Malaysian Airlines MH370 contact lost

[SLFgeek]

I suspect this is one of the reasons for believing the aircraft was intentionally disappeared, as it's consistent with someone turning the power back on shortly after they knew they'd be out of radar range. I believe someone mentioned a few thousand posts back that the SATCOM terminal is on the same power bus as some other communication equipment, so turning off the power to that bus would turn all that equipment off.

This is the kind of statement that bothers me. If the purposed individual turned off the power bus to make the aircraft invisible (as much as theoretically possible), why bother turning it back on at all ? What's the point ?

Was there something else that they needed active that required turning it back on, or was someone on flight deck randomly reseting CBs trying to get things back on and working.

Also, how would a pilot who normally flies with ADS enabled (for SSR) even know the precise range of that PSR ? Did enabling the power bus cause the ADS to become active again as well ?

Various drawings suggest that that the T7 has dual pairs of SATCOM antennas, mounted on either side of the hull (but near the top). I'm more inclined to wonder if changing course caused the re-establishment of SATCOM, than power being restored.

[Linktrained]

whitav8r
".. a minor bank angle..." (if stable) will cause a certain radius of turn, now beyond my arithmetic, with decreasing T.A.S. as well !

[BuzzBox]

Things get uglier when the 2nd engine cuts out...
1) TAC will (would?) drop out when it detects that both engines are out.
2) When the engines stop supplying AC power and hydraulic pressure, the hydraulic flight controls (including the TAC-adjusted rudder) will remain locked in place.
3) The Autopilot will drop out (disengage) with the total loss of AC power.
>> This creates a race condition in which the exact sequence of events will determine the flight control settings locked in after power is lost.

The suggestion has been made that the Autopilot would drop out first, with the rudder trip from TAC still fully applied. This would lead to an unfavorable condition while power is out. And, when power returns, the autopilot remains disengaged, so these flight control positions will be maintained.
-- This is what leads to the expectation of a very short glide period. The the plane will begin turning right at 2nd engine flame out, and continue that turn even when power is restored from the APC and/or RAT.
This is most likely the 'spiral decent' mentioned in the June 26th ATSB report.

That's not quite correct. The aircraft would have lost AC power when the second engine failed, triggering a series of events:

1. The flight control system would revert to secondary mode and the TAC would immediately drop out as a result of that reversion.

2. The RAT would immediately deploy and begin supplying electrical power via the standby electrical system and hydraulic power to the primary flight controls via the centre hydraulic system.

3. The APU would attempt to start automatically.

There would be several power transfers while all this was taking place, but the left and centre flight control power supply assemblies would remain powered throughout, via the main battery.

I don't believe the flight controls would remain 'locked in place' as you suggest. Some hydraulic power would be lost when both engines failed, but there would still be some hydraulic pressure and shortly thereafter the RAT would spin up and begin providing pressure to the centre system. With no further control inputs, the rudder would centre.

There is a theory that a spiral descent would be triggered by the extra drag from the RAT, which deploys from the RHS of the rear fuselage, behind the wing.

[JamesGV]

With such a high level of redundancy on the T7, it is hard to think of anything that can account for this tragic episode, other than something of a "catastrophic" nature.

A SATCOM "reboot" in flight !
Only with a simultaneous dual IDG and APU failure would that happen.
(BUGs/PMG's don't power SATCOM or TCAS)

[Pontius Navigator]

It is easy to calculate the radar horizon given height above ground for antenna and target.

Obvious assumptions in this case at 37,000ft (FL370) and radar at 15ft. This gives a theoretical range of 236 nms.

These assumptions are probably incorrect and unless your perpetrator was sufficiently knowledgeable a 3 minute (23 mile) gap between leaving radar cover (225 miles) and restoring power would be too close or ineffective.

In reality FL370 would be nearer 39,000ft in the tropics. That gives a radar horizon of 241 nms.

Then there is a property known as anaprop or anomalous propagation where the radar bends and goes further over the horizon.

Now in this instance we know the last radar contact fell well below the theoretical ranges. Did this mean the aircraft was at 32,000 feet or that the radar performance was well below optimum?

I submit therefore that for there to be a deliberate masking until beyond the radar horizon and resumption of power 3 minutes later was unlikely the timing was too tight for the probable variations in range but just pure coincidence.

