Almost immediately upon Blaine Alan Gibson’s discovery of the “No Step” debris fragment in Mozambique, questions were raised about the relative scarcity of marine life growing on it. These questions were redoubled after two more finds came to light, one from South Africa and the other from Mozambique, which both looked surprisingly pristine for objects that had been in the water for two years. I explored the issue in a post on this site entitled “Bioforensic Analysis of Suspected MH370 Debris.”
This weekend IG member Richard Godfrey addressed the question in a post on Duncan Steel’s website. “One possible explanation for this obvious difference between the flaperon and the other items,” he wrote, “might be linked to the differing routes taken by the floating debris.”
As a point of reference, I’ve reproduced the current chart from that post (above). Though in reality the currents are not nearly as deterministic as depicted–there is a randomness to the motion of floating objects that causes them to spread out, like a drop of ink in a bucket of water–it does accurately portray the overall movement of things. The black bar represents the area where Godfrey thinks the plane most likely impacted the water, northeast of the current seabed search zone. He points out that to get to the locations where they were found on the coast of Africa, the pieces would have to have either passed around the northern end or the southern end of Madagascar.
In the image below I’ve sketched out what these paths might look like, more or less. The pink oval represents the central gyre seen in the current map above. The yellow line is a hypothetical path proposed by Godfrey that the flaperon might have taken on route to Réunion. The orange line is a hypothetical path that the capsized boat which washed up on Mayotte may have taken during its eight-month drift from northwestern Australia in 2013-2014. I suggest this is a plausible example of a “north route.” The purple line is an even more hypothetical proposal for a “south route” that I just sketched out freehand after watching some drift simulations.
In the first part of his post, Godfrey tackles the question of whether the African debris might have traveled through water too cold to allow the growth of Lepas anatifera, the species of goose barnacle found on the Réunion flaperon:
If floating debris took a path passing slightly further south of Madagascar then it could remain in colder waters (especially between July and October) below 30S, under which circumstance barnacle attachment and growth is contra-indicated. Thus it might be that the three items found on the coast of Africa reached their destinations via such more-southerly routes… The Paindane item (‘676EB’) discovered at around 24S may well show some evidence of marine life, even though it most probably arrived via the southern route past Madagascar, mainly occupying cooler waters… The Mossel Bay find (‘Rolls Royce’) might not be expected to show evidence of marine life because it was discovered at around 34S and may well have spent most of its ocean transport time in the cooler waters below 30S.
To evaluate this idea, I consulted the newly published paper “Endorsing Darwin – Global biogeography of the epipelagic goose barnacles Lepas spp. (Cirripedia, Lepadomorpha) proves cryptic speciation” by Philipp H. Schiffer and Hans-Georg Herbig of Cologne University in Germany (preprint available here). According to this source, Lepas anatifera can be found in waters where the temperature is greater than 15 degrees Celsius. South of this line a sister species, Lepas australis, is found:
To get a sense of where this transition zone occurs, I traced it out on Google Earth and superimposed a surface-temperature chart lifted from Godfrey’s post along with the previously described drift routes.
The southern boundary of anatifera’s range is the red line that passes through the seabed search rectangle:
As is quite readily apparent, all the routes lie entirely within anatifera’s range. Note also that the southern boundary lies well south of the gyre, meaning that anything that drifts beyond it is going to be swept eastward. It’s entirely possible that a piece of debris might have neared Africa and then been swept south into cold water that killed the anatifera, but after that the piece would have been carried back towards Australia. In order to move back west it would have to have first drifted north back into anatifera habitat, where it would have had approximately a year to get re-colonized. Remember, Lepas reach sexual maturity in 60 days and achieve full size in six months to one year. So these pieces should have been carrying a load of biofouling similar to the Réunion flaperons even if their initial population was killed off by the cold.
Godfrey also raises another possibility: that the African pieces are clean because they passed through ocean regions too low in nutrients to permit the growth of marine organisms. To check this idea, I consulted with a NASA website that archives world-wide chlorophyll concentrations, which can be read as a proxy for ecosystem nutrient level. Here I’ve overlayed the same set of drift routes over a nutrient map for March 2014, when the water is near its warmest:
And here are the nutrient levels in September, when the water is near its coldest:
Broadly speaking, there is an area of relatively low nutrient levels in the middle of the SIO that grows and shrinks with the seasons, being biggest when the water is warmer. In the warmer latitudes transient high-nutrient patches can be found, but they are transient in time and space. The southern end of anatifera’s range experiences consistently higher levels of nutrients, as does the ocean between Madagascar and the African mainland.
