ER is 10,823 nautical miles. This flight broke the “great circle” (distance without refueling) and was accomplished on April 2, 1997 by “Super Ranger” a Malaysia Airliner 777-200ER flying eastbound from Beoing Field in Seatle to Kuala Lumpur.
The Boeing 777-200ER max seating capacity should have read 440 total(passengers plus flight crew).
I also forgot to mention that the wings were not on the object on South Georgia Island embedded in the snow.
@Victor, @DrB, @DennisW,
@Victor says “it appears the FFB shifted by 4 Hz due to the power down – dwell – power up cycle.”
@DrB says “I use a RMS value of 3.0 Hz for the expected dispersion of the CBFOs.”
@DennisW says “The AES (or SDU as the preferred name here) oscillator is simply not accurate enough to draw any reliable conclusions. This is oscillator physics 101 here.”
I have primarily been using the BTO to fit a flight path, with an acceptability criteria for the round trip RMS BTO Error of 29 µs.
Up until now, I have been using the BFO as a check, with an acceptability criteria for the RMS BFO Error of 7 Hz. It now appears that the RMS BFO Error should be relaxed (@Victor) or ignored (@DennisW).
@Victor showed, in a previous post, that the best BTO match was for a MH370 End Point around 33°S and the best BFO match was around 35°S. The BTO Error is relatively flat for LRC mode from 33°S to 27°S. The BFO Error for this range is < 10 Hz (please see link):
http://mh370.radiantphysics.com/wp-content/uploads/2017/11/GCBTOBFO2.png
In my view, in light of this new insight from @Victor, Ocean Infinity should consider searching up to 27°S, if resources and weather permit.
@Paul Glover said: The jetliner could have made that distance if it had been completely fueled…
We know that the aircraft was not fully fueled, and the range was therefore constrained to much less than the maximum.
SC has returned to the Secondary Search Area and picked up all the AUVs that were deployed.
It appears that Ocean Infinity have started the next cycle with the launch of AUV1.
https://www.dropbox.com/s/noqxbt1qqx577tm/SC%20Track%2014032018.pdf?dl=0
http://mpegmedia.abc.net.au/radio/local_sydney/audio/201803/ppm-2018-03-13-mh-recovery.mp3
@TBill
Ofcourse I know the American culture has accomplished many great things too with plenty of great scientists, great thinkers, artists, writers and so on.
Your culture in general is only based on different principles and thoughts than many other countries in the world. Which leads to resentment sometimes and misunderstanding.
I think the latter plays out at times. Not openly often but ‘under the waves’.
@Ge Rijn: Enough of your political and cultural commentary about the US, as it is a divisive distraction.
@Ventus
The only logical reason I can think of, given the early history of the first few days, is that they (Malaysia) know what happened, and why, and thus, where it is, or most likely is, and thus, no need for a clairvoyant at all. Further, since that is not where OI proposed to look, and are looking, they know that OI will not find it.
Interesting. I have had similar thoughts which actually motivated my rather weird post.
@Richard
In my view, in light of this new insight from @Victor, Ocean Infinity should consider searching up to 27°S, if resources and weather permit.
+1
@VictorI
I’m sorry if I ask something which is obvious for most here (I have missed quite some details lately):
– What is the mentioned “closed loop” aspect of the pre-compensation algorithm?
– Is it possible that the 4 Hz difference you observed originates from the 3 hours “dwell”? (oscillator drift)
– I was checking the logs again and it looks to me that everything after 15:59:56 was through HGA / class 3, so why would you expect the in-flight “jump” in FFB?
– A final point: if the FFB would drop by 4 Hz wouldn’t this shift the latitude of best BFO fit to the south by about 3 degrees?
@Ventus45
“The only logical reason I can think of, given the early history of the first few days, is that they (Malaysia) know what happened, and why, and thus, where it is, or most likely is, and thus, no need for a clairvoyant at all. Further, since that is not where OI proposed to look, and are looking, they know that OI will not find it.”
Ventus you are exactly right, Dennis, yes he’s right. OI are handing Malaysia a get out of jail free card. No wonder Like Going Lai looked so happy and relaxed in his recent press meeting. They’ve even got their men on board to make sure OI don’t stray from the designated (safe area).
Early on this blog and elsewhere there has been a lot of banter about the Malaysians not wanting the plane found (not sure myself why they would not want that). Nevertheless, recently many seemed to readily accept that the Malaysian deal with OI was genuine with no ulterior motives. I think there is a very good chance that Dennis and others have it correct. My initial reaction was exactly that – the Malaysians will agree to any search in an area that they believe does not contain the aircraft.
Malaysian officials did seem to know early on a lot about what happened. Perhaps a note had been left by Shah (perhaps more than one note) and they managed to squash it. (Something they seem to be good at). It could have been a scheduled email or other message.
In any case I see a lot of sense in Jean-Luc’s theory. It may require that that the ‘PIC’ ran out of fuel. Can that happen to an experienced pilot? I think the circumstances of Air China Flt 006, Air France 447 and even Asiana 214 tell us that experienced pilots can indeed make basic errors.
The Air China Captain violated a golden run taught very early in the instrument training process – believe your instruments – not your inner ear. He damaged the plane and came very close to a crash.
The Air France guys failed to recognize a basic stall (while their Captain was in the can working off the effects of his night with his mistress – not so professional either). By the time they collectively got their act together it was too late.
The Asiana crash was a little different, but it seems these guys were confused about the operation of some of the basic equipment (including the autothrottle).
Could Shah have run out of gas? He could have been under extreme pressure from the ground and perhaps from within the aircraft.
In any case, there could be some other explanation for a crash/ditching at sea to conclude the Jean-Luc theory.
Since we are pretty much all guessing, I would say Shah did not intended to kill the Pax. I think he was trying to achieve something and then land. He had to stay where he had that option as long as possible – not deep into the SIO.
Sorry, I have to comment on this.
Though I have no intention of discounting theories that suggest Malaysia knowing more, hiding information or not wanting the plane found, are we not overlooking the facts? Data (facts) seems to point the location of the plane in the region that is being searched.
@Arthur C. ‘Facts’ – and a basket of assumptions.
Hehe, true, “educated guesses”.
@ArthurC
“Though I have no intention of discounting theories that suggest Malaysia knowing more, hiding information or not wanting the plane found, are we not overlooking the facts? Data (facts) seems to point the location of the plane in the region that is being searched.”
Facts seem to point to the region that is being searched? You are joking, right?. There are no facts pointing to the region being searched. The area being searched represents a travesty of the facts, a manipulation of the data, or what passes for data, to divert attention away from ATSBs abject failure to find the plane, despite their assurances to the NOK that they would find it. At the start, Malaysia, at the highest level, told Australia, at the highest level, not at any stage to assume or suggest in public that the disappearance was due to malign intent on the part of the crew. For one thing, malign intent if substantiated, would send compensation claims through the roof, and for another could be politically very damaging for Malaysia. (which was the pilot’s intention btw) There is no scientific evidence to back up the new search zone. It is based entirely on false and phoney assumptions about retracted flaps pointing to a pilotless plane and consequently a terminus close to the 7th arc, and fanciful drift modelling that will serve to embarrass CSIRO for ever and a day. Not in their wildest nightmares, did ATSB ever think anyone would be dumb enough to come forward and offer to search this fantasy zone. Enter Ocean Infinity…
Take note for the record, the DSTG got it right first time, as will eventually have to be accepted.
OI have been duped into playing the role of fall guys in this charade. Malaysia will have no problems with OI going CV north. Just stay out of S38!
@Niels: To answer your questions in order:
1) Ian Holland’s paper was revised in January 2018 to differentiate between Class 1 log-ons using the LGA and the Class 3 log-ons using the HGA. See the section entitled “Effect of SDU Startup on the BFO”.
2) Yes, my guess is the 4 Hz shift in the FFB was caused by the thermal cycling of the crystal in the power down – dwell – power up cycle.
3) The shift of 4 Hz occurred between a Class 3 (HGA) log-on at 8:01:53 and a Class 3 (HGA) log-on at 15:59:55, with the power down – dwell – power up sequence occurring between those two log-ons. That’s also the sequence of events between the 15:59:55 log-on and the 18:25 log-on. I think there is a potential for another shift, although I don’t know the direction or the magnitude of the shift.
4) Shifts down in the FFB would make southern paths more likely, and shifts up would make northern paths more likely. I am not proposing that the power cycle causes another shift down. I think it is just as likely that the FFB shifted in the opposite direction. My observation is only that it could be very over-restrictive to assume there was no change in the FFB.
Victor,
Summary of FFB eyeball analysis of 9M-MRO data: For points before MH371 takeoff, FFB ~ 167 hz; after takeoff, FFB ~ 171 hz (with lots of scatter.) Points at 16:00 logon: FFB ~ 167 hz. Why the FFB changes after takeoff – unclear.
For MH370 (where we have ADS-B data, including ROC, tabulated at 1 minute intervals, unlike MH371, where the position reports are only tabulated at 5 minute intervals and no ROC is provided), no change in BFO offset is seen from before and after takeoff.
