Last month, I published an article in New York magazine about a secret Malaysian police report which included details of a simulated flight into the southern Indian Ocean. As Victor Iannello revealed in a comment earlier today, that information came from French journalist Florence de Changy, who had come into possession of the full police report but only shared a portion of it with me.
I have not seen the full report, but would very much like to, because I would like to form my own judgement of what they mean, and I think everyone who is interested in trying to figure out what happened to the missing plane, including the next of kin, are entitled to the same. Some people who have read the full reports have suggested that they give the impression that the recovered simulator files do not in context seem all that incriminating. Other people who have seen the full report have told me that the report contains material that makes it hard to doubt that Zaharie is the culprit. Of course, it’s impossible to rely on someone else’s say-so. We need to see the full report.
The reason I am writing this post now is that earlier today Florence published an article in Le Monde in which she describes having the full report as well as another, 65-page secret document on the same topic. Meanwhile, another French newspaper, Liberation, has also published an article indicating that they, too, have a copy of the report. And private correspondence between myself and a producer at the television network “France 2” indicates that he has as well.
Meanwhile, I know that independent investigators here in the US have the documents as well.
At this point, the secret documents are not very secret. Someone within the investigation has been leaking them like crazy, obviously with the intention that their contents reach the public. My understanding is that this source has placed no restrictions on their use. So journalists and independent investigators who have copies of these documents need to do their duty and release them — somehow, anyhow. Some people that I’ve begged and implored to do so have said that they fear legal ramifiations. Well, if it’s illegal for you to have these documents, then you’ve already broken the law. Use Wikileaks or another similar service to unburden yourself.
Free the data!
UPDATE 8/14/16: Apparently Blaine Alan Gibson has the document, too, according to a rant he post on Facebook. He reveals that the entire set of documents is 1,000 pages long.
Regarding the post by CliffG at 11.28 pm on Aug 22:
There must be enough people here who are fluent in French. Anyway, here is the extract from Florence de Changy’s book translated via Google Translate. Somewhat mangled, but quite similar to what was written in Jeff’s “The Plane which wasn’t there”:
Both Ukrainians come together in the last minutes of boarding, far more energetic than their travel companions. With their physical marine commandos, mussels in black T-shirts, they each carry a large cabin bag, they waltzed on the treadmill scanner, a gesture causes. Embedded between all the passengers on this flight, whether to try to appoint two hijackers, the Ukrainians would be the only ones present in the cliches attributes: age, physical condition, appearance, attitudes ..
Thanks for your detailed comments. The aero model crashed at 0:12:48, Exner’s simulation at 0:12:52, 4 seconds later. I didn’t worry about that difference, shown in the “Alt” graphic.
Yes, I didn’t realize that, but it is correct. Thanks for pointing it out.
When the second engine quits, the flight control system goes into secondary mode, causing the AP to disconnect. In secondary mode the envelope protections, including the bank limiter, are lost.
Agreed (except for the bank limiter). I have asked ALSM how the APU was controlled during the simulations but unfortunately he doesn’t know or doesn’t want to answer that type of question.
The cL/cD of about 17.5 is consistent with the FCOM data for idle descent, and is quite normal for similar airplane types. You can change the cD value and you’ll see that the model is not very sensitive to L/D. Incidentally, the cD includes the induced drag, and doesn’t change because cL is constant.
Inmarsat engineers have pointed out in their paper in the Journal of Navigation that the BFO value of -2 Hz at 0:19:37 is invalid and should be discarded. From an aerodynamic and stability/control standpoint I cannot imagine a condition in which an airplane rate of descent increases from 4000 fpm to 14000 fpm in 8 seconds without pilot input or control runaway. If recent Boeing simulations have indeed produced different results than the earlier ones, then what changes have been introduced in the simulations?
The pilot in Exner’s simulation applied 1 degree of right rudder trim, to make sure that the simulator would turn and bank. That ruddere trim determined the direction of the turn, rather than which engine failed first. Without such deliberate mistrim it is unlikely that the airplane would go to such extreme bank angles.
Agreed
You can trace the changes I made by comparing the edited version to the original document, still available at http://tinyurl.com/q2qpxnb.
@Richard Cole
Apologies for not replying earlier.
The DSTG Bayesian analysis produced a probability density function. The most probable path(s) crosses the 7th arc at local sunrise.