Godfrey writes:
Although it appears likely that the floating debris from MH370 was carried westwards towards Africa by the Indian Ocean South Equatorial Current through warm waters (i.e. where barnacle attachment and growth is feasible), these waters have relatively low concentrations of chlorophyll in the maps above, and therefore limited amounts of phytoplankton, and this militates against substantial barnacle growth.
The problem with this analysis is that the piece of debris which spent the greatest amount of time in the center of the Indian Ocean, with its low nutrient levels, is the flaperon, which has the greatest accumulation of Lepas, including some which have reached full size. The clean pieces, by contrast, have spent considerable time in the nutrient-rich waters near Madagascar.
Finally, I’d like to address an addendum to Godfrey’s piece by Don Thompson, who writes:
An alternative reason for the Réunion and Rodrigues items being barnacle-encrusted but not the other three might be as follows. The lepas (goose barnacle) colonisation may be a feature of proximity to coastlines inhabited by lepas colonies. Therefore, debris ‘dropped’ into a mid-ocean region (i.e. the crash site) might be expected to be ‘clean’ of lepas barnacles until free-swimming barnacle nauplii, released from reproducing coastal colonies, are encountered.
Again, Thompson has the situation reversed. Lepas are pelagic creatures which are adapted to rafting on the open ocean. Buoys placed far out to sea become heavily settled by them.
UPDATE 4-7-16: There seems to be some confusion about the lifestyle of the Lepas. Unlike some other genera of goose barnacle which can be found living in intertidal zones of the seashore (such as Pollicipes, a delicacy in Spain), those of the genus Lepas are obligate rafters, highly adapted to life floating free in the open ocean. Here’s an excerpt from Barnacles: Structure, function, development and evolution:
@ RetiredF4
Thanks for this. Certainly, the aircraft did not fly to meet any pre-set ping rings. The ping rings are derived from a small set of data that – in my understanding – cannot even be verified. Any possible flight path results from connecting a couple of dots with a line and nobody can say with certainty how the aircraft behaved between the BTO dots.
The discussion is getting weird and annoying.
@RetiredF4:
Thanks for reminding us of that excellent description of the B777 Fly-by-Wire system.
Incidentally, that explains how the airplane in ALSM’s simulator experiment maintained a steady turn at constant airspeed and bankangle in the 5 minutes 35 seconds after the second engine flame-out.
cant refuse, its wonderfull too…
https://www.youtube.com/watch?v=gmOvEwtDycs
and as a coincidence, our eastern friends are mentioning something like multipolar world, these days, and your friendly president told something about it also somewhere, but this quite old post (Mar 7, 2014) I found today mentions such things too, if its not anti-dated or so; because I NEVER watched the damn propaganda bullhorn before Mar 2014 and definitelly never will have something similar in head too without some wake-up
there is the link, excuse me; shuting up again…
https://www.rt.com/op-edge/ukraine-beginning-multipolar-world-430/
@Phil
Re our earlier discussion iro the second logon event:
This is my personal take on it.
Pilot wanted to create the maximum embarrassment for the Malaysian authorities.
His way of ensuring this was to disappear with his plane and passengers, but to avoid it being misconstrued as an accident.
By deliberately setting up the 2nd logon event, he would show the aircraft had flown until fuel exhaustion. The authorities would then be unable to explain it away as an accident.
When an aircraft suffers a catastrophic failure, it almost invariably crashes within an hour. Can you think of any that didn’t?
So he could prevent the authorities from having recourse to the accident theory to explain it away. And by staying silent and not transmitting a message if protest, he would ensure that it would always remain an enigma to hang over them. If he had broadcast a message of protest, the authorities spun it as an isolated, unfortunate political gesture of protest, to draw a line under it.
@ROB, It’s important to understand that airline pilots’ training does not include such technical details as the collection of BTO and BFO metadata by Inmarsat. Indeed, when Mike Exner asked the experts running the advanced flight simulation facility he visited how to reboot the SDU, they had no idea–the idea had never occurred to them. So: no, if Zaharie commandeered the plan with an eye to committing suicide, he did not reboot the SDU in order to embarrass the Malaysian government, or to achieve any other goal, because he wouldn’t have known how to reboot it or what the significance of the reboot would be for accident investigators after the fact.
The only way that anyone has come up with to make the SDU reboot conform with a suicide theory is that Zaharie might have isolated the left AC bus in order to stop the cockpit voice recorder, and this had the unintended consequence of causing the SDU to log off. One issue that I think remains outstanding is whether the subsequent reconnection of the left AC bus would have led the SDU to reconnect with Inmarsat in the way that it did–for instance, as part of the process the Inflight Entertainment system reconnected but other systems did not.
