Malaysian Airlines MH370 contact lost

[MtSpeedDemon]

Quoting MrDuck
"With the (reasonable) scenario most lately discussed..."

Because we don't KNOW:

If the scenario is what actually happened
If the boxes will have data or not (as others mentioned, this itself can be telling)
Heck, there is not a lot we do KNOW for sure other than the plane did not land as expected, millions of people are still fascinated by this and many families are in a lot of pain.

I know there are some other facts but they don't really tell us a whole lot. We reasonably believe this and that but we don't KNOW much at all. I think this is where the fascination comes from.

[HeyIts007]

"I think there is a pretty clear consensus that the track taken by the aircraft was the results of deliberate human inputs by someone familiar with the aircraft systems."

No, it isn't clear at all.

I think it's important to avoid the appearance of conflation of the notion of being on auto pilot or not with sinister or innocent intent. Not suggesting that's occurred, however it's easy for that impression to be formed. We simply don't know for certain. I would not presume the human motivation at this stage and I think to do that to some extent masks the technical process of discovery.

I'm wondering if auto pilot might be the only way to keep an aircraft under some resemblance of controlled flight, if it sustained damage to its pilot control mechanisms?

[aterpster]

H. Harry:

While no-one yet can presume they know what happened on MH370, it is clear from reading this document that investigators are working on the idea that the crew was unconscious for the larger part of the flight. Everything we know about MH370, and everything we've learned from previous accidents, would seem to point to the jet ending its flight after having spent a long time on autopilot."

It that premise is true, then it seems the crew would have been dead after a time rather than unconscious.

[oldoberon]

Originally Posted by Centaurus

'

Remember the Silk Air MI 185 Boeing 737 suicide crash? The investigation found that the two cockpit FDR circuit breakers had been pulled just prior the aircraft being pushed over into its dive. That meant no info after that was available to investigators. The CVR circuit breaker had also been pulled a few minutes earlier which is why the only CVR info that was available was the period before the CB were actuated by someone. If someone in the 777 went to the trouble of switching off the transponder and ACARS, chances are the same person may have tried to hide more vital information by disabling the CVR and FDR by some means (circuit breakers in the 777?)

If they are accessible and they are confirmed as pulled it virtually confirms this was not a fire/accident/decompression scenario especially if done prior to or during the initial deviation from flight path.

Assuming recorder CBs are accessible can any 777 pilot tell us how long with practice it would it take the pilot (LH seat) to pull all comms, CVR & FDR breakers , 10sec, 30sec ? and out of interest could it be done as quickly from RH seat.

[HeyIts007]

Why?....

Seems to me that unconscious pilots would imply an uncontrolled entry into the ocean and unlikely to be a gentle impact. Apparently the autopilot could have remained engaged following the first engine flame-out but would have disengaged after the second engine flamed-out. This would apparently result in a spiral decent into the ocean with no control inputs. Thus a severe impact would have been likely with debris scattered, yet no debris found.

[SpannerInTheWerks]

I suggest this Thread is held until any relevant further information is received?!

Nothing worthwhile is being added at the moment.

[Shadoko]

After a re-read of the last report, it seems that the length of the search zone is principally deduced from the aircraft fuel endurance (figure 20, page 22 (27 of PDF)). The length on the arc between the northern and the southern points is about 2800 kms.
It would have been "fair" if the fuel remaining quantity presumed included in the last ACARS have been given in the report... And what is the accuracy of the fuel quantity measurement?

The BFOs study then came to reduce this length, but "The potential aircraft location, where the derived flight paths cross the 7th arc, is very sensitive to variations in BFO frequency. A 10 Hz variation in the fixed frequency bias can result in the derived flight path at the arc moving 1,000 km." (page 42, 47 of the PDF)...

The general impression, IMHO, is that there is some tautology in the report, like presuming the flight was on AP after the south turn ... and finally finding it was.

Is there any news about the possible faulty line of the table 6 (Downlink Doppler) page 58 (65 in the PDF)? If the values are the same in the three columns (as it could be presumed), then nothing can be deduced about a 5° latitude location change from the registered BFO!

[Hyperveloce]

Here are the results of a MonteCarlo simulation, for constant speed/altitude final leg (from 18:29) like for the analysis B of the report ""Definition of the underwater search areas" available at the bottom of the web page:
http://www.atsb.gov.au/mh370/mh370-defi ... areas.aspx

The simulated ground speeds range from 320 to 520 kts and cover the final ping ring from -21 to -39° in latitude (see last link).

Only the top two deciles in the fitting criteria ( max_t|BFOsim(t)-BFOobs(t)|+µ * max_t|BTOsim(t)-BTOobs(t)| ) are displayed.