@Gysbreght, Thanks for the reminder about the Ashton et al. Journal of Navigation paper. You’ve prompted me to go back and take another look at it. A couple of noteworthy sections:
And:
Also worth noting is this statement from page 94 of the latest edition of “Bayesian Methods,” just released in book form by Springer last month:
I would be extremely interested to know why the ATSB decided these data points should be taken at face value, after all.
And btw, I find it quite striking that Inmarsat’s very first rough back-of-the-envelope calculation of where the plane might be matches up almost exactly with where the DSTG’s elaborate calculations ultimately put it (page 8), after having wandered up and down the 7th arc as various mitigating circumstances were added and subtracted. This could either be a happy coincidence or an illustration of the power of first impressions, when the fundamental facts of the case remain clear.
@Jeff Wise
The simple BTO-only solutions (constant speed, true/GC course) always placed the search area roughly around 90E. Figure 6 of the JoN paper is one those. It is when other data types are included, or more complex paths, that more northerly destinations are indicated. The JoN paper’s use of BFO data is an example. That required a more complex path and was finally dropped.
The function of the DSTG paper (at the core is a BTO-only solution) was to establish the width of a plausible search area. The use of the aborted satellite phone BFO data (or not) is one of the few variables.
@CliffG, Here’s my attempt to translate Florence’s description of the Ukrainians:
@Richard Cole, Yes, I suppose a strong case could be made that it was only luck that the final answer wound up looking so much like the first one.
BTW, thanks once again for tweeting your latest chart of the seabed search progress. Would you mind tweeting or emailing an image that shows the whole area? I’m particularly interested in getting a big-picture overview of what has now been searched.
Gysbreght: You stated:
“I noted several errors which I have corrected in the version below, and I have added some calculations based on those data.”
What do you mean by errors? The transcript was not the result of calculations. It was a list of what was observed on the instruments during a simulation. How can you say the observations were in error? You don’t have the video do you? Typos? Please explain.
All: As for the right turns in our simulations: That has been covered many times. The bottom line is that the simulator can not be used to determine the direction of the turn MH370 made. After FE, any plane will tend turn in the direction of the rudder trim bias, if any. We used 0.5 deg right rudder trim because that was typical for the planes Paul flew. But in a fleet of many B777s, there will be some that require a little left rudder trim at cruise, and some that need a little right trim. Apparently, 9M-MRO was trimmed with a slight left rudder trim at cruise. This is irrelevant to the simulator trials. We know MH370 turned left from the BTO data and the geometry of the problem, not the simulator.
Regarding JON Paper:
There appears to be some confusion about how the Inmarsat 400 kts speed was chosen. Isat did not chose a priori to use 400 kts in their model. An error minimization software routine chose 400 kts. Isat ran many trails using a range of speeds and altitudes. They checked the BTO and BFO errors for the set of trials and found that the BTO/BFO error was lowest for the path they published. From a pure math POV, it may make some sense to accept that answer. But we know that the BFO in particular is subject to several small error sources, including some small bias drift errors, and we know 400 kts is not a normal cruise speed. So the Isat path was not taken by most experts as being the most likely path. Two years later, with the benefit of debris finds and negative search results further south, it is reasonable to ask: Could the error minimization technique used by Isat be telling us something previously deemed wrong or unimportant? If we knew that the BFO data was accurate to 0.1 Hz, then the Isat method could be trusted more. Unfortunately, we know it is not that accurate, so I believe it is a coincidence that the Isat path points to a POI more consistent with what some debris drift models suggest.
@airlandseaman, Thanks for this clarification. Could you shed light on the question of why the ATSB is now deeming valid the 0:19 BFO values that Inmarsat and the DST said should be discarded?
@all @ROB
The debris-hunters seem to find a lot of possible debris:
http://video.foxnews.com/v/5094247534001/possible-flight-mh370-wreckage-found-on-south-africas-coast/?#sp=show-clips
Another piece is found near Durban:
https://www.dropbox.com/s/cg4crz2kkf5u2ci/Durban22-8.jpg?dl=0
Of special interest to me are the same knife-like cuts in this panel that are also visible in a lot of other pieces.
A special for @ROB:
https://www.dropbox.com/s/on4mtgk1qt7wrhd/MH370%20sunrise.jpg?dl=0
@Jeff Wise
> I suppose a strong case could be made that it was only luck that the final answer wound up looking so much like the first one..