[Shadoko]

About the RAT.
If the log-on sequence of 18:25:xx was the result of a power loss and come back, was not the RAT to be extended at this time?

[Ulric]

Shadoko, you are thinking along the same lines as I am.

[sky9]

The RAT comes out when a double engine failure is sensed. The APU also automatically starts and provides power until that also runs out of fuel. The log on was probably caused by the APU starting with the fuel available in its fuel line which rapidly ran out.

[Propduffer]

The 18:25 login is attributed to the removal of power from the left AC bus which powers the sat transmitter and modem. This "power failure" was a local event, only affecting items powered by that bus.

The power failure which brought about the 00:19 login attempt was a total failure as explained above.

[Shadoko]

sky9: from what I read (I don't know about that), I understand that RAT extends (and APU starts) if both AC transfer buses lose power in flight (even with engines running). That could be from failures or if both general switches (and both backup generators switches?) are switch off.

Propduffer: the log-on sequence could not be from power failure. It could only happen from power recovering (if the lost was from an electrical "problem"). Nobody knows for how long the SATCOM was unpowered (or not functional) because there was no communication at all between 17:07 and 18:25. The non-response to a request at 18:03 proves that the link was not functional around this time. This doesn't prove that the link was out of order all the time between 17:07 and 18:25.

Could someone confirm that the cockpit door cam is on the same electrical bus that the IFE?

[BuzzBox]

If the log-on sequence of 18:25:xx was the result of a power loss and come back, was not the RAT to be extended at this time?

Not necessarily. Three things will cause the RAT to deploy automatically:

1. Loss of AC power to both AC transfer busses in-flight.

2. Both engines failed and centre system hydraulic pressure low.

3. All three hydraulic system pressures are low.

Given that the aircraft flew on for a number of hours after the log-on attempt at 1825, it seems unlikely that any of the above events occurred at that time.

As Propduffer said above, the log-on at 1825 could have been caused by a power interruption to the satcom system. Possibilities include the tripping and resetting of the satcom circuit breakers, or some other short-term interruption of power to the AC bus that supplies the satellite data unit.

[Propduffer]

The fact that the transponder was turned off just before 1:21:13 and no ACARS message was generated to report that fact seems to tell us that ACARS was shut down not long after the 1:07:48 report.

[Propduffer]

Possibilities include the tripping and resetting of the satcom circuit breakerThese CB's (there are three of them for the sat xmit system) are located in the electronics bay. The only way to turn off the satellite transmitter/modem from the flight deck is to drop the left AC bus.

[BuzzBox]

The only way to turn off the satellite transmitter/modem from the flight deck is to drop the left AC bus.

Agreed. However, deliberately shutting down half the aircraft's AC electrics, even for a short time, would be a major step to take, with serious repercussions. I'd be very surprised if a pilot would do that just to interrupt the satcom system. There are far easier ways of disabling the ACARS & satcom if that was the intent.

The satellite log-on at 1825 may have been caused by a power interruption on the aircraft, or it may have been caused by something else. In my experience, aircraft occasionally 'lose' the satcom for a short time time in-flight and subsequently log-on again automatically.

If the log-on was caused by a power interruption, then possible causes include a malfunction in the L Main AC Bus, causing the system to isolate itself automatically; or deliberate actions taken by the flight crew to achieve the same result. That's not something the flight crew would normally contemplate, except perhaps in the case of an electrical fire where they might attempt to isolate part of the system. As I said above, I think it is highly unlikely they would isolate half the aircraft's AC electrics just to turn off the satcom.

[Propduffer]

Shutting down half the aircraft's AC electrics, even for a short time, would be a major step to take, with serious repurcussionsIn light of the actions taking place aboard 9M-MRO at that time, I wouldn't think repercussions from de-powering the left Bus and the left FMS would be of much consequence to a person in the act of diverting the plane from its scheduled flight path.I'd be very surprised if a pilot would do that deliberately just to interrupt the satcom system.I don't think it was done to disable the sat system - in fact I believe that was an unknown and unintended consequence of de-powering the left AC Bus. IMO the bus was taken down to de-power the left FMS which would allow direct pilot control of the aircraft without interference from the FMS. The FMS would provide physical resistance to control column movements which are outside those Boeing expects an airliner to experience in "normal operation." There are far easier ways of disabling the ACARS & satcom if that was the intentThere is no other way to disable the satellite transmitter/modem from the flight deck other than to de-power the left AC Bus. The CB's for this are in the EE Bay.