The more red the fitter: the hottest spot ranges from 28° to 32° south along the last ping ring (best run at 31°):
https://drive.google.com/file/d/0B3s...it?usp=sharing
https://drive.google.com/file/d/0B3s...it?usp=sharing
https://drive.google.com/file/d/0B3s...it?usp=sharing
It seems very close to the results of the analysis B of the report (see page 27) when the fitting criteria (on the BFO and the BTO/angles of elev.) uses only the last 5 hanshakes (handshakes after 18:29). When the criteria use all the handshakes, the hottest spot shifts 1.5° to the south.

[Ambient Sheep]

I am no pilot (I'm a software engineer, for what it's worth), but I'm pretty certain that hundreds of pages ago it was stated that the 777 FDR & CVR breakers are downstairs in the EE bay, unlike in earlier Boeing aircraft where they are indeed cockpit-accessible.

[HeyIts007]

the 777 FDR & CVR breakers are downstairs in the EE bay, unlike in earlier Boeing aircraft

How does that impact the investigation? Just curious. i.e. whether circuit breakers were deliberately triggered or automatically triggered due to an another technical incident for some reason. The end effect in terms of aircraft performance resulting from the disabled function is surely going to be the same, notwithstanding the effect on the aircraft of other factors that might have triggered them.

[andrasz]

How does that impact the investigation?

I'd say quite significantly. Can someone confirm that the FDR CB is indeed in the EE bay and that the FDR cannot be disabled from the cockpit by pulling other CB-s higher up the power grid ?

My assumption was that whoever initiated the departure from the original flight plan (and there IS consensus this must have been intentional, though most likely the last hours of the flight were purely on autopilot) with an objective to hide all traces would start by pulling the FDR CB, followed by all comms systems. If this is not possible, the FDR might indeed hold some clues on at least the initial sequence of events.

[Ulric]

Are there solutions for constant heading? i.e. rhumb line navigation?

[Pontius Navigator]

Originally Posted by HeyIts007

I'm wondering if auto pilot might be the only way to keep an aircraft under some resemblance of controlled flight, if it sustained damage to its pilot control mechanisms?

I believe this has been done where the physical controls were disabled but the autopilot worked.

I have no idea of the relevance to 777.

[OleOle]

@Shadoko : I agree that the line in Table 6 must be an error.

Regarding the sensitivity to errors in the BFO I feel that between all the computer simulations one tends to forget the basic facts.

Basically the only independent information in the BFO is the line of sight (LOS) speed of the satellite towards the aircraft. It is what richardC10 calls D2_satellite in his paper.

At a given velocity of the sat, D2_satellite is only a function of the angle between sat velocity vector and the line of sight. I.e. all points of equal D2_satellite are located on a cone around the sat's velocity vector. The intersection of this cone with the surface of the earth gives a line of position.



This principle is used for satellite based location of emergency buoys (e.g.: FAQ - Argos) only that those systems are designed for that purpose and have a much more favorable geometry as opposed to the BFOs measured by inmarsat.

Basically this Doppler derived line of position contains the BFO information which is uncorrelated to the BTO information. The intersection of the BTO line of position (ping arc) with the BFO line of position (doppler cone) defines the position of the aircraft.

On the last ping arc, D2_satellite is greatest due south of the sub-satellite point and it is null on the equator abeam of the satellite. In the middle region between these two extreme ends the variation in D2_satellite can be approximated as linear. According to my calculations at 0:11 UTC D2_satellite is ~12m/s due south of the satellite.

The radius r (not on the surface of earth) of the 0:11 ping arc is ~4000 km
=> the quarter circle section between due south and equator has a length of (pi * r)/2 ~= 6000 km

In the region where MH370 is suspected the variation in D2_satellite can be approximated as linear, then 1 m/s in D2_satellite variation corresponds to
6000 km / (12 m/s) = 500 km per 1 m/s D2_satellite variation.

At 1.6 GHz (L-Band uplink) 1 m/s relative LOS-velocity is equivalent to ~5.3 Hz Doppler shift.

Thus the 1000 km position error for 10 Hz bias error make perfect sense. Actually the error should be much smaller and if it is measurement noise and not fixed bias error, those errors partially cancel out over the series of BFO measurements.

[Hyperveloce]

Originally Posted by Gysbreght

Hi Hyperveloce,

A couple of questions:

I suppose you used the BFO calibration given in the report. What fixed frequency bias did you use?