Not sure I was saying that…
I will work on the wider view map – needs some updates to the various boundaries.
Thanks Richard!
Yes, there were a few typo’s. Values read from the instruments with one digit changed by +/- 1, or things like that. But the error that came to light just now is different. Apparently the data came from two files with different time scales. It seems that when these two files were stitched together, a timing error occurred at around 0:12:30. The ‘patch’ inserted a short climb in the descent. In any case the altitude trace in your file requires changes of rate-of-descent that are physically impossible. Removing the rows that were obviously erroneous, required the headings in the remaining rows to be changed to their reciprocals for continuity.
ALSM, as you know very well, what you write is not true. Since that has been covered more than once, I will not cover it again, but I can if you insist.
@Ge Rijn
That was very kind of you to post the sunrise for me. Do you know what the time tag was? It was before 00:19UTC, I’m thinking.
@Ge Rijn
Interesting comment from Peter Foley, concerning the outboard flap. I said I thought the flaps were down. He says the flap appears to have been retracted inside the wing. If Boeing back this up, it might explain why no RH inboard flap parts have turned up, also nothing from the LH wing.
Here is a revised scenario for your consideration: 2nd engine flameout at 00:17:30UTC (approx) RAT manually deployed shortly beforehand. Pilot begins manually controlled descent. APU come on line a minute later, and runs for long enough to reboot the SDU, but flames out shortly after.
When the RAT is the sole source of hydraulic power, flaperon outboard actuators are operating, only.
Pilot comes in for a high speed ditching, but the RH wingtip or RH engine pod catches the swell (which is running from right to left) He attempts to level the aircraft, causing the RH flaperon to deflect down. The RH wing continues to dig in, ripping off the the flaperon and forcing off the seal panels, and tearing off the RH outboard flap. The inboard flap stays on the wing.
Because the RH flaperon inboard actuator is inoperative, the flaperon is subjected to torsion, and the inboard hinge fails before the outboard hinge. There is a crease or buckle in the flaperon leading edge, close to the outboard actuator attachment point, which could have been caused by torsion of the flaperon body.
Gysbreght: You are correct that the the observations came from one long file (17:25 total) and a shorter one at the end (00:57). I don’t recall what happened to cause the GoPro to turn off briefly. But the time between the last frame in the first file and the first frame in the second file was only 2 seconds. In the transcription, I did neglect to add the 2 seconds to the times in the second file, but that hardly has any significance to the simulation results or the transcribed record. The times were off (2 seconds slow) after the file change. That is no reason to scrap 8 records as you did. Just add 2 seconds to each of the last 57 seconds of the observations and stop with the nonsense.
Re: “ALSM, as you know very well, what you write is not true.” No, what you wrote is a lie. Contrary to your idiotic claim, I do know what I wrote is true. I was there. You were not. It consistently turned right after FE when the trim was set to 0.5 deg right. Your personal attacks…accusing me of lying and knowing it… are despicable.
@ROB
I thougth @David told only the outboard actuator of the right wing flaperon worked under RAT. To me this seems an inbuilt redundancy to keep the plane level. But I don’t know for sure. There must be a good reason though to keep this flaperon working also under RAT. The only reason I can imagine is to keep the plane level under some kind of ultimate ‘auto pilot’ restraint in all circumstances as good as possible.
Expert anwsers on this welcomed offcourse.
The sunrise-graphic I stole from following twitterer I think (but its gone now on his account so I’m not sure).
Very interesting twitterer IMO.
Indeed the graphic shows the situation around 0:19.
From his account also this graphic (thanks mr. Hazzard..).
The left wing spoilers show to be actuated under RAT as is the right wing flaperon.
How this is all possible I don’t know but it seems he knows what he is talking about:
https://www.dropbox.com/s/ykq1wm037lk2gyr/MH370ditchingapproach.jpg?dl=0
https://www.dropbox.com/s/2arvd0aznqbogv5/MH370ditching.jpg?dl=0
@BrockMcEwen
‘Confirmation bias’.. seeing faces in the clouds, patterns where there are none.
The instinctive/intuitive drive to try make sence of every situation as soon as possible.
An in-built survival skill refined by lecturing and expirience often very usefull but failing when a situation shows up never expirienced before.