The idea that MH370 represents an attempt to humiliate the Malaysian Government has always had its supporters, but the undeniable fact is that the overwhelming public reaction to its disappearance has been, “How can a plane disappear in this day and age?”, not “Why didn’t the Malaysian Air Force launch fighter jets to intercept it?” Indeed, as Retired F4 has explained, there would be no reason to expect the Malaysian Air Force to be able to intercept, nor indeed to want to. More generally, there was nothing about the disappearance of MH370 per se that was embarrassing to anyone. Certainly the incompetence on display in the aftermath has been dismaying, but it’s hard to imagine a perpetrator saying to himself, “I bet if I turn around the plane at IGARI, MAS OPS will be so asleep at the wheel that they’ll I’m actually over Cambodia when I’m really over Penang, and that will make the Malaysian government look bad.”
@DL: You wrote, “The ping rings are derived from a small set of data that – in my understanding – cannot even be verified.” I don’t know what you mean by this; the BTO data are the most precise and reliable information we have about the last six hours of the plane’s flight. Your impression that they cannot be used to determine how the aircraft flew is inaccurate. And while you may find the discussion weird and annoying, we are attempting to resolve a point that is crucial to understanding the mystery. It is exactly this kind of discussion that this comment thread exists to serve.
@RetiredF4, We may be back to discussing autopilot modes for the umpteenth time, but you’re actually proposing something that, as far as I can recall, has not been suggested before: that the plane flew for the final six hours, or some portion thereof, with the autopilot switched off. I don’t know if this is possible or not, but I think it deserves some consideration. The problem, as I see it, is that in order to reach the kind of area that Richard Godfrey is proposing, near Broken Ridge, the plane would have to gradually undergo a heading shift from, say, 180 to 150 degrees true, or about 5 degrees per hour or 1/12 of a degree per minute. I think it would be impossible for a human pilot to achieve this kind of consistent precision. With the autopilot turned off, is it possible that the plane could fly in this manner? I’m not sure, but I suspect not. Is the autopilot-off condition you’re suggesting different from the uncontrolled state that Mike modelled in the simulator? I’d certainly like to hear people’s thoughts.
@Brock, Thank you for clarifying. Yes, by “straight” I mean all the different kinds of heading modes that are accessible by autopilot, some of which result in slightly curved tracks, but all of which in practice put the plane in the ATSB seabed search area.
@Neils, The key issue is the size of the coincidence required. I’ll try to address this point in greater depth in a future post.
@Middleton, you wrote “if there were a set of legs / waypoints programmed into the FMC (either geographical, or custom: lat/long pairs) with speeds and altitudes associated to each waypoint…” You can’t program different speeds into the autopilot. Victor came up with a clever way to alter the speed by slowly descending, but that still resulted in the plane winding up in an area of the 7th arc where the seabed has been searched.
@Jeffwise
A am having to disagree with you on whether the pilot knew that an SDU reboot would send out a logon request. I firmly believe he would have been aware of it.
The person involved was an experienced airline captain. He would have known about the SATCOM logon sequence before takeoff, and of the SATCOM availability (or otherwise)messages that appear on the cockpit MDF.
I accept that he would have been unaware of the handshake interrogations generated each time SATCOM/ACARS is silent for more than an hour.
@jeffwise
The location of heading / speed changes needed to produce an apparent (!) curved path are (as Dennis also stated) not related to ping ring positions. So IMO the coincidences involved are not very big. For example, to reach S20/S21 a possible path is ISBIX, POLUM, IKASA. Based on the BTO/BFO produced it would turn out as a curved path in the explicit path generation tools.
So what I conclude is that (apparent) curved paths can be produced with and without autopilot engaged. (Thanks @RetiredF4), and with and without pilot input (depending on requirements on the few heading and speed changes).
Jeff,
Re: “…but you’re actually proposing something that, as far as I can recall, has not been suggested before: that the plane flew for the final six hours, or some portion thereof, with the autopilot switched off.”
Frankly I am quite disappointed. How come it was not discussed before? I have strong impression that non-AP modes are “taboo” for understandable reasons. Neither camp wants to hear about them.
Re: “the plane would have to gradually undergo a heading shift from, say, 180 to 150 degrees”.
Haven’t I presented the two notes, where the heading changed from 180 to 150 deg? The main objection from ALSM was the stability without AP. More on the way: this time with VNAV enabled. Niels has also presented his study of curved trajectories.
But if nobody reads these notes, does it make sense to continue?
@Jeffwise
And the fact is that the disappearance of MH370 has indeed been a high embarrassment to the Malaysian government, that is undeniable. It’s clear to me that they guessed his intentions from the very beginning, and ever since that moment, they have been engaged in a desperate damage limitation exercise. It’s the only explanation for their patently obstructive behaviour.
And I never at any time questioned why the Malaysians failed to send up fighters to intercept.