For all, regarding DSTG Fig 5.4, it should be remember that what is plotted is the BFO residual. It includes contributions, not only from SDU oscillator drift, but also from the flight path estimated from the ACARS 5-minute position reports, ACCESS-G wind model, satellite motion, channel number, EAFC+SAT calibration, and the actual BFO data. We have zero information regarding any of the latter. Before leaping to conclusions, it would useful to make sure that any jump in the BFO is not due to, say, the firing of the hydrazine thrusters for station keeping purposes.
@sk999: I only looked at data while on the ground at KLIA, so flight path estimation and wind model errors don’t matter. The actual statistics for the HGA log-ons are:
7:58:13-7:58:48, FFB=171.1 Hz, RMS error=0.8 Hz, N=25
15:59:55-16:00:23, FFB=166.8 Hz, RMS error=0.3 Hz, N=8
16:00:28-16:28:12, FFB=150.5 Hz, RMS error=0.9 Hz, N=81 (after calibration)
16:29:46-16:29:52, FFB=149.9 Hz, RMS error=0.9 Hz, N=16 (left gate)
Victor,
Yes, I also get FFB of 171 for 7:58 logon, but do not trust the channel dependent offsets, so would give it less weight.
16:29 was not a logon (?). Engines were being started; plane was still on the apron; not yet taxiway or runway.
@Ventus 45. The ABC audio adds to the earlier report on it. For those who have not listened to it, at 4:25 the currently drafted final report,
“…is already being reviewed by various states and organisations involved in the investigation ….it offers a number of safety recommendations based on a number of possible scenarios; and that includes pilot hijackings and cargo management, or should I say mismanagement and that is a tantalising clue that the investigation team they can’t quite rule on board systems disasters triggered by some sort of flash fire caused by illicit cargo; and that is where the finger of suspicion has always landed on the supposed consignment of two tons of mangosteen fruit. If it wasn’t fruit was it a dangerous number of black market lithium-ion batteries and perhaps a corruption of that KL-to-Beijing cargo business on Malaysian Airlines.
First, a list of “possible scenarios” will be controversial as to its omissions should this report come to be published.
Second, the final report being circulated now doubtless will suggest to some that this final search is not being taken seriously by Malaysia.
Third, the cargo-as-cause has not been discussed for a while. It will be interesting to see how in any account that is combined with subsequent navigation and lack of descent.
Fourth, he lists pilot hijacking but not ‘straight’ hijacking though that might have been missed inadvertently in the info provided or reported.
Fifth, he talks about black market lithium batteries. Black market I think has not come up before. It has overtones of risks of quality.
@sk999: Why less weight at 7:58? The RMS error is only around 0.8 Hz for the data set of 25 points, yet the offsets (when applied) are around -4 Hz.
I agree the data at 16:29:46 was not a log-on. It looks as though the plane was moving between 16:28:15 and 16:29:41, as the BFO error grows, so the plane had likely left the gate, but not yet on the runway.
Victor,
R.E. 7:58, it’s not the rms, it’s the particular combination of channel Unit Id and chnnel unit name (i.e., bandwidth and frequency). I prefer to cross-correlate data from that combination with other equivalent combinations collected at similar times. Just not enough data here to do that, although I haven’t dug in deep.
R.E. 16:28:15 to 16:29:41, this raises an interesting issue, possibly resolved here before, but not to me. At this time, the SDU is receiving valid navigation data, right? Needed to logon via the HGA. Except that all documentation says that ground track data (needed for the Doppler compensation calculation) is not available, since the ground speed is less than 25 knots. What is the SDU doing? Simply not computing a Doppler compensation value? Not unreasonable, given that the aircraft is hardly moving, and the SDU knows that the wheels are still on the ground But it would mean that the BFO now reflects the absolute speed of the aircraft, which, even though slow, is still enough to affect the BFO. In any case, that is the explanation I have though might explain those anomalous BFOs around 16:29.
@David
I was waiting for somone else to comment on the last part of the tape.
You have summarised it well from 4:20 or so.
But one point you missed.
Specifically 3:52 – 4:02.
Interesting.
@Ventus45 @David
The two possible causes mentioned so far seem more inwardly directed to MY as far as responsibility. I am thinking MY does not have too much freedom to shift blame to others.
@Ventus45. On the face of it OI will leave the search scene at the end of the 90 days if the aircraft is not found. That being the contract duration, unless the contract were extended I assumed they would. Finding it beyond that would bring no reward otherwise, unless ex-gratia.
What I think you allude to is the definity of, “… the plane is not found by the end of that 90 days OI will depart the Southern Ocean; we have that from the Malaysian Government”. This makes it sound as if there will be no extension, period.
It is possible also that that would be the end of the season for OI irrespective due to likely weather, their next job bearing down or expiry of contracts they have with their own staff, some of whom will have their own plans, and others; with a renewal in the next search season still possible.
@sk999
25 kts uncompensated would result in a substantial BFO effect. Potentially some 40Hz-50Hz or so depending on track obviously.
The oscillator power cycling, or retrace as it is commonly called, is harder to quantify and depends on the particular oscillator (even within the same family of oscillators). A reasonable rule of thumb is a part in 10^8 for ovens. That would amount to 15Hz or so at L1.
@sk999
Sorry, being a GPS person L1 is on my brain. I should have said L-band.
@David
” … WILL depart …. ” is the operative statement, if true.
How can he say that ?
OI obviously have their own plans. Voyages are planned a long way out. There is much to consider, logistics, crewing, ship maintenance (but is is pretty new), a whole host of things. Some people think the intended destination was CI and that is “the next best bet”. Well if that is so, OI can do donuts around CI for a year if they want to, the weather is NOT a factor up there, but I doubt it.
As you say, OI may have their own “time window” for other jobs, and/or the “season” (south of 30S) is a factor as well, as is crewing, but I think it is unlikely that OI will simply walk away from this.
I don’t think being “paid” by Malaysia has anything to do with it at all. If they find it outside the terms of the “contract”, then it becomes a lawyers feast in the salvage courts. They will get paid either way, whether they find it in the “contract” or not, ie, after the contract expires.
SC is making progress on the S.E. side of the Secondary Search Area, with the second cycle of AUV launches.
This area is of particular interest because of the potential MH370 End Point following a track of 180°T toward the South Pole.
https://www.dropbox.com/s/165pm9bpv334sqs/SC%20Track%2015032018.pdf?dl=0
@ventus45. You doubt that OI being paid reliably is an issue but were that so why negotiate a reward for the current search of a specific amount, presumably to be paid within a time limit, in the first place?
Once the contract expires and the Malaysians forward a final report to ICAO, I suspect it will become open season for spot searching under constraint akin to that for sunken warship searches.
Any money in return for success would be most uncertain both in amount and timing.
Personally I see the OI effort currently as a venture, having a reasonable prospect of some return and of costs being covered but also with personal accomplishment and philanthropy having played a large part in accepting the risk.
Unless new information or insight lifts the probability/sq NM of any new search field above the meagre and reopens the reward offer I doubt OI and others such as SeaTrepid would do other than go for other opportunities, current search costs being written off.
So I think it unlikely that the search will continue beyond the contract or into a new season.
@Victor, @sk999
At 16:27:59, the aircraft began to signal the ‘OUT’ message, ie doors closed & park brake off. That required a single R-ch burst (BFO=88) to reqst a T ch slot, then the T-ch burst (BFO=88), and the final R-ch acknowledge (BFO=85).
At 16:29:17, the aircraft began to signal the APU Shutdown Report, i.e. at least one engine running, supplying AC power, and APU then shutdown. That required a single R-ch burst (BFO=95), the T-ch burst (BFO avg=98.8). However, the final R-ch acknowledge associated with this exchange seems to have been missed by the GES, as the GES requested its retransmission.
Between 16:27:59 (beginning of ‘OUT’ message correspondence) and 16:29:27 (completion of APU Report correspondence), it’s safe to assume the aircraft had been pushed back, at least one engine started (taxi out may have been undertaken on one engine), and the AC power source switched from APU to at least one engine IDG. However, no Log On occurred during this period so the SDU ‘held-up’ during the power transfer.
@DennisW: Based on a small amount of research on the subject, retrace and thermal hysteresis (some references make a distinction between the two effects, and others do not) are almost guaranteed to cause a shift in the FFB before and after the inflight power down – power up cycle.
@Don Thompson: Thanks for digging in deeper. I think we can have high confidence that 9M-MRO was not stationary between 16:28:15 and 16:29:41, and therefore the BFO values in that period should not be used to calculate the FFB with assumptions of no aircraft motion. (The RMS error is way too high.) However, I do think the short burst between 16:29:46-16:29:52 can be used. That’s significant because we have two values of BFO through the R4 channel, which is what was used for the hourly handshakes after the 18:25 log-on.
@Victor
@Don Thompson
The aircraft is likely to have been pushing back from the gate between the ACARS ‘OUT’ message at 16:27:59 and the APU shutdown report at 16:29:17. During that period it would be normal for both engines to be started simultaneously and the APU would then be shutdown during the after start sequence. The aircraft may or may not have been stationary when the APU was shutdown at 16:29:17, depending on the length of the push back.