That’s what happened after MH370 dissappeared.
Back to the confusion..
IMO you try to prove with all means you’ve got, the Inmarsat-data could be wrong.
And all the drift-models (you not name) are not significant.
Your approuch is very usefull. An acadamic critical eye to all ‘confirmation-bias’ regarding the Inmarsat-data, the 7th arc and forward drift analysing.
Realy necessary a voice like this against all odds.
A few points.
Drift analysis is not the same as debris finds. Forward drift analysis can predict (more or less) where debris can be expected to wash ashore before they are found.
Reverse drift analysis on the other hand cannot at all predict where landed debris came from. Simple as that. Your latest work shows this obviously as do all the reverse drift analysis.
The same with Inmarsat-data and possible flight paths. They are not the same.
In this sence the Inmarsat-data are a kind of forward drift analysis.
They predict an abundance of possible flight paths but in a restricted area and (till now) only based on the assumption that after FMT MH370 turned suddenly into a ghost-flight (whatever that strange idea came from IMO).
The Inmarsat-data might be spoofed, faulty or whatever wrong.
But IMO if you leave the credibility of the Inmarsat-data you can as well stop the search effort.
That will only leave reverse drift studies and the ones you attempted with almost the whole IO as a possible crash area.
If this must be it must be but I don’t see any plausible reason yet to discard the Inmarsat data. Going that way might turn out to be the truth but it won’t lead to finding the plane.
I assume that’s not the way to go you prefere also.
@Ge Rijn
Good point. But we’re agreed that the RH flaperon is operative under RAT power, and its the RH flaperon that arrived on Reunion, showing distinct signs of having been dragged through the water (ALSM will disagree)- strongly suggesting the aircraft was under pilot control, and the RAT was powering the hydraulics, when the aircraft ditched.
Then a useful conclusion might be that the pilot hadn’t planned to glide as far as possible after fuel exhaustion but instead, to ditch as quickly as possible. The 00:19 BFO data could be said to support this, too.
So it’s still possible the wreckage might be somewhere within the 120,000sq km search zone.
@ROB
I forgot to link the twitterer. Interesting source IMO:
https://twitter.com/DearMRHazzard
Hope he doesn’t mind.. It’s all for the good cause..
@GE Rijn
“The left wing spoilers show to be actuated under RAT as is the right wing flaperon.
How this is all possible I don’t know but it seems he knows what he is talking about”
LouVilla = DearMRHazzard
I carried out two scenarios on FSX with the great PMDG 777-200LR Addon (The same Zaharie Ahmad Shah used until 2014). The first one was a level flight approach to the water surface and immediately before ditching i make a move to the left with the yoke. The Right Wing Flaperon extended to the bottom on the maneuver.
The second scenario should simulate a failed ditching maneuver with a banked turn to the left as it happened on Ethiopian Airlines Flight 961 in 1996 near the Comoros Islands.
The Right Wing Flaperon extended to the bottom every time if someone in the Cockpit is moving the yoke to the left.
This maneuver is only possible from the Seat of the PIC because the RAT not supplies power to the Displays and Instruments on the side where the FO sits.
From @Annette’s account: @airlandseaman picture I posted before.
Looks like another cabin-piece like the Rodrigues-piece to me.
Must be frustrating those guys on the ships finding nothing while laymen find piece after piece.
https://twitter.com/aussie500
If just those black boxes would seperate on impact and drift…
Please explain then how it is possible that an airplane descending with 60° bank at 15,000 fpm, suddenly levels off, essentially maintaining altitude with 60° bank during 19 seconds, then again suddenly continues descent at 15,000 fpm.
Then you write:
There are two angles to this: first, the amount of rudder trim and, second, why was it applied.
1). I recall that in a first post on this subject you wrote that at first 2° was tried and found to be too much, therefore reduced to 1° which was used in all but one of the tests in the simulator. In several subsequent tests you mentioned 1°, and only recently changed to 0.5°. The EXCEL file we are discussing states 1° right rudder trim.
That same file also states that when the right engine failed, the trim changed from 1° right to 2° left. In other words, the thrust asymmetry resulting from one engine inoperative and the remaining engine at maximum cruise thrust corresponds to 3° of rudder trim. Therefore 1° of rudder trim is quite significant and is more than what would correct small asymmetries that may typically be found on B777’s in service.