The SDU was disabled for the first hour to stop email and text messages going out. After one hour of depressurization, he no longer needed the video camera to monitor the cabin. He then re-energised the SDU, and disabled the IFE from the overhead switch.
@Jeff, RF4: according to 7BOEING7 (from airliners.net) with the autopilot off, and no controlled inputs from a pilot, the aircraft will bank over, and fly in circles until it runs out of fuel. This is for the B777. A B787, OTOH, if you take your hands off the controls, will continue to fly an “inertial” heading: it will stay pointed in the same direction, but will be affected by cross winds and Coriolis forces, but not be married to any particular heading, magnetic or otherwise. It does this only because of automatic controls, however. My understanding is that modern airliners, unlike a Cessna 172, especially at high altitudes, are inherently unstable, and only maintain trim through constant controlled inputs.
@Gysbreght: After flameout, when the autopilot kicks off, “you end up in secondary mode and per the ops manual, in secondary mode envelope protection is not available.” per the same 7BOEING7:
http://www.airliners.net/aviation-forums/tech_ops/read.main/368400/
Warren Platts,
Re: “B787, OTOH, if you take your hands off the controls, will continue to fly an “inertial” heading: it will stay pointed in the same direction, but will be affected by cross winds and Coriolis forces, but not be married to any particular heading, magnetic or otherwise.”
That is approximately what my two previous notes were about: Coriolis + wind. The modelled trajectory fits BTO and BFO. It terminates around 99E, and the location 19:41 is consistent with Kate’s testimony.
It certainly has been mentioned before, including various references to the description of the flight control system in the FCOM and FCTM.
The uncontrolled state that Mike modelled in the simulator was different in three respects:
1. After dual engine failure the FCS reverts to secondary mode. In that mode the flight control laws are essentially the same as in normal mode, but there are no envelope protections for overspeed, overbank and stall.
2. Mike ensured that the airplane was out-of-trim at autopilot disconnect by setting the rudder trim to 1 degree right. Therefore the airplane started to yaw, turn and roll immediately.
3. There probably was a short interruption of hydraulic power after the second engine flame-out and rundown until the APU had started up and came online.
During that period the Flight Control System (FCS) would have commanded the ailerons to counter the roll induced by the rudder trim, but without hydraulic power the the ailerons did not move and therefore the airplane rolled to 35 degrees of bank. When the APU had come on line about one minute later the ailerons moved to stop the roll and the FCS then maintained the roll and pitch attitudes and airspeed for about five minutes. The airspeed was the selected speed of 200 kIAS shown on the PFD’s in Mike’s screen grabs, which is the maneuvering speed (minimum safe airspeed) at a weight of 174 tonnes.
In RetiredF4’s situation the FCS would maintain the roll and attitude and the pitch attitude for the trimmmed speed . With autothrottle OFF the engines would maintain the thrust commanded by the thrust levers, which is an EPR (Engine Pressure Ratio) for Rolls-Royce engines. As the weight reduced over the hours due to fuel consumption, the thrust required to maintain altitude would become less, and therefore the airplane would slowly gain altitude.
@Oleksander: Yes, when I first saw the very first Inmarsat plot, the 400 kt path screamed “Coriolis” to me. I even created a spreadsheet to calculate this effect. However, I have since learned this would only work for a 787, not a 777.
@Gysbreght: Not sure what you mean by “the FCS would maintain the roll and attitude”, but if you mean that the FCS would keep the plane from banking over with the autopilot off and no inputs from a human pilot, that is incorrect.
Warren Platts,
I posted first results on Coriolis+wind about a year ago. Then a technical in-depth note in early September; then the second revision in early December:
https://www.dropbox.com/s/8vrt72o783262he/TN-CTS-Rev1.1.pdf?dl=0
Yes, it is consistent with the initial ATSB results. I already revamped the model with VNAV engaged (FPA followed by ALT hold), LNAV disengaged, and the trajectory is nearly the same.
But as I said, curved trajectories are taboo: apparently neither camp is interested in further investigation for apparent reasons.
@Warren Platts:
RE: “Not sure what you mean by “the FCS would maintain the roll and attitude””
Please ignore the “and”. It was inadvertently left in when I edited the original wording “roll and pitch attitudes”.
RE: “if you mean that the FCS would keep the plane from banking over with the autopilot off and no inputs from a human pilot, that is incorrect.”
Yes, that’s exactly what I mean, and you are wrong. Turning the yoke commands a roll rate, the more you turn the ypke the higher the roll rate, just as conventional controls. Yoke neutral commands zero roll rate, i.e. the bank angle is constant.