@Victor
Any offset would effect all BFO readings equally, right (including sat call C channel)? I know there could possibly be further drift during the flight.
@TBill: Yes. That shift would be superimposed on the corrections based on channel.
@sk999 said: But it would mean that the BFO now reflects the absolute speed of the aircraft, which, even though slow, is still enough to affect the BFO. In any case, that is the explanation I have though might explain those anomalous BFOs around 16:29.
I agree. With no AES compensation, the change in BFO is maximum for motion roughly along the E-W direction (more precisely, 88°-268°). So for motion west (towards the satellite), a speed of about 3 knots produces a +6-Hz shift in BFO due to uncompensated Doppler shift.
The MH370 research has become so highly technical, scientific and sophisticated, that I am asking myself for a very long time, whether something terribly basic could have been missed. Kind of missing the forest for the trees for being routine-blinded.
I have no particular supposition in mind concerning the above. But as an example, is there any way to have more fuel available than ACARS reports? I.e. is there a way to tamper with the data or to trick the system into believing there is less fuel remaining than there actually is?
A car’s fuel gauge can be affected by turns and ascents/descents.
Can the plane’s attitude, RoC or manoeuvres affect the plane’s fuel gauge?
> Victor Iannello says:
> We know that the aircraft was not fully fueled, and the range was therefore constrained to much less than
> the maximum.
@Ventus
Seems to me that interview you posted above is quite interesting. The other idea I heard for the first time was that if the digital FDR is not found, they can still reconstruct a lot of the info from the electronics which send the data to the flight data recorder.
@Niu Yunu: Not only would the ACARS reports would have to be hacked, but the fueling receipts would have to be tampered. I wouldn’t rate that very likely.
@Victor Iannello:
I don’t rate it very likely either.
If ACARS reports cannot be tampered with, could the plane’s fuel gauge itself be misled (intentionally or unintentionally) by the plane’s attitude or manoeuvres at the time of ACARS reports?
If cargo paperwork tampering is considered (see above), I don’t see why
fuelling paperwork tampering is beyond question.
@Niu Yunu: That a pilot could and would deliberately manoeuver a plane to give false fuel readings timed with the ACARS reporting times is close to a virtual impossibility.
Because the timing (of ACARS reports) is unknowable to the pilot?
Or because attitude/manoeuvres don’t affect the fuel gauge?
Or both?
@Niu Yunu: The timing of ACARS reports is unknowable, and the effect of attitude on fuel readings, if there is one, is unknowable. I am politely advising that there are more productive avenues to pursue.
As @sk999 states, the ADIRU doesn’t pass ground speed to its output buffer below 20kts. There is no freq compensation during pushback and slow taxi.
@Andrew,
Thank you for the reply on pushback and engine start. The FCOM does state APU off in the ‘Before Taxi Procedure’. I wasn’t sure if local procedures might have guided taxiing/single engine use.
@VictorI
Cultural differences are not a diversion in this case. It’s not only about data. Pure rationalists never seem to understand the bigger picture. Ration is only a tool to serve subconsience motives.
A very important tool but secondary. It can only try to make sence of things in a rational way while motives often are not rational at all.
Never mind. Nothing new anyway. Nothing found yet.
About cultural differences, instead enjoy this video.
People are the same wherever you go:
https://www.youtube.com/watch?v=Yu154T7hhes
@DrB
I believe there is value in computing a lot of flight paths like you did. Would it be possible to get the BTO/BFO calcs for them at the handshakes? Thank you.
> Victor Iannello says:
> I am politely advising that there are more productive avenues to pursue.
Sorry Victor, I didn’t want to get in your way. Apologies. I’ll remove myself.
I just thought it was an interesting idea, because …
> Victor Iannello says:
> […] deliberately manoeuver a plane to give false fuel readings timed with the ACARS reporting times
… because this would virtually guarantee that MH370 won’t ever be found,
as the pilot could fly to an area considered out-of-fuel-range, which will never be searched.
@TBill,
I can provide BTO/BFOs for a few routes if you tell me which ones you are interested in.
@VictorI
Thank you for the reference and the explanation.
“my guess is the 4 Hz shift in the FFB was caused by the thermal cycling of the crystal in the power down – dwell – power up cycle”
And:”… I think there is a potential for another shift, although I don’t know the direction or the magnitude of the shift.”
We would probably need the data of quite a number of these thermal cycling “events” to properly characterize the possible shift in FFB.
@Niels: Yes, I don’t see an easy way to bound the effect without experimental results. However, based on what we see on the ground at KLIA, and based on any of a number of references on crystal oscillators, the effect is known and appreciable. I find it interesting (but also frustrating) that for LNAV (great circle) paths with autothrottle engaged, there are a large range of paths in which the RMS error (assuming FFB=150 Hz) is dominated by the mean error. That means that if we allow a shift in FFB at 18:25, paths ending even north of 27S satisfy our criteria for BTO and BFO error. For instance, there is an LNAV path crossing the 7th arc at 26.6S latitude with BFO RMS error equal to 2.8 Hz if the FFB had shifted from 150 Hz to 157.3 Hz.
@Don Thompson
RE: “I wasn’t sure if local procedures might have guided taxiing/single engine use.”
For what it’s worth, I can’t find any mention of single-engine taxi procedures in the MAS manuals. In any case, it’s a very short taxi from gate C1 to RWY 32R at KUL, so there’s nothing to be gained by a single-engine taxi.
@Victor,
“That means that if we allow a shift in FFB at 18:25, paths ending even north of 27S satisfy our criteria for BTO and BFO error”
I expect that other parameters must also change in order to satisfy the criteria. For example, speed and fuel burn assumptions [or PDA].
So, back to my hobby horse . . . what range of average speeds around the point of closest approach, 19:41 to 20:41, would be necessary to fit the more northerly endpoints.
Which flight number?
The ATSB report, Underwater Search Definition (Aug 18, 2018), shows the flight paths of two flights that were used to test the ATSB’s flight reconstruction techniques. These paths are shown in Figs 30 and 31 of the report. The flight numbers are given as MH021 and MH009. The destinations are not given, but one can match them up with flight paths shown in the DSTG book, Figs 9.9 and 9.11 as being flights to Amsterdam and Frankfurt respectively. Further, the JON article specifically mention that MH21 traveled from KL to Amsterdam. (To be clear, these are flights at the same time as MH370 and thus involved aircraft other than 9M-MRO). Everything is consistent, right?
Not exactly. Communications between these flight and Malaysian ATC controllers are documented in the transcripts in Factual Information (Appendix 1.18A). What flight numbers do we encounter?
16:03:58 Lumpur Delivery: “Malaysian One Six is cleared to Amsterdam …”
16:26:01 Lumpur Delivery: “… Malaysian Six is cleared to Frankfurt …”
So it would appear that the flight to Amsterdam was actually MH16 and the flight to Frankfurt was MH6.
Which was it? I have no explanation.
Erratum – ATSB Underwater Search Definition date should be Aug 18, 2014.
@sk999
As I understand it, MH’s westbound flights are even numbered and eastbound flights are odd numbered.
MH005 = Frankfurt to Kuala Lumpur
MH006 = Kuala Lumpur to Frankfurt
MH016 = Kuala Lumpur to Amsterdam
MH019 = Amsterdam to Kuala Lumpur
MH020 = Kuala Lumpur to Paris
MH021 = Paris to Kuala Lumpur
I can’t find any flights labeled MH009. Something seems to be screwed up.
@Andrew, sk999. Pic ostensibly of MH009, Zurich to KL 2003.
https://www.jetphotos.com/photo/211487
@sk999
Steve, the flight numbers referenced in the labels for those figures are indeed odd. As Andrew has pointed out above MH16 was KL to Amsterdam (while the return flight is now called MH19 when 9M-MRO went missing the AMS-KL return flight was MH17, it was redesignated to MH19 after 9M-MRD was shot down while operating with the 17 flight number); MH06 was KL to Frankfurt.
MH009 used to be the Malaysia Airlines flight from Zurich to KL but it hasn’t operated for quite some time; I’m pretty sure that the cancellation of that service pre-dates the loss of 9M-MRO.
@David
Thanks. Yes, MH009 was the old Zurich to KUL flight.
SC has 7 AUVs in the water on the S.E. side of the Secondary Search Area, in the second cycle of AUV launches.
The first AUV will be ready for pick up tomorrow afternoon.
https://www.dropbox.com/s/x8ehoi1h30r8j4a/SC%20Track%2016032018.pdf?dl=0
@sk999, @Andrew, @Mick,
Here is a link to the MAS Timetable for the period 1st Dec 2013 to 29th March 2014.
https://www.dropbox.com/s/k49wrsy3rcm8dl3/MH%20Timetable%20November%202013.pdf?dl=0
The ATSB Report is incorrect.
Fig. 30 is not of MH021, neither is it of MH020 to Paris, which according to the timetable was operated by an A380.
Fig. 31 is not of MH009, neither it of MH010 to Zurich, which according to the timetable no longer operated, you had to transfer via Amsterdam.
The DSTG Report is correct.
Fig. 9.9 Is MH016 to Amsterdam and shows a take off at 16:36 UTC, which fits the clearance at 16:03:58 UTC. The timetable shows a departure of 23:59 local time and operated by a 772.