I also drew your attention to the report on the accident of AirAsia QZ8501, an A320 that crashed in the Java Sea on 28-12-2014, from which I quote:
2. Why was rudder trim applied at all? B777’s in normal cruise are controlled by the autopilot and do not require lateral trim. The only reason to trim would be to ensure that the airplane is in trim in case the autopilot disconnects. On 15 november 2015 you wrote: “… we used 1 degree right trim in cruse (sic) because in Paul’s experience, all the 777s are slightly “bent”, and most require about 1 degree rt rudder trim in cruise. ” How does Paul know that? Either it is standard practice to trim a 777 in cruise laterally, or it is not. If it is standard practice, then the pilot of MH370 would have trimmed out any asymmetry early in the cruise, and the airplane would have been in trim when the autopilot disconnected. If trim is not applied routinely, the Paul would not know that “most require about 1 degree rt rudder trim in cruise”.
I do not blame you for trusting your pilot buddy, but I suspect that you have been misled.
@LouVilla @airlandseaman
Ha, I saw your name here before.. Just found your twitter few days ago, interesting.
If I understand you well only the PIC would have some control under RAT to manipulate the right wing flaperon and the left wing spoilers?
I think this could be quite interesting for it could mean a PIC (in the left seat) had the means to level out the plane after an ’emergency descent’ under only RAT power.
Is this been simulated in the simulator-tests @airlandseaman and others have done?
@GeRijn
If only the RH flaperon was powered by the RAT, you couldn’t roll right (unless you had LH spoiler) because the flaperon does not deflect up as much as it deflects down. You would have no control authority. So It seems both flaperons must be operative with the RAT?
Do a thought experiment.
@Ge Rijn
I meant RH spoiler there
@Ge Rijn, An interesting facet of the debris collected so far is that a disproportion percentage of it comes from the wings and control surfaces, rather than the fuselage.
I wonder if this might be due in part to the fact that the 777 fuselage is made of aluminum, while the wings etc are made from composites (with a fair bit of honeycomb, it would seem).
@ROB
I’m not a pilot so our dear blog-host or any other pilot will correct me if I’m wrong for sure.
Just my basic aircraft-education..
If only the right flaperon is available you have means to control the right wing up and down movements a bit. More up than down.
But if also left wing spoilers are available and work together with the right wing flaperon you are able to decrease the lift on the left wing enforcing the effect of the right wing flaperon.
What @LouVilla suggests in his graphics (IMO); when the right wing flaperon goes down the left wing spoilers go up reducing lift on that wing and keeping the plane level.
Only thing that at least must be confirmed is that some left wing spoilers are also actuated under RAT power.
@Ge Rijn, I find it hard to imagine an aircraft engineer devising a system where controls would be asymmetric, e.g. pilot could command a roll to the left but not the right. Essentially he’d be devising a flight mode where a crash would be all but guaranteed.
@PatM: appreciated.
@JS: profuse apologies – please link me to your most recent/pertinent request for feedback, and I will happily digest and respond.
@Ge Rijn: thanks for your support. Re: “Going that way might turn out to be the truth but it won’t lead to finding the plane.
I assume that’s not the way to go you prefere also”: actually, if I had to choose, I’d much rather find the truth than the plane. I can’t speak for the next of kin, but I strongly suspect that would be their preference as well.
Furthermore, while I agree with you wholeheartedly that the debris by itself can’t pinpoint seabed wreckage, releasing ourselves from the Inmarsat straight jacket may well lead us straight to it, if other evidence previously dismissed (acoustic recordings, wider debris set, eyewitnesses) is re-examined – this time without bias against.
Gysbreght:
You must have a terrible memory. You state:
“I recall that in a first post on this subject you wrote that at first 2° was tried and found to be too much, therefore reduced to 1° which was used in all but one of the tests in the simulator. In several subsequent tests you mentioned 1°, and only recently changed to 0.5°”
None of that statement is true. It is garbage you made up. We never tried 2 deg and I have not changed the story. Moreover, sim pilots have no way of telling what amount of trim would be “correct”. This was a simulator, not a real aircraft. Real aircraft have some biases built in. It is not intentional, of course, but it is unavoidable in real life builds. Wings do not get bolted on exactly the same on every aircraft, etc. Simulators do not have built in bias as far as I know. So your fabricated memory of what was done makes no sense at all.