@Niels, ISBIX-POLUM-IKASA produces speeds of 126, 378, and 447 knots, respectively between 19:41 and 21:41. These seem unrealistic to me.
The coincidences may not seem big to you, but if you take the trouble to generate a large set of random routes, as the DSTG has done, you will find out that your initial impression was mistaken.
I will add that a lot of people have spent a lot of time messing around with routes into the SIO, and there are no routes via waypoints that end up with BTO/BFO/autopilot modes anywhere outside the seabed search area.
This kind of route, as a result, requires a suicidal pilot who is flying out over a vast, featureless ocean, in the dark, making random changes in direction and/or speed for no apparent reason. And before you say “because he wanted to throw investigators of the scent,” recall that when Inmarsat unveiled their “new mathematical technique” for deriving position from BTO and BFO metadata, they considered it utterly novel. No airline pilot could have anticipated it.
@Oleksandr, Non-AP modes are not “taboo” because no one wants to consider them, but because they don’t make any sense. If you read through the comments on this site you will hear a lot of assertions that such routes are practical but no demonstrations of actual plausible routes.
I don’t know what the longitudinal stability of a 777 in secondary mode will be, whether it will result in a spiral dive in less than five minutes as in Mike’s simulation, or whether it will stay more or less upright, but I don’t think it’s realistic to think it would enter a stable 1/12 of a degree per minute left bank and maintain it for six hours.
Jeff,
It does not enter 1/12 degree per minute turn ineed – there is no such a mode. The turn is due to Coriolis and wind. Attitude (ATT) hold mode of B777 is a good candidate, for example. Or FPA, ALT etc with LNAV disengaged.
You may see EY440 – it is affected by cross-wind in holding pattern mode. But the problem is that there is a minimum bank angle, which does not suite MH370 case, unless there was electronic glitch, which allowed for zero bank angle (technically possible).
@Oleksandr, Your paper is quite clever but it seems to rely on a flight mode that I’m not sure exists. You say attitude hold (ATT) with the bank set to zero degrees, which would require a “glitch” — that is, an entirely hypothetical condition.
@Warren Plats
“@ Dennis: You’ll like this one:
According to the testimony of 6 Swiss Citizens making a cruise between Perth and Singapore via Jakarta, the following evidences were spotted on March 12 while approaching the Sunda Strait:
1430LT – latitude 6°, longitude 105°, speed 17,7 knots:
life jacket, food trays, papers, pieces of polystyrene, wallets,
1500LT:
a huge white piece of 6 meters long to 2,5 meters wide with other debris,”
http://www.baaa-acro.com/2014/archives/mas-777-missing-over-the-gulf-of-thailand-with-239-people-on-board/
Actually not far from the 7th arc, assuming a westerly current flow…”
Wow! I rate swiss people as quite reliable, this is very very interesting piece of info…
@ROB, Please show me a single source which suggests that the disappearance of MH370 itself, rather than the govt’s inept response, has been perceived as a cause of embarrassment to Malaysia.
Also, your last paragraph describing the SDU and IFE is both inaccurate and nonsensical.
@StevanG: strange nevertheless. I mean if you were in a sailboat going past life jackets, food trays and wallets, wouldn’t you at least stop to pick up the wallets?
@Gysbreght: Yes, one would expect from the comfort of an armchair, or even a rated simulator, that neutral yoke would yield a zero bank rate. However, in real life, that is not the case. You cannot shut off the autopilot, take your hands off the wheel, walk away, and expect the airplane to fly straight. A 787 will fly straight, but a 777 absolutely will not. Cf. the link to a.net I posted above. “Take your hands off the controls in level flight in a simulator the airplane will stay in level flight, the real airplane won’t.” 7BOEING7 is an actual 777 pilot whose job is to take new planes out for test flights before they are handed over to the airlines. He’s tried it in person. The aircraft will not fly straight in autopilot off, hands off mode.
@Oleksander; Jeff: the reason autopilot off scenarios are not talked about much (anymore) is because they are IMPOSSIBLE, or at least certainly not consistent at all with the BTO lines of position. See above.
Actually, I think when this was discussed two years ago, there was some uncertainty if the airplane would simply fly in circles, or if the overbank would overcompensate causing the plane to bank to the other side, etc. I suppose it could be possible that one could get a sinuous flight pattern where the airplane is constantly banking back and forth at 35 degrees each bank. So the airplane would be going at a constant cruising airspeed, but it’s average ground speed would be slower. In such a case, I guess it would still be subject to crosswind and Coriolis perturbations…
@Warren Platts:
I didn’t say the airplane flies straight. Neutral yoke commands zero roll rate, and the computers will command the ailerons and spoilers to achieve that within system capabilities. Whether that results in flying ‘straight’ depends on the wind variation, turbulence and perhaps coriolis effects.