Fig. 9.11 is MH006 to Frankfurt and shows a take off at 16:49 UTC, which fits the clearance at 16:26:01 UTC.
The timetable shows a departure of 23:59 local time and operated by a 772.
@Brian said: I expect that other parameters must also change in order to satisfy the criteria. For example, speed and fuel burn assumptions [or PDA].
So, back to my hobby horse . . . what range of average speeds around the point of closest approach, 19:41 to 20:41, would be necessary to fit the more northerly endpoints.
For endpoints north of 34S, the LRC speed schedule fits better than constant Mach number. The speed does not change for more northerly paths. The BTO RMS error is between 31 μs and 32 μs over the range of endpoints between 32S and 27S latitudes. For each case, the groundspeeds between 19:41 and 20:41 vary between about 485 and 479 knots. Remember, the position at 19:41 is changing for each case. Relative to your closest approach observation, the groundspeed is not constant, and the positions at 19:41 and 20:41 do not fall exactly on a ping ring, so the results will be a bit different.
Regarding PDA, the position at 19:41 moves north for more northerly paths. You have heard me before discuss the possibility of a descent to FL200 and a hold, which would reduce fuel consumption. However, we don’t know how the plane was flown between 18:22 and 19:41, so we don’t know whether there was “too much fuel” to allow northern paths.
@Richard Godfrey
Thanks – that makes sense.
@VictorI
What is your best guess latitude on the 19:41 ring? Mine is currently ~6N. With that 19:41 latitude and an 00:11 latitude of ~30S indicates a great circle average ground speed of ~480 knots.
@VictorI
BTW, the track is ~171 and the FFB needs to be high by ~6Hz. BTO fit appears OK.
@DennisW: I don’t really have a best estimate of the latitude at 19:41, as I can only report a possible range. For great circle paths using LRC speed, the latitude at 19:41 varies between 3.8N and 9.2N for 7th arc crossings between 33.0S and 26.6S. The corresponding shift in FFB over that range would be between +2.9 Hz and +7.3 Hz.
@VictorI
Thx. I regretted posing the question retrospectively. My fixation on 6N has to do with analytics I cannot defend (in addition to Richards drift work).
I’m pleased to see that @ALSM finaly has pin-pointed the ‘blue panel’ on the map:
https://twitter.com/Airlandseaman
@Ge Rijn
please forgive me Victor,Slightly off topic I confess. But all work and no play makes Jack a dull boy.
I’m fascinated by the thoughts of the “blue panel”. I saw something similar in the British Museum when they mounted an exhibition of ancient roman murals from Pompeii.
@Victor Iannello:
is there any chance you would let Brock McEwen come back to this blog?
He was banned 2 months ago (for reasons not fully known to me and which at any rate are none of my business). I just wanted to say that I often found his postings interesting and thought maybe he could get a second chance after his extended absence? (perhaps on condition of him promising to not re-engage in whatever he did that annoyed you)
All,
Thanks for info r.e. flight numbers. One hopes that DSTG and ASTB were at least able to associate the BTOs and BFOs with the proper ACARS position reports. The DSTG punted on giving flight numbers other than for the KL-Beijing flights. Presumably it was Malaysia Airlines that was responsible for matching up the satellite data tagged, with the AES ID, and the flights themselves. Perhaps we now have a better understanding of how they had so much trouble keeping track of MH370.
@Perfect Storm: No.
@Rob
I understand ofcourse why you are ridiculing me about this ‘blue panel’.
I’m pleased to see @ALSM put in on the map, which implies he takes it serious. That has some more weight to me then your comment on this.
Fact: this ‘blue panel’ and associated debris field of ~13 spotted other pieces on 28-3-2014 could well fit all known data:
-It would fit the eastward current along Broken Ridge shown in the latest CSIRO/Griffin drift-analysis (the one where the CSIRO-hotspots were based on).
-It could well fit the drifting-time needed from the 7th arc till 32.4S/97.8E where the piece/debris was spotted on 28-3-2014.
-It could also fit the Inmarsat-data and fuel-data.
-The area is not searched yet. Probability increases while nothing is still found south or west from this area.
I think it also could be usefull when a reverse drift-study was done on the ‘blue panel’ like @Richard Godfrey has done on the debris spotted around ~29S west of the 7th arc used in @VictorI’s proposal associated with the debris spotted on 29-3-2014.
Excusse me VictorI/RichardG. Correction.. I remember now @VictorI did the reverse-drift-study on the pieces spotted west from the 7th arc around ~29S himself.
SC has started the collection of the AUVs on the S.E. side of the Secondary Search Area, following the second cycle of AUV launches.
The collection process will be completed tomorrow afternoon and then I expect SC to depart for Fremantle and crew change, refuelling and resupplying. Fremantle Harbour is currently expecting SC to arrive 22nd March 2018 at 04:00 local time.
I expect that the Tertiary Search Area on the S.E. side will be completed at the start of the next swing around 30th March 2018.
https://www.dropbox.com/s/zdmtbsq991oai87/SC%20Track%2017032018.pdf?dl=0
Fascinating reading technical stuff that you guys come out with. Way above my pay grade.
I contend that there was an initial pilot initiated high speed descent. Ask any airline pilot what are immediate actions for double engine failure at 40,000 feet. Considerations of depressurisation, core lock and more but also in MH370 case with the autopilot disconnect and ADG deployment he would need to increase speed by diving to get the ADG up to speed for hydraulic pressure to the flight controls. Could then trade kinetic energy back to optimum glide speed at say 30,000 feet to extend the glide. The ATSB assumption of an uncontrolled dive is wrong.
@Rob
To elaborate further on the ‘blue panel’ again.
You can think of it what you want, but this is the only piece photographed in the aerial search that at least has some clear resemblence of a possible aircraft piece.
Namely the left-wing flaperon of a B777.
It shows the same trapezium shape (not rectangular). The estimated surface-area by the trained spotters on the plane was 4m2. Which also fits.
Other features which I mentioned in lenght before also are positive indications imo.
Like to state again the ‘blue panel’ is not ‘blue’. It is white.
The blue color is caused by a thin sheet of water covering the main center area of the piece.
The forward edge, trailing edge and short edge on the ‘south’-side show the piece is white.
And imo this is even another indication the piece could be the left wing flaperon floating up-side down.
The hollow shape of a flaperon floating this way would cause a thin sheet of water to center in the middle area leaving the named edges free of water.
Once again the photo I enhanced and shared long time ago:
https://www.dropbox.com/s/stc5oqydhyysl2o/Xflaperon1.jpg?dl=0
@Byron Bailey, are you saying that a pilot-initiated vertical descent after both engines ran out of fuel, would be a sensible maneuver if the main goal was to enter into an extended glide?
@Ge Rijn, sorry, but this white rectangle whose pictures you have shared with us, doesn’t even look a bit like the flaperon.This has already been dicussed extensively many times and ultimately rejected.
Why can’t you let go of certain things which have been dismissed or debunked a long time ago, and move on to more sensible discussions? May I say this: my impression is, that the assessment of the debris is way beyond your limited expertise. It’s beyond my limited knowledge as well, as I’m ready to admit. Therefore I leave the in-debth discussions of the debris to those who know what they’re talking about – which of course doesn’t exclude us from asking probing questions 😏
That said, why don’t you continue to build a sensible scenario which includes the possibility of an extensive glide, although many think this is an unlikely scenario? I have dismissed this possibility so far because as a non-pilot I couldn’t come up with a good reason for putting the plane into a vertical high speed dive before entering into a glide. But if this maneuver does make sense after all for a pilot whose engines aren’t running anymore, the scenario may need to be revisited, and we could move on to the next question: what motive could the pilot (most likely Zaharie Shah) have for performing a long final glide? Making the searches more difficult by gliding away from the 7th arc would only make sense if ZS knew about a final arc. But there’s of course another possibility: it could’ve been a final challenge for a skillful pilot.
Ultimatly I hope that an extensive glide away from the 7th arc didn’t happen, because it would make finding the plane almost impossible…
@Byron Bailey. “Ask any airline pilot what are immediate actions for double engine failure at 40,000 feet.”
Would a professional pilot let it get to such a position, at altitude, sitting waiting for double flame out?
If he wanted to glide for distance why would he prompt a dive first?
Related, why would he not aim for a powered ditching or a powered dive?
ADG. Do you mean the RAT? Why do you suppose that would not work at the likely IAS?
Why do you suppose he would not have windmilling hydraulics anyway, that is before the APU came on line?
Since the outflow valves normally would close when cabin altitude rose above about 11,500 ft why would decompression concern him?
What would be the decompression rate above that cabin altitude with those closed, no bleed air?
If he was diving to avoid decompression why did a high descent rate take two mins from fuel exhaustion?
Would his left engine have relit?
If he glided for some time what is the rationale for the IFE not connecting?
I hope you will actually answer these questions. All of them.
@Byron Bailey: First, welcome to the discussion. I apologize for the barrage of questions, but I hope you will take the time nonetheless to provide thoughtful answers. Contributors here are both knowledgeable and curious, and include @Andrew, a senior airline pilot who has considerable experience flying and training on the B777.