In real aircraft, it is always best practice to trim the rudder and aileron control neutral positions. The purpose is to minimize fuel consumption (minimize drag) when the AP is engaged. And when the AP is not engaged, trimming the rudder eliminates the constant left or right rudder pressure a pilot must exert to keep the plane flying straight. The AP will work fine whether the plane is trimmed or not, but the fuel consumption is minimized when the pilot manually trims the aircraft.
The amount of rudder trim we used was 0.5 degrees right for all our sim’s except one. It was an arbitrary choice based on Paul’s 7 years of experience flying left seat in 777s. (If I rounded off to 1 deg in some statement somewhere, get over it.) For the last simulation, with both engines flaming out at the same time, we set the rudder trim to zero in order to test our theory that the plane would tend to turn post FE in the direction of the cruise trim position. When the trim was set to zero, the plane did start to turn eventually, but it took much longer. Unfortunately, we ran out of sim time before we had a chance to verify with a left trim simulation. But after the simulations, I discussed our simulation with a number of 747 and 777 pilots and they all confirmed what we suspected based on the simulator. I also conducted several experiments in a SEL airplane confirming this.
Lost in this worthless discussion is the much more important finding that: The sequence of engine flame outs has no effect on the post FE turn direction. We expected going in that the plane would turn rather rapidly toward the last engine out. And it would, if not for the TAC. But the TAC acts very fast after any change in asymmetric engine thrust. The TAC automatically and rapidly trims the rudder about 3-4 deg to compensate for the first engine out, and it returns to the manually set rudder trim position after the second engine flames out.
@GE Rijn
“If I understand you well only the PIC would have some control under RAT to manipulate the right wing flaperon and the left wing spoilers?”
This is how it looks like when both engines are out of fuel and the RAT just deployed. I´m not sure if the FO is able to make inputs to the yoke. He has no displays on his side. Therefore, i´m assuming he can´t.
http://www.bilder-upload.eu/upload/231b9a-1471979395.jpg
“If only the RH flaperon was powered by the RAT, you couldn’t roll right (unless you had LH spoiler) because the flaperon does not deflect up as much as it deflects down. You would have no control authority. So It seems both flaperons must be operative with the RAT?”
Both Flaperons are fully operative with the RAT…..
Turn to the left :
http://www.bilder-upload.eu/upload/f037f8-1471980133.jpg
Turn to the right :
http://www.bilder-upload.eu/upload/941a50-1471980181.jpg
Off topic
Just thought of sharing this to loosen things up
https://youtu.be/K6S4O-VtZBI
https://youtu.be/cqZc7ZQURMs
https://youtu.be/x-Yi762sQTo
3 of my all time favourites. Hope, I am not kicked out 😀
@LouVilla
‘Both flaperons are fully operative with the RAT’.
This would be important information IMO.
Is this only in FSX or also in reality?
Ge Rijn
This is in FSX. I think it´s realistic. Real Pilots and Boeing created this virtual PMDG aircraft.I can´t imagine an real B777 Cpt. like Zaharie Shah flew this aircraft if it was unrealistic to it´s real counterpart.
@airlandseaman:
If you ignore most of what I wrote then further arguing is pointless. Let readers make their own judgment. Did you read your own EXCEL file? On the single test with rudder trim neutral you once had a different story. You don’t remember the “jerk” that rocked the cab? Do you really know the inputs that were made to the simulator behind your back?
You wrote at 3:11 PM: “In real aircraft, it is always best practice to trim the rudder and aileron control neutral positions. The purpose is to minimize fuel consumption (minimize drag) when the AP is engaged. And when the AP is not engaged, trimming the rudder eliminates the constant left or right rudder pressure a pilot must exert to keep the plane flying straight. The AP will work fine whether the plane is trimmed or not, but the fuel consumption is minimized when the pilot manually trims the aircraft. ”
That illustrates how little you understand of “real aircraft”. Lateral equilibrium requires the rudder in a certain position. It makes absolutely no difference to the fuel consumption whether that rudder position is commanded by the autopilot or the trim setting.
I asked you how the APU was controlled in the simulations. Please reply to that question.
Gysbreght:
You are right. It is pointless trying to communicate with someone as wrong and bull headed as you are.