[BuzzBox]

IMO the bus was taken down to de-power the left FMS which would allow direct pilot control of the aircraft without interference from the FMS.

Surely there are easier ways to take 'direct pilot control of the aircraft without interference from the FMS'. Why not just use other autopilot modes such as HDG or TRK? Failing that, why not just disconnect the autopilot and fly manually? Far easier than taking down half the aircraft's electrics.

There is no other way to disable the satellite transmitter/modem from the flight deck other than to de-power the left AC Bus. The CB's for this are in the EE Bay.

I beg to disagree. The CBs for the satcom may be in the EE bay (I can't remember to be honest), but on the 777s I flew there was definitely a way for the flight crew to force the system to log-off.

[Propduffer]

In over 5,000 hours flying the 777 I never found the need to de-power an entire AC bus in order to take 'direct pilot control of the aircraft without interference from the FMS'.And presumably you've never had the need to turn off the transponder either, yet these are things that happened aboard MH-370 that night.

We have to address the facts as they present themselves. One fact that won't go away is that there was a satellite communications blackout from 9M-MRO which began at 17:07:48.9 and lasted until 18:25:27.4 when the sat modem initiated a cold login to the satellite. A cold login would only be caused by restoration of power unless you prefer to think that this sat modem which had performed flawlessly since the plane was delivered from Boeing, all at once developed a bug the same night the plane was diverted to the southern ocean.

The sat receiver/transmitter/modem receives its power from the left AC Bus in the EE Bay. Down there there are three circuit breakers which provide the only way to turn power off to the sat equipment. There is no way to power off the sat equipment from the flight deck other than to bring the left AC Bus down. Your "on the 777s I flew there was definitely another way of disabling the system from the flight deck" reeks of the kind of stuff that comes out of the Malaysian transport/defense ministry. It appears that you parsed your words to give yourself as many outs as possible.

Let me state it clearly: there is no way to remove power from the satellite modem/transmitter from the flight deck on a 777 without powering off the left AC bus.

You can use your knowledge of the Boeing 777 to shed light on what happened aboard MH-370, or you can use your knowledge of the aircraft to obfuscate the issue. Your choice.

But I suggest that you give some thought to what you are doing.

[BuzzBox]

...presumably you've never had the need to turn off the transponder either, yet these are things that happened aboard MH-370 that night.

Turned off, or failed, who knows for sure? I certainly don't. The suggestion that it was deliberately turned off is pure speculation at this point. The only known fact is that it stopped transmitting.

It is certainly possible that the left AC bus was de-powered for a time, I never said that it wasn't. However, your suggestion that it was deliberately de-powered in order to 'take direct pilot control of the aircraft' is, frankly, preposterous.

Further, it IS possible to force the satcom to log off a satellite from the flight deck, WITHOUT de-powering the system.

[Propduffer]

it IS possible to force the satcom to log off a satellite from the flight deck, WITHOUT de-powering the systemPlease tell us how that's done.

(Without rolling the plane inverted)

[Propduffer]

Who do you think you're fooling?

[DJ77]

Gentlemen, may I recall this thread is about resolving the Inmarsat data. Speculations should go to ""Rumors and News".


Someone suggested that the satcom re-logon at 18:25 may have been caused by power interruption. This is not by far the only possible technical explanation of this event.


Propduffer, you appear entirely focused on finding ways everyting was done intentionnally. That just reflects a lack of imagination or knowledge about all that could go wrong aboard an airplane.

[Propduffer]

Someone suggested that the satcom re-logon at 18:25 may have been caused by power interruption. This is not by far the only possible technical explanation of this event.Other than a power failure or the modem coincidentally developing a bug that very same night, what other "technical explanation" do you have that would cause a modem to initiate a login?

[tawnybird]

Gentlemen,

Whether or not this is the proper thread for this particular discussion (though it seems appropriate enough considering it's mostly technical aspects), I dare interject nevertheless.