Some comments. (There is some overlap with David’s questions.):
1) What is an ADG? Do you mean IDG? It is unlikely that windmilling in a high speed descent would allow the IDGs to produce electrical power to power the electric pumps, if that is the claim.
2) Level D simulations* show that the right and left hydraulic systems would remain pressurized due to the windmilling of the engines. That, combined with the pressure supplied by the RAT for the center hydraulic system, would provide more than sufficient authority of flight control surfaces, obviating the need for a high speed descent to regain control.
3) Once bleed air is unavailable to the air packs, the outlet valves would automatically close. There would be no rapid depressurization that would necessitate a rapid descent. Even if a rapid descent was necessary, procedures would be to deploy speedbrakes and descend at Vmo, which would not cause the (BFO inferred) descent rate of 15,000 fpm.
4) In a nutshell, I fail to understand how an intentional high rate of descent is consistent with an extended, controlled glide. Please further explain your rationale.
*Level D simulations for the behavior at fuel exhaustion were performed by @Andrew, as described here.
@Sabine Lechtenfield
May I say my limited expertise has identified as the first one; the outboard flapsection, the flaperon closing panels and the vortex generator, identified by @ALSM, as definitely being the right/engine vortex generator.
Trust me I took a very close look at the ´blue panel´ too.
It cannot be conclusivly identified but sure is not debunked yet at all and sure has resemblance to a B777 left wing flaperon floating up/side down.
Together with the associated spotted debris field at 28-3-2014 it has to be taken very seriously imo.
> Victor Iannello says:
1) What is an ADG? Do you mean IDG? It is unlikely that windmilling in a high speed descent
> would allow the IDGs to produce electrical power to power the electric pumps, if that is the claim.
Victor, I think ADG means Air-Driven Generator here, another term for Ram Air Turbine (RAT) used e.g. by Canadair.
@Ge Rijn, I agree with you insofar as the blue color is probably an optimal Effekt and not real.
You provided a link to Mike Exner’s Twitter account and claimed that ALSM put the panel on the map. I can’t find any tweet which deals with the floating panel. Could you be more precise?
Correction: I wanted to say “optical effect”. Autocorrection… 😣
@Byron Bailey
Recently some of us have indeed been discussing that the final descent could be intentional based on the data. I personally combine this with the thought of picking a crash spot (eg; Broken Ridge at 32.5S).
@Peter Norton said: I think ADG means Air-Driven Generator here, another term for Ram Air Turbine (RAT) used e.g. by Canadair.
Thank you. I’ve never seen that abbreviation used for the RAT of a B777.
@Byron Bailey said: Ask any airline pilot what are immediate actions for double engine failure at 40,000 feet.
The “Gimli Glider”, Air Canada Flight 143, was a B767 that ran out of fuel at 35,000 ft. Rather than initiate a rapid descent as you claim would be expected, the pilots chose to descend at the optimum glide speed of 220 knots. The hydraulic pressure supplied by the RAT proved to be sufficient at this airspeed.
@Byron Bailey
RE: “I contend that there was an initial pilot initiated high speed descent. Ask any airline pilot what are immediate actions for double engine failure at 40,000 feet.”
I strongly disagree with your contention that a pilot initiated high speed descent was necessary.
1. Depressurisation: As others have already said, the outflow valves would automatically close to limit the cabin altitude. The cabin altitude would continue to rise due to air leakage, but at a much slower rate. As Victor mentioned, an emergency descent at Vmo with full speedbrakes would not produce a descent rate anywhere near the 15,000 ft/min suggested by the BFO analysis. In any case, why would a pilot intent on destroying the aircraft and killing the occupants care about depressurisation? The crew oxygen system would have provided more than enough oxygen for his own needs.
2. Hydraulic Pressure: The RAT is able to supply its full rated capacity at a minimum speed of 115 KIAS. The pilot did not need to “increase speed by diving to get the ADG up to speed for hydraulic pressure to the flight controls.” In this case, the L & R hydraulic systems would remain pressurised by the windmilling engines while the C system flight controls would be powered by the RAT.
@Victor
Your SPOLE path has an interesting quality.
Below is Time_BFO Measured_BFO Predicted_BFO Error
Arc3 20:41____141.4____137.4____4
Arc4 21:41____168.4____164.1____4.3
Arc5 22:41____204.4____199.7____4.7
So that path is closest to constant error, assuming a bias shift at 18:25. No other path so far looks just like that. Unfort that is only meaningful if the reported BFO are almost exactly accurate, excepting for bias shift.
I assume maneuvers before Arc3 after Arc5 so above is the constant flight period assumption.
@Andrew, @David: We all had similar reactions to @Byron Bailey’s comment, which is not surprising since contributors here have been sharing and digesting information on this topic for a long time. I hope Byron chooses to respond, as he has been one of the most vocal proponents of a controlled glide after fuel exhaustion, and an informed dialogue with him would be helpful.
@TBill: I don’t agree with your rationale for ignoring the BFO errors at 19:41 and 00:11 when there are paths that produce low RMS errors with those included.
One thing you will see is that if you allow for a shift in the FFB equal to the mean BFO error between 19:41 – 00:11 (inclusive), more northern paths maintain low BFO error, and also the “problem” with the large BFO error at 19:41 disappears.
@Sabine
“what motive could the pilot (most likely Zaharie Shah) have for performing a long final glide? Making the searches more difficult by gliding away from the 7th arc would only make sense if ZS knew about a final arc. But there’s of course another possibility: it could’ve been a final challenge for a skillful pilot”
I for one can see how an extended glide following a flight that had been piloted in such a manner so as to achieve the maximum possible distance on the available fuel, for example a constant M0.82 at FL400 on a great circle toward a pre-selected waypoint, could be considered a natural option for a pilot to pursue. We shouldn’t forget that although the pilot may have been unaware of the potential path-reconstructing properties of ping arcs, he may nonetheless have thought it prudent to gain as much additional range as possible after flameout, if for no other purpose than to make it more difficult to determine where the flight could have ended, based on maximum cruise range estimates. Intentionally allowing the SDU to reboot once the fuel ran out may have been to signal that the flight had continued until fuel exhaustion (if indeed it was intentional. It cannot be proven it was intentional, but it would arguably be unwise to dismiss the possibility) If signalling that the plane had flown to fuel exhaustion was part of the plan, then it would make sense to glide further on as far a possible, as an insurance.
The pilot had planned to make the plane disappear. His actions suggest this, the path the plane followed between 1722 and 1822 suggest this. It is very difficult to argue otherwise. In the context of this, noth flying the maximum distance possible, followed by an extended glide makes local sense.
With all due respect, hoping that he didn’t perform an extended glide is not going to make our task any easier. Facing up to the unpleasant realities of the situation is going to be much more useful.
@Victor
“One thing you will see is that if you allow for a shift in the FFB equal to the mean BFO error between 19:41 – 00:11 (inclusive), more northern paths maintain low BFO error, and also the “problem” with the large BFO error at 19:41 disappears.”
Can I use the orig Iannello/Godfrey McMurdo report path as an example of what you are saying?
@TBill: Choosing LRC speed matches the BTO better for the McMurdo path, although there is little difference in BFO error between constant Mach number and LRC. If you want, I can send you the path data.
@VictorI
“The “Gimli Glider”, Air Canada Flight 143, was a B767 that ran out of fuel at 35,000 ft. Rather than initiate a rapid descent as you claim would be expected, the pilots chose to descend at the optimum glide speed of 220 knots. The hydraulic pressure supplied by the RAT proved to be sufficient at this airspeed.”
What you say is true. This also applied to the TranAt flight, but both these flights had passengers and cabin crew that were alive and needed to be brought down safely, and the pilots were gliding toward target airfields, which necessitated a carefully controlled gradual descent rate particularly in the early stages when it wasn’t certain if the aircraft were capable of reaching the airfields.
But the situation with MH370 is different.
It’s quite possible the plane was at 40,000ft when the RH engine flamed out. When the LH engine followed suit a few minutes later, the aircraft could have begun to loose altitude at a faster rate than the pilot had anticipated, but without gaining much kinetic energy in the form of increased airspeed. The pilot may have thought it was a better bet to fly aggressively, drop the nose to both gain kinetic energy and ramp up the ground speed, distance covered, then climb back up to 30,000ft and set up a glide.
@Victor
OK I think I have the LRC from Richard’s prior post of 18 flight paths. But I am mainly looking BFO trend at the moment.
@Rob: With an experienced pilot at the controls at fuel exhaustion, if the intent is to glide, the only rationale way of flying is to control airspeed with pitch, as was done with the Gimli Glider. It makes no sense to allow wildly fluctuating airspeed or vertical speed. A 0.8-g descent would be quite extreme, and if there was recovery to an altitude, there would be a penalty in the total energy (potential + kinetic). A 15,000 fpm descent is simply not consistent with an intention to glide as far as possible.
@TBill
My previous post of 18 flight paths included BTO and BFO for each data point throughout each flight.
@Sabine Lechtenfeld
RE:”.. ALSM put the panel on the map. I can’t find any tweet which deals with the floating panel.”