I do not use an a priori AES fixed frequency bias (nor 150 Hz, nor a uniform distribution between 150-5 and 150+5 Hz like in the report), in my simulation the AES frequency bias is estimated (along with another parameter via a linear regression) for each simulated flight so that its simulated BFO profile is the closest possible (rms) to the measured BFO profile. The AES fixed bias obtained through this procedure are indeed close to 150 Hz and I get the following probability density function (an histogram actually):
https://drive.google.com/file/d/0B3s...it?usp=sharing
It is not the uniform distribution between 245 and 155 Hz used in the report, 80 % of the estimated freq. biases are between 150 and 155 Hz.

Originally Posted by Gysbreght

At each arc crossing of the segmented track there is a change of heading. The BFO changes with heading, so at each crossing there are two values of BFO, one for the arriving segment and one for the departing segment. Did you use both?

For this simulation, the reference trajectories are rhumb lines connecting points (constructed under varied constant speed hypothesis) on the varied handshakes:
https://drive.google.com/file/d/0B3s...it?usp=sharing
For each of the reference (non mutated/randomized) trajectories corresponding to each of the ground speed hypothesis (between 320 and 520 kts), the handshakes sample each time the new course taken arriving at a new ping ring. But in its subsequent mutations or crossovers, the relative timing between the trajectory and the hanshakes is also randomized (a handshake can occur slighty before or after the real one/related event on the trajectory).

Originally Posted by Gysbreght

Assuming that your 'goodness-of-fit' criterion is the root-mean-square of the deviations between the calculated and the measured values of BFO, which speed gave the best fit, and what were the deviations at each point?

The best fit is reached between 400 and 410 kts of ground speed, the goodness of fit I use is not a rms (linked to the 2-norm) but the max gap over the flight (linked to the infinite norm), for the best fit, it is less than 5 Hz for the BFO and less than 0.1° for the angle of elev/BTO.

[Hyperveloce]

Originally Posted by Ulric

Are there solutions for constant heading? i.e. rhumb line navigation?

There are rhumb line trajectories intersecting all the ping rings, with variable speeds, they depend where your starting point (at 18:29) is:

best bearing (so as to minimize speed variations) as a function of the starting point (lat/long)
https://drive.google.com/file/d/0B3s...it?usp=sharing
the blue circle portion is the 18:29 ping ring

associated (for the best bearing) mean speed as a function of the starting point (lat/long)
https://drive.google.com/file/d/0B3s...it?usp=sharing

associated variability (std) of the speed as a function of the starting point (lat/long)
https://drive.google.com/file/d/0B3s...it?usp=sharing
the best rhumb line trajectories exhibit a 27 kts standard deviation (speed variations) from their mean speed.

[Shadoko]

Sorry to disrupt the BFO things, but why the 00:11 handshake has been at this time and not at 00:15?

A phone call at 18:41 (channel released at 18:40:56.354) seems to have zeroed the one hour timer at 18:41, so the handshakes at 19:41, 20:41; 21:41 and 22:41. All of them in a ± 30 seconds interval. The next would have been at 23:41, but there was another phone call at 23:13:58.407 with the channel release at 23:15:01.886. This might have reset the one hour timer and without any contact, we would expect the next handshake at 00:15. And later, it was at 01:15:56 (so a two hours + 56 sec from the time of the phone channel release) that a not answered handshake from the sat happened, because the "partial handshake" of 00:19 would have not reset the timer.
But why a 4 minutes early HS at 0:11? From the Inmarsat data, it has been clearly initiated by the sat.

Just a remark. At 00:19, the whole larger search zone (the grey zone, page 42, 47 of the PDF) was in daylight, with the sun just rising near 40S 85E, and about 20° above horizon near 17S 107E (at sea level).

[getonit]

You guys having this discussion are obviously a whole lot smarter than me, as I was just a corporate pilot and now in "management". Would it make sense to take a 777 run it out where they think it went and see if these "pings", data points, signals and see if they match via all of these magical formulas? Just my thought.

[Ulric]

Thanks. The reason for asking about the rhumb line is that I'm looking for any assumption which might narrow down the possible heading. It occurred to me that the South Magnetic Pole might be a testable hypothesis if it could result in plausible tracks.

[sk999]

Would it make sense to take a 777 run it out where they think it went and see if these "pings", data points, signals and see if they match via all of these magical formulas?

Probably not. First, the route it ran would not be the actual route of MH370. Second, for calibrating the BFO (which is the more problematic measurement) one would want to carry the actual SATCOM equipment onboard MH370 and be able to return 3-F1 to its original, eclipsing orbit and inclination. Neither is going to happen. The formulae have been tested on myriad other aircraft, so the basic equations are not in doubt.