Things you read on the internet should be treated with caution.
@Warren, You simplified everything nicely and then unsimplified it. I don’t think if a plane is flying without an autopilot or anything to level the wings, and its bank angle is increasing, there is any reason the bank angle would spontaneously decrease. I think Mike’s experience with the simulator is correct: the bank angle will increase until the plane is in a spiral dive.
@jeffwise
Jeff, I have ISBIX between the 19:41 and 20:41 positions, their coordinates being:
N7E93.3
S1.5E94.1
So I don’t know how you came to your numbers for speed?
The 19:41 position is rather northerly as this was a calculation for the path VOCX, BEDAX, ISBIX, POLUM, IKASA and the VOCX waypoint may not be the right position (in terms of longitude) at about 19:00.
@Niels, According to my measurement made in Google Earth, your 19:41 and 20:41 positions are 511 nm apart. That’s better than 126 but still doesn’t gybe well with 378 and 447 nm during the subsequent two hours.
@Warren
I know, but they are swiss people, they don’t touch anything until it’s not theirs 😉
@ Jeff, You are no doubt correct. If the aircraft was not in secondary mode, the overbank protection would presumably kick in(but not after a dual flameout). However, with the autopilot off, there would be nothing to keep the nose up, so a spiral dive would happen, as you and Mike say. A “sinuous” flight mode is no doubt baseless speculation–just me thinking out loud.
@Gysbreght: I think that in a neutral yoke position, the ailerons/flaperons will simply be in a neutral position. If they are actively moved to compensate, the IRL observed banking when the controls are let go would not occur. Yes, one should treat internet lore with caution. lol!
@JEFF
“your last paragraph on the SDU was nonsensical and inaccurate”
If I may elucidate one more time.
He began depressurizing the plane straight after the signoff. To prevent emails and text messages going out he could have simply punched the IFE/seat power switch off (overhead panel) but this would have also turned off the video camera, needed to monitor the other side of the door during the first hour. He needed to be aware of possible intervention while the depressurization was taking effect.
So to get round this, he de-energised the SDU rather than use the IFE/seat power switch.
After one hour, he no longer needed to monitor the cabin (for rather obvious reasons) and was able to restore normal electrical power, re-energizing the SDU in the process.
He switched the IFE/seat power switch off, but not before the IFE logon request had been transmitted.
@StevanG
Yes, I saw that when WarrenP first posted it. I certainly found it interesting, and had not seen it previously. I avoided making comments because of the “pet theory” feedback you tend to get when you comment on something that supports what you said previously. I was surprised that there was virtually no other comments on it. I did find it odd that there was no mention of any attempt to retrieve anything, but then again if they were in a sailboat, that platform is not well-suited to retrieving things from the water.
@Oleksandr
I did read your report earlier, and thought it was extremely well done. I try to read all the serious analytics that come to my attention. It takes a lot of time, obviously, especially a report as complex as yours. I also get a lot of requests relative to path feasibility from people who are not mathematically inclined. I am sure you get your share of such requests as well. I try to respond to them.
My position remains that there is a very broad range of terminal possibilities that can be obtained by modeling the ISAT data with different flight dynamic assumptions. As a predictive tool, the ISAT data is by no means deterministic. You went the “extra mile” in my opinion by reconciling your terminus with plausible “cockpit” inputs. I stop at feasibility, and take my lumps relative to the detailed flight mechanics.
@Warren Platts:
“I think that in a neutral yoke position, the ailerons/flaperons will simply be in a neutral position. If they are actively moved to compensate, …”
Yes, I understand that you are thinking that, but it is not correct in a fly-by-wire airplane.
Not sure what you mean by “IRL observed banking”.
@jeffwise:
“I think Mike’s experience with the simulator is correct: the bank angle will increase until the plane is in a spiral dive.”
In the one simulation that Mike has selectively made available to us, the bank angle increased to 35 degrees within the first minute (in that first minute no bank angle values were read from the GOPRO video) and remained constant in the following five minutes. Only then (5 min 35 sec after 2nd flame-out) did the airplane suddenly enter into a spiral dive, but unfortunately Mike is unable or unwilling to explain that sudden change of airplane behavior. Perhaps the pilot shut down the APU?
Gysbreght: Re “…unfortunately Mike is unable or unwilling to explain that sudden change of airplane…”
Did you really expect me to take this obvious click bait! Well, I will just add this: I was in the simulator for 4 hours. No two sim’s unfolded the same. However, all had a steep spiral descent. You can take it from there.
@jeffwise,
Furthermore, ATSB, MAS, Boeing and others have indepently tested the dual flameout scenario in various simulators, and no one has reported a spiral dive ending in 90° bank and Mach 1. In a reply to a former poster on this blog, the ATSB has explicitly denied having encountered that in their tests.