Click on the graphic in @ALSM’s tweet (14 march) to enlarge it and you’ll find in the far right side the pin-point with ‘Blue Panel’ next to it.
And it’s at the right location too at ~32.4S/97.8E (as expected from @ALSM):
https://twitter.com/Airlandseaman
@Richard
Thank you I have downloaded the paths previously.
Thanks, @Ge Rijn, I found it.
@Rob, yes, your reason why ZS could have tried to glide away from the location where he ran out of fuel, does make some sense, even if he wasn’t aware of a final arc. But going into a steep descent first still doesn’t make a lot of sense to me. I hope that Byron Bailey will explain his view some more. But I know no incident at all, where a pilot choose to go into a steep descent when he wanted to maximize the length of the glide. So, if Zaharie Shah did indeed initiate a steep dive and then recovered the plane in order to enter into a glide, he must’ve had other motives for executing such a peculiar maneuver. Extending the length of the glide cannot have been one of the motives.
@Rob
In any intentional scenario I see no rational for an extended glide ‘as far as possible’ for the reasons you suggest. A relatively short glide would also meet the same objectives as you mention.
I see a rational though for a pilot with the intend to end the plane (and himself) in a ditch the way @Byron Bailey also suggests.
After the second engine flame-out the pilot would have known the end was very near. There would have been no reason to extend this with a glide ‘as far as possible’. Rather he would have chosen to speed-up things by inducing a high speed descend, recover and glide as short as possible to a (pre)chosen destination.
This could also explain the two minute time-delay between fuel exhaust and final BFO’s. A pilot would have to prepare himself (and the plane) for some time to execute an action like this.
@Byron Bailey wrote “Ask any airline pilot what are immediate actions for double engine failure at 40,000 feet.”
Why would “any airline pilot” attempt to maintain cruise altitude with the obvious fuel exhaustion scenario looming?
@Ge Rijn wrote “I’m pleased to see @ALSM put [blue panel] in on the map, which implies he takes it serious.”
GR: that is rather a rash assumption on your part.
@Ge Rijn
“In any intentional scenario I see no rational for an extended glide ‘as far as possible’ for the reasons you suggest. A relatively short glide would also meet the same objectives as you mention.”
Well, the question of short versus long is largely academic when you consider that a 13:1 ratio glide starting from 38,000ft ends just beyond the southern boundary of the ATSB search area. Our pilot would have been quite capable of performing a 17:1 glide if he wanted. So I’m saying that a piloted glide from 38,000ft say, can comfortably accommodate a steep descent phase, and still have enough rang to have ended outside the original search area.
The ATSB dismissed the extended glide on the grounds that a downward acceleration of 0.68g apparently coinciding with an 8 second snapshot of the descent profile, the timing of which was determined by the combined required APU/SDU startup times, and evidence of retracted flaps, and very contestable evidence the the debris condition precluded a soft ditching. They were on a very sticky wicket.
@Rob
You said to Sabine:
With all due respect, hoping that he didn’t perform an extended glide is not going to make our task any easier. Facing up to the unpleasant realities of the situation is going to be much more useful.
You cannot pick and chose your “realities” and remain credible. The debris (and lack of debris) distribution have relegated your scenario to the list bin a long time ago.
@Rob
“list” = “dust” above
@Rob and @Dennis,
I might add that my hopes and wishes have exactly zero influence on the current search, and they also don’t change what actually happened. I could as well say, I hope that Schrödinger’s Cat is alive😉
As a cat lover I seriously prefer that alternative.
@Don Thompson
@ALSM would not put the ‘Blue Panel’ in a graphic he published himself on Twitter if not taking it serious.
But I agree @ALSM has to speak on this for himself.
I hope he does.
@Rob
Like @Dennis says; the original search area is in the dust-bin for a long time. No-one ever is going to search beyond 38S anymore and for very good reasons.
The only assumption you had that still stands is the possibility of a glide and ditch-like impact.
I stick with what I said. I ran the scenario in an Emirates B777 Simulator in Dubai shortly after the ATSB End of Flight report. Boeing use the term RAT -( I now fly a Challenger – ADG ) but a ram air turbine actually drives the air driven generator to supply hydraulic power ( priority ) to the flight controls plus emergency electrical power – I have done several test flights where the minimum flight envelope speed was 140kts. B767 and B777 are very different animals as B777 is fly by wire. Read my statement Sabine nowhere do I state a ‘ vertical’ descent
@Byron Bailey: I’m sorry, but you are not making sense. There is no need to descend at 15,000 fpm to reach 140 KIAS. Please explain what you are trying to say. And please answer the questions that we posed. Saying “I stick with what I said” is no help.
Andrew says:
When AC power is lost, the RAT automatically deploys to power the standby electrical system. The APU also starts automatically if there is fuel available.
I don’t know if this information might be of any value to this discussion, but yesterday I watched a documentary on TACA Flight 110, performed by a Boeing 737-300 which suffered a dual engine flameout. I don’t know how accurate the documentary is, but subsequent to the flameout the pilot says “manual APU power on” (seemingly as an instruction to the co-pilot). That called my attention, as it gave me the impression that the APU had to be started manually (not automatically as mentioned above). Sure, it’s not the same aircraft type, but I thought I’d share it here in case it might be helpful.
@Byron Bailey wrote “I stick with what I said.”
Oh my, that was a productive engagement.
@Byron Bailey
RE: ”I ran the scenario in an Emirates B777 Simulator in Dubai shortly after the ATSB End of Flight report.”
A number of end-of-flight simulations have been run by different parties. Would you care to share the results of your simulation for discussion?
RE: ”I have done several test flights where the minimum flight envelope speed was 140kts.”
Nevertheless, the Boeing maintenance training manual states that the RAT is capable of suppling its rated capacity (10 gpm @ 2850 psi) at a minimum airspeed of 115 KIAS. In other words, the RAT will function at ANY speed within the aircraft’s ‘normal’ flight envelope. A high speed dive is not required for the RAT to operate and power the centre system flight controls.
RE: ”B767 and B777 are very different animals as B777 is fly by wire.”
Nobody said they were the same. Victor used the ‘Gimli glider’ incident to show that a high speed dive is not necessary following an dual engine flameout. Why is the B777 any different in that regard?
@Perfect Storm
Since I’m here… from the B777 AMM
The APU will automatically start for these conditions:
* Airplane is in the air
* Power removed from the left and right transfer buses.
The ELMS contains the APU autostart logic and sends the start
signal to the APUC.
The ram air turbine will automatically extend during an APU
autostart.
@Perfect Storm
RE: ”it gave me the impression that the APU had to be started manually (not automatically as mentioned above). Sure, it’s not the same aircraft type, but I thought I’d share it here in case it might be helpful.”
Thanks for your comment. I’m not very familiar with the B737-300 systems, but I don’t think its APU is designed to auto start. The B777 APU is designed to auto start if both transfer buses lose AC electrical power, without any action by the crew. Nevertheless, the DUAL ENG FAIL/STALL checklist requires the crew to manually start the APU (APU selector to START, then ON) as a backup, in case the auto start fails.
I’m not familiar with the B737-300 systems, but I don’t think its APU is designed to auto start.
@Andrew, @DonThompson: thank you for the technical insights
@all
Personally, I find the range of expertise and commentary here fascinating. However, transcending that is the notion of who is “in charge”. What is really going on, and who is calling the shots?
My propensity in similar situations was to open a big can of “whoop ass” to get answers. But then again I was in charge, not an observer. What in the world is going on? Is there someone who is leading this endeavor? If so, who?
@Byron Bailey
A word of advice: Some very smart people contribute to the discussion on this website. If you want to be taken seriously, you need to justify your comments with detailed reasoning, expect to be challenged and be prepared to engage in the discussion. Comments such as “I stick with what I said” won’t earn you any credibility.
@Andrew
@Perfect Storm
Regarding the APU start sequence on the September 2005 version of the FCOM for the B737-600, -700, – 800 states;
‘APU Start
The automatic start sequence begins by moving the APU switch momentarily to START.’
There is no mention of an auto-start capability.
@Mick Gilbert
Thanks – the FCOM (page 7.30.2) for the B737 ‘Classic’ (-300/-400/-500) reads the same.
SC is heading for Fremantle.
@Byron Bailey
Something else on the RAT. The turbine is governed and has a constant rotational speed independent of the airspeed (above ~115KIAS):
https://aviation.stackexchange.com/questions/19232/is-there-a-maximum-airspeed-to-deploy-a-ram-air-turbine
@Andrew, @Mick Gilbert:
Thank you for solving this little mystery.
After all it doesn’t seem like something could be gained from it for understanding MH370’s final moments, unfortunately.
@Byron Bailey: If you are trying to make the case for an extended glide after fuel exhaustion, you undermine your own case by making false claims and by using senseless arguments.
But here, I’ll help you. The BFO data represents two points spaced 8 seconds apart in time. It’s physically possible that after a steep descent, the pilot chose to carefully recover and enter into an extended glide, even if we can’t explain the reason for either the steep descent or the extended glide. Perhaps the pilot had second thoughts about the steep descent, and self-preservation instincts took over. Perhaps the pilot thought it would be fun to push the plane to its limits before maximizing the distance into the SIO after fuel exhaustion. Perhaps after a steep descent and a long glide, the pilot chose to end the flight with a high-speed impact, which could explain much of the debris. We simply can’t be sure about how the plane was flown after the last SATCOM transmission.