Everyone else:
What I stated about the economics of trimming an aircraft is correct. Gysbreght is clueless. At UAL, pilots are trained to trim planes to minimize drag and thus minimize the operating cost. It is a fact.
The engineers and pilots among you will appreciate the fact that an aircraft can fly straight and level with almost any combination of cross controls (e.g., right aileron and left rudder), provided you add enough power to compensate for the increased drag. Minimizing drag is not something the B777 AP is capable of doing (although it would be a nice thing to have, and possible with today’s technology). It must be done manually in the B777 by manually adjusting the aileron and rudder trim together to minimize the inevitable residual cross control induced drag.
BTW…I routinely fly towplanes. After a glider releases, it is common practice to descend and land as quickly as practicable. If near the airport, we do that using extreme cross controls (full left rudder and hard over right ailerons, while keeping the power at a high setting. This results in the towplane rolling over on the left side and spiraling down very fast without shock cooling the engine, which causes the cylinders to crack. Not good. So, trust me. A plane that is poorly trimmed will take more fuel to fly straight and level.
“At UAL, pilots are trained to trim planes to minimize drag and thus minimize the operating cost. It is a fact.”
It is not. The purpose of lateral trim is that the airplane will be in trim when the autopilot disconnects.
@DrBobbyUlich – You said, “I have been investigating all possible combinations of altitude, speed, and navigation method. I can tell you that the fuel is inadequate to reach any point to the southwest of the search area.”
I believe you are basing that on data in the FCOM, which I believe is conservative. Therefore, your PDA of -0.4% is a viable path. The ATSB update of 3 December 2015 said,
“Boeing analysed the aircraft performance for various scenarios ranging from MRC to maximum cruise thrust limit (MCRT). The starting time for the analysis was at 18:28 and the end time was the flame-out at 00:17 (i.e. 2 minutes prior to the 7th arc). This gave a time interval of 5.8 hours.
The Boeing analysis gave a series of ranges and time intervals for different cruise altitudes. It was noted that a constant altitude of FL350 or higher gave sufficient range to reach the region on the arc corresponding to the DST Group analysis. Applying the assumption that a series of step- climbs had been performed during cruise, produced a range greater than that required to reach the region of interest on the arc.”
A couple of things that I do not understand is the various analyses use an altitude of FL350 and a speed around LRC. If the altitude in FS is FL400 and speed prior to the last radar was in the neighborhood of 500 kts,why are these dynamics not being evaluated? Can they not make a track that meets the BFO? A conscious pilot who is still trying to get away as fast as possible could continually vary the altitude to achieve maximum efficiency for 500kts. The resulting impact area would be far enough west for all of the debris to drift away from Australia.
Interesting to watch how people behave when they feel they know it all. With a little knowledge they become arrogant. Stick to facts rather than personal attacks. Personal attacks occurs when they run out of making arguments for their statements that they claim.
What I did was because I had a reason to, but I never attacked anyone. I still stick to everything I’ve said including my predictions, and that will lead to the supernatural.
@LaurenH
“applying the assumption that a series of step- climbs had been performed during cruise,”
During the SIO flight it kept the same altitude after the turn from Indonesia till the “end”.
@Lauren H.: “I believe you are basing that on data in the FCOM, which I believe is conservative.”
I can tell you from personal experience that the data a manufacturer puts in the FCOM are not deliberately conservative. Is some cases where several factors cannot be quantified precisely, the data may represent a conservative envelope of possible conditions, such as variations of the holding pattern in the case of holding fuel.
Boeing does not say what the precise definition is. A typical definition would be the average performance expected from a new aircraft at delivery from the factory. That performance can be guaranteed within a certain tolerance for variations between airplanes and measurement accuracy in pre-delivery flight test.
A customer airline does eleborate comparisons between competitive airplane types before choosing a supplier. Therefore a manufacturer cannot afford to be conservative in the published performance data, except perhaps in the case of holding fuel which enters into the reserve fuel carried, but is not counted as actually consumed.
@all
“A conscious pilot who is still trying to get away as fast as possible”
Good catch. May I just add that after Indonesia, in the clear, the speed was slown down and constant. They seem to know this, except for the people here.