@ALSM
“Did you really expect me to take this obvious click bait!”
No I don’t. You always ignore inconvenient questions. There are many things you are not telling us.
Jeff,
The scenario I studied was roughly based on my previous discussions with Gysbreght, which is briefly summarised by him above:
“With autothrottle OFF the engines would maintain the thrust commanded by the thrust levers, which is an EPR (Engine Pressure Ratio) for Rolls-Royce engines. As the weight reduced over the hours due to fuel consumption, the thrust required to maintain altitude would become less, and therefore the airplane would slowly gain altitude.”
In my understanding it is possible to control functionality of the AP via MCP. And there is a plenty of combinations of LNAV and VNAV modes. To ensure flight stability only properly working VNAV is required. The selection of LNAV “TRK HOLD” or “HDG HOLD” are not required for this.
Besides the ATT mode, there is, for example, FPA mode, which automatically switches to “ALT” mode (altitude hold) upon reaching specified altitude. The trajectory is similar to the trajectory depicted in my September’s and December’s papers: only residuals are getting slightly worse, but still acceptable. As a matter of fact, only nearly linear increase of altitude from ~FL100 at 19:00 to ~FL200 by 00:20 is required to catch right wind conditions to fit BTO and BFO.
Warren,
“the reason autopilot off scenarios are not talked about much (anymore) is because they are IMPOSSIBLE”
I am not sure possible or not, since there were a lot of contradictory statements. I am not qualified to comment on this, so I will wait for a final resolution. However, what I am sure is that AP functions can be partially enabled/disabled, and “TRK Hold” or “HDG Hold” LNAV modes are certainly not required to maintain the stability.
Re: “, or at least certainly not consistent at all with the BTO lines of position. See above.”
I am not sure what you tried to say. AP has nothing to do with BTO (and BFO). What did you mean? See what?
Dennis,
Yes, I totally agree that there is a “very broad range of terminal possibilities that can be obtained by modeling the ISAT data with different flight dynamic assumptions”.
And they have to be scanned one by one.
@ Dennis & Stephen: I figured it had to have been dealt with and disposed of here a long time ago as a “pet theory” ;). Did some googling, and cannot easily find any record of any maritime mishaps in that area at that time. Nor can I find any other mention of the “Swiss citizens”, except at the Bureau of Aircraft Accident Archives, which just so happens to be based out of Geneva… However, a couple of Chinese rescue ships were deployed to the Sunda Strait just a few days later. Not sure if they found anything.
@Gysbreght: “Not sure what you mean by ‘IRL observed banking’.”
IRL = “in real life”. The gentleman who posted that information, albeit anonymously, is a highly respected poster at a.net, not trollish at all, and there is no reason to think he is lying. In his job as a 777 pilot, he does not do revenue runs with passengers, so he is free to experiment a bit, like shutting off the autopilot and letting go of the yoke to see what happens. According to his experience in REAL 777’s–not simulators–when you are hand flying (autopilot off) and let go of the yoke, the aircraft eventually banks over. He says he did not let it go all the way to the overbank protection, but presumably it would kick in. When it banks, it turns. The actual mechanics of how that happens I can only speculate on. What is not speculation is THAT it happens.
@Oleksander: related to the above, since an autopilot off, pilot incapacitated B777 will bank over, the nose will tend to point down as well unless opposite rudder is applied, causing a dive, and probably an eventual crash. Even if the aircraft continued until fuel exhaustion, it would simply circle, and therefore, never reach the 7th arc. That is why an autopilot off/pilot incapacitated scenario is pretty much impossible.
Just a quickie. There’s what appears to be a damaged? large twin engine aircraft in this photo from Bing maps, on Gan airstrip in the Maldives. It doesn’t appear in any of the historic images of the airport on google earth. No idea of the date of the image, however it was uploaded to YouTube in April 2014 by a chap who thought he had found something of note.
I can’t figure out if the slides are down, there, or if the wing is damaged.
See what you think, anyone who wants a look:
http://binged.it/1qBZTMS
Hope this link works. I’m trying to find out if there were any incidents involving larger aircraft on this island but not turned anything up so far.
Warren,
Yours:
“since an autopilot off, pilot incapacitated B777 will bank over, the nose will tend to point down as well unless opposite rudder is applied, causing a dive, and probably an eventual crash.”
contradicts to RetiredF4’s post 3:23 AM:
“The aircraft could have flown on a straight or curved path with some random minor heading and altitude deviations due to outside environmental influencies without human input and without autopilot or autothrottle systems being active.”
Could you please both agree?