Propduffer is correct in this instance.

Buzzbox, with due respect, you first agree that the only way to turn off the satellite transmitter/modem from the flight deck is by dropping the left AC bus.

Then, seemingly out of thin air, you disagree all of the sudden and provide no basis for this change of mind, other than to claim that this is 'sensitive information'.

DJ177, with due respect, I feel it unfair to accuse Produffer of a lack of imagination. He has clearly stayed solidly grounded in fact, and the burden is on you or others to factually discredit his claims, which you have failed to do.

[JamesGV]

@DJ77

I asked about the "log on" at 18.25
It was suggested it was likely a power interruption.

The T7 has a high level of redundancy so I was trying to discover "intentional" and "non intentional" reasons for this from (limited) OFFICIALLY RELEASED information (and where the location of the a/c was).

...and what a "thankless task" that is !

1. After 1.21am MYT (around IGARI) there is nothing.
2. There is no "relevant" data gleamed from the 2.03am MYT/2.05am MYT Inmarsat transmission.
3. There is no "officially released" Primary Radar data. The ATSB Report is "not recognised" by Malaysia, although the "last contact" given by the Malaysians as 2.22am MYT is noted
4. A "new log on" is given at 2.25am MYT, which places the a/c in the Banda Aceh area.
5. There follows a number of "handshakes" recorded by Inmarsat along with 2 "telephone calls", which correspond to the activities MAS undertook.

Like it or not, they are the facts. There is NO overlap in events.

[BuzzBox]

Tawnybird:

I did not disagree with the suggestion 'that the only way to turn off the satellite transmitter/modem from the flight deck is by dropping the left AC bus'. I said that 'it is possible to force the satcom to log off a satellite from the flight deck, WITHOUT de-powering the system'. There is a difference.

The basis for that statement? A few thousand hours flying 777s in airline operations.

The satcom is not completely autonomous from the flight deck. Honeywell was kind enough to provide the flight crew with some control. Most of those functions are not normally used by the flight crew, but they are available.

[JamesGV]

@BuzzBox

As a way of "resolving" this "log off and Log on" issue which is believed to have occurred at "some time prior to 1825 UTC" and "at 1825 UTC", you are saying with your time on T7's it is possible to "force" a "log off" without de-powering the system...and without leaving the flight deck.

Assuming you did this. The system has power but is "off line", how would you "log on" again ? Would this be something that would be "automatic" ? Or a "manual" function ?

[DJ77]

@ tawnybird
At least, by saying that there is no need to "de-power" the left FMC in order to take control of the T7, BuzzBox has a more solidly grounded point than Propduffer.


@JamesGV
Formally, the log on at 2.25am-MYT is on the relevant Inmarsat-computed ping arc. It is assumed close to Banda Aceh only due to the existence of the radar track of an UFO believed to be MH370. I mean this has never been a certainty.
As I understand the works of RichardC10, Gysbrecht or Hyperveloce, when the 2.25 log on is assumed close to Banda Aceh it appears impossible reconcile the Inmarsat data and possible trajectories without large variations of ground speed. According to the end of flight analysis in ATSB report of June 26th, the unresponsive crew scenario appear more probable (given available evidence). I am vainly trying to figure out how this could add-up with large speed variations an neither can I find any logic for intentionally fiddling with speed.

[JamesGV]

@DJ77

Taking your first point....

"It (the "log on" and therefore what we believe was MH370) is assumed close to Banda Aceh only due to the existence of the radar track".

Are you saying that IN CONJUNCTION WITH the "last/lost conatct" Primary Radar data is it possible to establish that a plot for the "first arc" is possible ?

As in, it is IMPOSSIBLE to establish a plot for this "start point" using Inmarsat data/BFO/Doppler Effect alone ?

[PrivtPilotRadarTech]

Search radars operate at frequencies above 1 gHz, where atmospheric refraction is less of a factor. The Butterworth radar track data photo that was posted some time ago went out to 200 miles, with a gap from around 100 to 130 miles. Using an online radar horizon calculator it looks like flying at 5000', then climbing thru 10000' at 130 miles would produce a gap like that. That would jibe with the claim that MH370 flew "as low as 4500'". However, the radar horizon at 200 nautical miles is about 23000', which does not jibe with the reported last altitude data of 29500'. In my experience, a huge target like an airliner is easily detectable at max range if it is above the radar horizon. Max range of a search radar is typically around 250 nautical miles, and it is a hard limit, determined by the pulse repetition frequency.