Now, to be clear, I believe the scenario that is most consistent with the evidence is a high-speed impact that occurred shortly after the last SATCOM transmission, and I think it is very sensible to define the search based on this scenario. However, we can’t know that for sure. I would place the probability of an impact close (within 25 NM) of the 7th arc to be 80%. However, it is not easy to estimate a probability for human behavior, especially since abnormal human behavior likely was responsible for the flight into the SIO.
@Victor
To add to that, @Deplorable Edward (pilot) has been making some 4th anniversary tweets. He says:
“From a pilot perspective, on my last flight I would not want the autopilot to be on much. I’d have it off and have fun with dives and climbs and steep turns”…and
“I could, of course, be totally wrong—any predicted path that fails to consider a live, conscious, pilot to the end won’t find #MH370 If found in current search area I would be thrilled for the NOK…”
I always liked Edward’s ideas though he is not using BTO/BFO. If we do not find MH370 now, then up around L894 (21S) is a warm spot for me and Edward.
What is Byron’s explanation for the flaperon edge damage? In your view, Victor, I think it could be high speed dive with maneuvers to pull out?
Consider, for example Birthday waypoint flyover, with U-turn at Arc7 gliding back to about Arc6 to hit the waypoint. Another quote from the deplorable guy above:
“With respect to the 7th arc, my presumption is #MH370 was flown westward in the final minutes to allow the pilot to see better. The wreckage will not be E of the arc IMHO.”
@TBill: As I told Ed on Twitter, we can’t construct a search based on “hunches”. He agrees that a progressive search pattern is better. As for saying that as a pilot, he would not want the autopilot to be on much, I responded that 6 hours of hand-flying at FL350+ would not be much “fun”. He didn’t disagree. His main objection seems to be the rigid assumption of automated flight the entire way without allowing for at least occasionally some amusement.
Frankly, I have not seen any hypothetical impact point along the 7th arc that employs a set of assumptions that collectively have high probability. I see no alternative but to continue the progressive search as far north as possible at a width of +/-25 NM, recognizing that even that search strategy may fail.
I have no idea how @Byron Bailey explains the flaperon damage, as so far we have not been able to get him to supply clear explanations to his claims. Why he chose to make unsubstantiated statements on a forum of this type at this time is unknown.
People have been struggling with the obvious inconsistencies of a “rational type rated pilot”, apparently leaving it until after total fuel exhastion, to execute the “end game”, whatever the paln initially was, either a dive early from high altitude (with the obvious rish of in flight breakup) or a extended glide to either ditch, or make a final dive from the glide at low altitude.
One proposed possible explanation.
Let’s assume Z had set his final planned top of descent waypoint sometime between the FMT and say, 20:00 utc. Precisely when it was set does not initially matter (though generating a precie path will dpend on where and when it was actually set). In any case, let us assume that this TOD waypoint was somewhere that was consistent with about 40 minutes fuel remaining, ie, about 23:30 to 23:40, to allow a standard descent rate to say 5,000 feet, to execute the end game.
Everything was going according to plan, but now, after all the high pressure excitement of the past few hours, it was now totally quiet and boring. He was very tired, and his body clock was now well past his normal bed time anyway. He unintentionally “nodded off” at some point after setting the TOD waypoint. The aircraft flew on as programmed.
The final TOD waypoint is reached at 23:40, and the aircraft then follows a path as per earlier end of route dicontinuity discussions. The chimes go off, but Z is now in REM sleep, and does not awake.
About 35 minutes later, the right engine runs out of fuel, but he was so sleepy, he still did not awake to the chimes.
The aircraft flew on, and slowly descended single engine, but all was otherwise normal.
Left engine ran out of fuel soon after, autopilot disconnected, but he still did not awake to the chimes.
The aircraft soon developes a high bank spiral descent, with increasing “g” and buffetting. The buffetting and “g”‘s finally wake him up now.
He is initially disoriented, but he quickly realises no power, no fuel, and the time, he realises that he had nooded off, and commences recovery action, successfully.
The explanation for the “8 seconds” could be as simple as that.
Victor, I agree with everything you said.
There may be various reasons why ZS might’ve recovered the plane from a steep descent and then entered into a glide. You named a few valid reasons. But if he did perform such a peculiar maneuver, it cannot be explained by textbook piloting. However, if ZS was really alive and conscious until the very end of the flight, he had been done with textbook piloting a long time ago. Sorry for being very drastic, but we are talking about the moment when he was about to commit suicide! There are no rules or recommended procedures for that.
A note to those who believe that Zaharie Shah has abducted and eventually crashed the plane: it’s highly unlikely that he was nothing but a cold blooded machine, who went through with his plans without a mistake, doubts or moments of hesitations. That’s the reason why end-of-flight scenarios with a conscious pilot are so difficult to predict.
But as Victor said in his last paragraph: right now there’s nothing which contradicts the scenario of a crash after a steep descent near the 7th arc. That other scenarios can’t be excluded, isn’t a contradiction. If the renewed search, which seems to be executed with great diligence by OI, produces the plane, we will all be very happy. If the plane won’t ne found – well, that would be a result, too, even if it wouldn’t be what we all hoped for. But only then it is time to re-think the search strategies. OI simply cannot hunt down all crash locations which are favored by us for varying reasons – even If everybody believes, his or her scenario is the o only plausible solution 😋
@ventus45: We can construct many hypothetical scenarios in which the pilot recovers from a steep descent, and it will be impossible to decide which is most likely. The question is, what is the probability of a recovery versus a continued steep descent until impact, and how do we incorporate those probabilities into a search pattern. Also, as @David will remind us, the missing IFE log-on has to be explained. It’s possible the APU ran out of fuel, but our best estimates are there should have been sufficient fuel for the IFE log-on. That’s another indicator of an impact soon after the last SATCOM transmission.
@TBill @VictorI
To shime in. We know now the (by @Andrew’s simulation) the flaperon was not floating freely and therefore probably not trailing edge-damaged or seperated by flutter forces. The outboard flap section would be even harder to explain this way. This leaves other high aerodynamic forces that could have damaged and seperated those pieces in flight. But there are no examples this happened in previous high speed dive events/crashes (SilkAir, ChinaAir) relating to those control surfaces.
And even if something like this happened it remains very hard to explain why the big right hand engine cowling piece had his complete forward edge lip intact and the nose gear door piece showed no compression damage whatsoever.
This all is just not compatable with a high speed nose down impact.
@all
Thinking ahead to the future search area Site 2, we have:
Site2 Area1 (Pink)
Site2 Area2 (Dark Blue-Broken Ridge)
Site2 Area3 (Pink)
I hesitate to share this drawing, but here is my “artistic” approximation of the Site2 Area2 BR region: Site2_Area2_BR
The Site2 Area2 BR (dark blue area) seems to mainly comprise the fairly thin zone of the steep cliffs, and for some reason they toss in the “tongue” at the lower right which would apparently be a steep slope down to the bottom of the trench.
Site2 Area1, presumably the pink area below Area2, should perhaps be called Diamantina Escarpment. It does not appear to be an easy-to-search area either. It should be noted that current search area Site1 Area3 also has some of the Escarpment volcanos/etc, as we now know from the search last couple weeks.
@Ge Rijn
The controlled ditch has a number of issues, which you simply ignore:
1> The lack of IFE login – A big issue for me.
2> The BFO data – It is unambiguous.
3> Running out of fuel – If a PIC intended to make a controlled ditch he would have done so before fuel exhaustion.
4> OI search strategy – OI has more information than we do, and they are searching +/- 25nm.
5> Debris analytics – You are drawing conclusions from pictures, and have no pedigree in this domain. People who are qualified have actually breathed on the debris, and have advised OI to search +/- 25nm.
@DennisW
-1; The lack of IFE log-on is still not explained by the ATSB or anyone.
The ATSB gave one possibility; the IFE was switch off before the log-on.
-2; The BFO data is unambiguous. I agree.
-3; Unless a PIC intended a controlled ditch after fuel exhaustion.
-4; Could well be true. And +/-25Nm could well be sufficient. But the risk of being ~20Nm short in this search in certain areas I would not take if I was OI.
-5; On debris analytics. Only based on pictures me and others outside the official investigation have done several identifications of debris before they were confirmed by the officials.
@DennisW
You said to Ge Rijn “OI search strategy – OI has more information than we do, and they are searching +/- 25nm.”
And “Debris analytics – You are drawing conclusions from pictures, and have no pedigree in this domain. People who are qualified have actually breathed on the debris, and have advised OI to search +/- 25nm.”
Dennis, you are an intelligent person, but you are assuming here that OI have more information than we have. However, You don’t know that as fact. On the contrary, it is very unlikely that OI have more information than we do. OI are going purely on the so-called expert advice of ATSB. The +/-25nm limit is based on 1) the interpretation of the BFOs 2 minutes after running out of fuel as indicating an unpiloted descent, 2) Retracted flaps indicating unpiloted descent. 3) State of the debris indicating a high energy impact therefore unpiloted descent. 4) The desire to keep the search area within practical limits.