Either case, what does happen if AP is “On”, but LNAV is disengaged?
@Warren,
“The gentleman who posted that information, albeit anonymously, is a highly respected poster at a.net, not trollish at all, and there is no reason to think he is lying. In his job as a 777 pilot, …”
Apparently you believe that the gentleman’s job is what he claims it is. Your quotes of what he writes makes me doubt that. But then Boeing employs dozens of pilots, and not all are equally conversant with all the ins and outs of complex systems in unusual situations.
Your source is effectively saying that the Boeing engineers who designed the flight control system failed to achieve their objectives. That would mean that the B777 is inferior in that respect to comparable Airbus products. The following is a quote from the Airbus A330 Instructor Handbook:
@Warren
“Did some googling, and cannot easily find any record of any maritime mishaps in that area at that time. Nor can I find any other mention of the “Swiss citizens”, except at the Bureau of Aircraft Accident Archives, which just so happens to be based out of Geneva… However, a couple of Chinese rescue ships were deployed to the Sunda Strait just a few days later. Not sure if they found anything.”
me neither, however I haven’t seen any mention of CI or any other theory on their site so why would they plant false info?!
I’d say these witnesses are the most reliable of all (if the story is actually true and I believe it is), who knows where that debris was by the time chinese ships went to the area
If the ac experienced a serious electrical fire near IGARI, which prompted the pilots to completely depower the plane, and fly utilizing vestigial mechanical flight controls, for the next hour, letting circuits cool and smoke & fumes to dissipate…
And if the pilots risked partially repowering the plane near MEKAR, in preparation for emergency landing near Banda Aceh …
And if that reignited the electrical fire , whose fumes now quickly incapacitated all aboard in under 10 minutes or so …
Such that final approach was not initiated…
Then when would such a fire, energized for the remainder of the flight, have burned out ? Is it possible, that some slowly smoldering fire gradually ate through the e/e bay, eventually interrupting power to the SDU some six hours later? If so, then the second reboot derived from the effects of fire, and not from fuel exhaustion, such that modeled flight paths reaching the 7th arc prior to fuel depletion, could then be considered as far as fuel allowed ( implying a search area further E,S ) ??
To add to Gysbreghts post:
Concerning the roll mode we have to consider aditionally the dampening mode of the roll law. Bank angles with less than 5 ° are cancelled actively out to zero once the force transducers on the control column or the Side Stick senses no control input anymore. Above this threshold value the aircraft would maintain the last bank angle by not allowing any uncommanded roll rate
I take the story about Warrens friend and his experiment with some grain of salt. There are no jobs out there where you fly an 777 with an empty cabin around and one can fly in test pilots terrain. And as we found out in year long discussions on the AF447 desaster, most pilots use those systems without ever knowing how they work in detail.
Zero roll rate does indeed mean exactly that, no roll rate for whatever reason. The roll rate is not measured at the input side, but is from roll rate sensors. Any increase or decrease of bank would trigger such a signal and the flight control computers should actively stop this roll rate immidiately by using the respective flight control surfaces. Therefore it would take some time until the bank angle excceeds the initial 5° threshold value and progresses to the value of the envelope protection. And even then the envelope protection should protect the aircraft from crashing. As long as all the systems work as designed.
@Jeff
You commented to Neils:
“This kind of route, as a result, requires a suicidal pilot who is flying out over a vast, featureless ocean, in the dark, making random changes in direction and/or speed for no apparent reason.”
There is no apparent reason for flying into the current search area either that I have heard.
The changes in speed and heading associated with my suggested route would have to do with simple stalling while waiting for something to happen in KL. Of course, I have encountered problems with that scenario, and I am not embarrassed to articulate them.
The single most puzzling thing to me is something no one else seems to be troubled by. As Don pointed out, the fuel remaining instrumentation is pretty darn accurate. The PIC must have known that fuel exhaustion was imminent (note the correct usage – not eminent or immanent). In this circumstance a non-suicidal (and conscious) pilot would have attempted to communicate the aircraft position in some manner. There is nothing to suggest that such an attempt was made.
Basically the accuracy of the instrumentation coupled with no position report attempt screams at me. I cannot reconcile these facts with a pilot profile or any sensible causality. The deliberate turn West near Igari followed by a deliberate FMT seems impossible to reconcile with a subsequent zombie flight. Perhaps a combination of a deliberate diversion followed by some unanticipated “event(s)” on the aircraft?
Those people advocating a controlled ditch need to account for the lack of a position report. These things seem incompatible to me. What would be the point of a controlled ditch without a survival motive (position report)?
I can only hope that the smart people on this thread would step away from their spreadsheets long enough to think about what might have happened. I fail to see the stigma they seem to associate with doing so.