@Trond
So you now supernaturally feel you know it all already and you are smoking the winners cigar swinging in the chair while watching sunset over sea? Congrats 😉
@Cliffg
sure that distraction of media/people from Crimea takeover was something coming to mind as the operation was expected to be as quick as possible without any lost lives and it went so in fact, as some kind of blitzkrieg done right, followed by help of locals and negotiating with Ukrainian army units to not go against their own citizens; they even used fake ammo in rare shootings; it sounds all crazy, sure
@falken
I act as if I know about the things I type here. So therefor treat me accordingly. I didnt’t come here to draw attention to myself.
There are at least 3 endings in the SIO. The 1st place is between the pings under the silt.
The 2nd is in the current search area, which was fruitless. I claim nothing to be hidden here. The 3rd is the sim ending, for the fuselage. I may be wrong.
Did they find the Indian military aeroplane that vanished? There are more cases than just MH370.
@Airlandseaman
Re the purpose of trimming…
Lancaster Flight Engineer Joe Nutt recalls the hazards of night raids over Germany: “In order to fool the enemy controllers, the route home was not always a reciprocal of the outbound. Usually, it resulted in a longer way back, so fuel economy was always in the engineer’s mind. With a Lanc lightened of its flare load, and a fair amount of fuel, cruise IAS could be held with relatively low revs and consumption, 2000rpm + 2 boost gave a fuel flow of around 165 gal/hour. The engineer would inch the revs down when the IAS needle went a hairsbreadth over 155kts. Flying for range, the engineer, nav and skipper liaised to get the best out of the aircraft and its remaining fuel. The engineer with his fuel calculations, the nav with his estimated times of arrival, and the skipper trimming so that it flew true, with no deflection showing on the turn and slip, squeezed the last drop of aerodynamic efficiency from the Lanc”. From “Lancaster at War 2”
Happen you’re right about trimming. Some things don’t change.
@Jeff Wise,
You said: “Could you shed light on the question of why the ATSB is now deeming valid the 0:19 BFO values that Inmarsat and the DST said should be discarded?”
I think the original assessment of the 00:19:29 and 00:19:37 BFOs was that they were unexpected and therefore suspect. There is an important distinction in play here between “they are wrong” and “they may not be right.” To my knowledge neither Inmarsat nor ATSB has ever published any technical explanation of why either of those was “wrong” or random.
The 00:19:29 BFO, just like its prior counterparts earlier in this flight at 16:00 and 18:25, appears at face value to be valid and trustworthy. I believe that is the reason why they are now accepted by the ATSB (and presumably also previously accepted by Inmarsat). The same cannot be said, however, of the 00:19:37 BFO, which remains unvalidated (but perhaps not unexplained). Attributing all the BFO deviations from the values ~8 seconds prior to changes in rate of climb is unsatisfactory because of the high vertical accelerations that would be required both between 18:25:27 – 18:25:34 and between 00:19:29 – 00:19:37. Thus the 18:25:34 and 00:19:37 BFOs are still considered suspect. Some time ago I proposed a theory of a “SDU software feature” that appears to explain both values, and I still believe they contain useful information regarding a combination of the aircraft’s turning rate and climb rate. However, my efforts to get ATSB and Inmarsat to confirm or deny that possibility have gone unanswered. I wonder why?
I believe the DSTG report is actually outdated (or obsolete). It presents the “starting point” information they were given (more than a year prior to the report publication) and the results of their analyses. It does not, in my opinion, necessarily represent a current view of the satellite data, rather a historical record of the process they used and the results they obtained. I think it is a good assumption that the ATSB would not publicly adopt a revised position on the reliability of the satellite data without having the concurrence of Inmarsat, who are the experts. I don’t think the DSTG was ever tasked at any point in time to make a judgment on the reliability of the satellite data – then or now. They simply used whatever they were given way back when they started. Then they wrote it up “for the record” more than a year later.
@Gysbreght/airlandseaman,
“That illustrates how little you understand of “real aircraft”. Lateral equilibrium requires the rudder in a certain position. It makes absolutely no difference to the fuel consumption whether that rudder position is commanded by the autopilot or the trim setting.”
The rudder is not commanded by the autopilot during normal flight. Pitch and Roll, yes! Yaw, no except for autoland.
Yaw damping (although under the AP category) does not provide trim input; TAC only adds to what the pilot has input. In reality the only thing that can truly “trim” the rudder is the pilot.
OZ