OI had no option but to take the ATSB advice on trust. Unfortunately, the wisdom and reliability of that advice is highly doubtful. Retracted flaps do not necessarily indicate an unpiloted descent, neither does the condition of the debris.
If the pilot’s plan was to deliberately run out of fuel at 40,000ft then fly his plane with only the RAT providing hydraulic power, sans APU, sans flaps, then ditch just forcibly enough to sink it as rapidly changing as possibly, in as few pieces as possible, then the debris does in no way conflict. You don’t need a pedigree in the domain of debris analysis to see that. The ATSB certainly don’t have one.
When Byron Bailey first had the temerity to suggest the flaperon damage suggested a soft ditching, the ATSB went out of their gourds and published an official refutation in their next progress report (putting the record straight, was the term used as I remember). Coincidentally I had independently suggested exactly the same around the same time ie. that the flaperon pointed to a soft ditching, and they felt it necessary to point out why I had to be wrong. In short, the ATSBs conclusion about the plane being unpiloted was NOT up for discussion. It was fact, get over it.
It was fact because it had two particular advantages, it limited the search area and it kept the Malaysians happy. Nothing has changed!
The lack of an IFE logon remains an open question. It is not necessarily evidence of an unpiloted descent. The plane refuses to be found within 25nm of the 7th arc – now that IS evidence of a piloted descent.
You said to me “You cannot pick and chose your “realities” and remain credible. The debris (and lack of debris) distribution have relegated your scenario to the list bin a long time ago.”
I do not pick and choose my realities. I leave that for you.
OI have much to ponder on as they make their way back to Port Henderson, still empty handed.
> 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.
On the contrary, in the “hard ditching scenario”, you don’t know if you survive the touchdown badly injured and then
– have to swim until you drown
– or get eaten alive by sharks (given that you are possibly bleeding)
– or get stuck in the cabin, slowly drowning in the water pouring in
– or have to spend a long time in the broken plane before it sinks
… all of which results in unimaginable agony.
Why would you opt for such an unpredictable plan ?
@Peter
Good points, IMO. It also dovetails with my objection of running out of fuel before you do it (ditch soft or hard), when you have much greater control of the impact with engine power.
Frankly, after giving the Ge Rijn and Rob responses the dignity of extended consideration, I have to put them in the lame category.
Reading up on Suicide by pilot, I realize that while rare, these tragedies occur more often than I had thought (the ratio of pilot suicide crashes to pilot-error crashes is given as 1:3). Even often enough to have become subject of studies, from which I would like to quote:
« In a study of pilot suicides from 2002-2013, 8 cases were identified as definite suicides, with 5 additional cases of undetermined cause that may have been suicides. […]
Most cases of suicide by pilot involve general aviation in small aircraft. In most of these, the pilot is the only person on board the aircraft. In about half of the cases, the pilot was using drugs, usually alcohol or anti-depressants, that would ban them from flying. Many of these pilots had mental illness histories that they had hidden from regulators. […]
In a 2016 study published in Aerospace Medicine and Human Performance, Kenedi et al. systematically review suicide and homicide-suicide events involving aircraft. They note that “In aeromedical literature and in the media, these very different events are both described as pilot suicide, but in psychiatry they are considered separate events with distinct risk factors.” The study (1999-2015) reviews medical databases, internet search engines, and aviation safety databases and includes 65 cases of pilot suicide (compared to 195 pilot-errors in those years) and 6 cases of passengers who jumped from aircraft. There are also 18 cases of homicide-suicide involving 732 deaths. Pilots perpetrated 13 homicide-suicide events. Compared to non-aviation samples, a large percentage of pilot suicides in this study were homicide-suicides (17%). […]
They state: “[…] 5 of 6 homicide-suicide events by pilots of commercial airliners occurred after they were left alone in the cockpit. […] No single factor was associated with the risk for suicide or homicide-suicide. Factors associated with both events included legal and financial crises, occupational conflict, mental illness, and relationship stressors. Drugs and/or alcohol played a role in almost half of suicides, but not in homicide-suicides.” »
Another underwater aircraft-shaped image story here. Despite the patent lack of image resolution apparently the author could determine that the airplane was riddled with … wait for it … bullet holes!
@Mick Gilbert: This was a conscious effort to spread a fake story to generate clicks, and reporter/editor Kate Schneider should be ashamed.
@ DennisW
2> The BFO data – It is unambiguous.
The BFO at 0019:37 is not a certainty as indicated by the notation provided in the ATSB report (The Operational Search for MH370), and therefore the claimed decent rate at that moment comes into doubt. You will recall that this was originally an outlier and discarded by the investigators, it was then brought back based on Holland’s study.
The note, I believe, was provided to satisfy elements within the investigation team who were not in consensus as to the final moments of the flight.
@Pilatus
The note, I believe, was provided to satisfy elements within the investigation team who were not in consensus as to the final moments of the flight.
I have the same recollection. My sense at the time was that the lack of consensus was due to “polling formalities” rather than specific technical objections.
> Peter Norton says:
> 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.
I looked into each listed incident of pilot suicide provided here for a reality check. My result is that all incidents are nose-down, high speed crashes (into the ground, buildings, mountains or river/ocean), invariably leading to instant death.
2 notable exceptions stand out from the crowd however:
• On July 13, 1994 a Russian air force engineer stole a plane, which he circled until it fuel exhaustion and crashed.
• In the 1999 Air Botswana incident, the pilot “threatened to crash it into an Air Botswana building, saying he had a grudge with the airline’s management. He demanded to speak to Botswana’s vice-president and was about to be put through to him when the plane ran out of fuel. Officials in the control tower told [the pilot] that there were people in the Air Botswana building, whereupon he crashed the stolen plane into 2 other ATR-42s on the tarmac. All 3 planes were destroyed in a fiery crash, and [the pilot] was killed.”
So out of 26 incidents we have 2 pilot suicides that lasted until fuel exhaustion,
1 of which included negotiations*.
* (I mention this although I personally consider the MH370 negotiations theories unlikely.)
@Victor Iannello
Yes, I should have placed an ‘Errant Nonsense Ahead’ caution on that post. It looks like The Daily Star Sunday (Northern and Shell Media, UK) first ‘broke’ that story, it was then picked up by The Sun (News UK) and has then wormed its way into other News publications. I expect that The Australian will be running in due course.
Here is my summary of the individual cases:
pilot suicides:
• 1976-09-26: pilot crashed into building, instant death
• 1977-01-05: pilot “set full power on both engines and aimed at the Connellan complex before plunging the aircraft into the centre of the building”, instant death
• 1979-08-22: pilot crashed into a Bogota suburb, presumably instant death
• 1980-06-01: pilot crashed into building, instant death
• 1982-02-09 JAL350: pilot engaged thrust-reversers in flight, intending a crash with instant death
• 1982-09-15: pilot crashed into airport, presumably instant death
• 1994-07-13: pilot circled the aircraft until it ran out of fuel and crashed
• 1994-08-21 RAM630: pilot entered a steep dive, crashed into mountain range, instant death
• 1994-09-12: pilot tried to fly into walls of the White House (instant death), but hit the South Lawn and died on impact
• 1997-04-02: pilot crashed into mountain, instant death
• 1997-12-19 SLK185: pilot executed rapid, nearly vertical dive, instant death
• 1999-10-11 Air Botswana incident: pilot negotiated until fuel exhaustion, subsequently crashing into 2 planes on tarmac, instant death
• 1999-10-31 MSR990: 1st officer pushed nose down, steep descent, crash into ocean, instant death
• 2002-01-05: pilot crashed into skyskraper, instant death
• 2010-02-18: pilot crashed into building, instant death
• 2013-11-29 LAM470: “aircraft was cruising at FL380 when it began to lose altitude abruptly. The aircraft descended rapidly at a rate of ~100 ft/s. After 6 minutes of losing altitude […] the aircraft crashed into Bwabwata National Park and exploded. The explosion completely destroyed the aircraft and killed all 33 people onboard instantly.”
• 2015-03-24 GWI18G: first officer crashed into mountain, instant death
hijackings:
• 1964-05-07 PCA773: passenger shot both pilots before shooting himself, instant death
• 1974-02-22 DAL523: man intended to crash into the White House (instant death), unsuccessful, committed suicide by shooting himself in the head
• 1987-12-07 PSA1771: “plane crashed into the hillside of a cattle ranch, exploding on impact. The plane was estimated to have crashed slightly faster than the speed of sound, at around 770 mph, disintegrating instantly. The aircraft experienced a deceleration of 5000 G.”
• 1994-04-07 FDX705: crew member tried to crash the plane into the FedEx headquarters
• 1994-12-24 AF8969: hijackers intended to blow up the plane over the Eiffel Tower
• 2001-09-11 AA11: hijackers crashed into WTC
• 2001-09-11 UA175: hijackers crashed into WTC
• 2001-09-11 AA77: hijackers crashed into Pentagon
• 2001-09-11 UA93: plane crashed into field at 563 mph
[Comments here are closed. Please continue the discussion under the new post.]