[Ulric]

If we accept that the Inmarsat arcs and the radar trace are correct then we have a timing anomaly. If MH370 maintained its last known course - the radar track - then we have a problem because it would cover only about 140NM between two ping arcs which are timed at 1 hour and 12 minutes apart. It is speculated therefore that the aircraft took a large and unexplained northerly diversion (the hook referred to in the Australian report) in order to maintain both a plausible airspeed and the timings of the satellite pings. There is as yet, no reasonable explanation of why this should have happened. This northerly excursion would possibly have taken MH370 as far as Port Blair in the Andamans.

In reality there are an almost infinite number of possible courses which could account for the timings including all sorts of arbitrary "holding patterns" and at present we must accept that there is no evidence which can tell us what the real path of the aircraft was during this period.

[Propduffer]

If we accept that the Inmarsat arcs and the radar trace are correct then we have a timing anomaly. If MH370 maintained its last known course - the radar track - then we have a problem because it would cover only about 140NM between two ping arcs which are timed at 1 hour and 12 minutes apart. It is speculated therefore that the aircraft took a large and unexplained northerly diversion (the hook referred to in the Australian report) in order to maintain both a plausible airspeed and the timings of the satellite pings. There is as yet, no reasonable explanation of why this should have happened. There is a reasonable explanation which does not require a northward excursion. The portion of the 18:29 (The BTO data was obtained at 18:29, not 18:25) ping ring plots as nearly north - south (NNW-SSE) at the point it is likely to have intersected the flight path of MH-370. The exact point where the flight path intersected the ping ring was at about 6°32'34.04"N 96°42'17.66"E. this is about 20nm east of where the Malaysians have vaguely given it's position at 18:22. I accept the Inmarsat BTO data as accurate and thus ignore the Malaysian position estimate at 18:22.

As you say, the 19:40 ping ring does lie only about 140nm from the 18:29 ping ring - and it is probable that 9M-MRO did cross the 19:40 ring or pass very close to it about 18 minutes after reaching the 18:29 ping ring.

However at that time (about 18:47) there were no satellite communications in progress, so there was never any BTO data collected to reflect that time and place. Thus the 140 nm distance between the two ping rings is meaningless as far as 9M_MRO's flight path is concerned.

What happened at about 18:47 was that 9M-MRO turned south - and when the 19:41 Inmarsat transaction took place and logged the BTO data, 9M-MRO was at about 2°38'57.58"N 94°42'33.49"E (plotted for a 376kt speed) or about 1° 3'43.56"N 94°42'30.43"E if the plane was flying at 470kts.

The 19:40 ping ring also traces an almost north - south path at this longitude so 9M-MRO was on or near that ring for over an hour on its journey south but it was only at 19:41 that the BTO data was collected.

[JamesGV]

Oldoberon

That's what I mean (Woodpecker is "lost in the trees" methinks).

I am trying to arrive at 2.25am MYT (1825 UTC) and the events surrounding that event.

The SATCOM sys has to go "off line" (either offline or de-powered) prior to this login. (For Woodpecker... the 1825 UTC login !).

For a login, it has to have logged off.
It can't be a dual IDG failure (APU auto starts)
It "could" be a dual IDG failure and APU failure.
BUT what restarts SATCOM ? So it can't be that.
So a "forced" log off ?
Then why does it log on again ?

You see a "mysterious power outage" doesn't explain anything.
As in.... the sys "logged off" before it "logged on" again.

[Ornis]

MH370: New evidence of cockpit tampering as investigation into missing plane continues - Telegraph (http://www.telegraph.co.uk/news/worldnews/asia/malaysia/10933917/MH370-New-evidence-of-cockpit-tampering-as-investigation-into-missing-plane-continues.html) The Australian report revealed attempted log on at 2.25am, three minutes after it was detected by Malaysian military radar, as the plane was flying north of Sumatra.

David Gleave from Loughborough University said the interruption to the power supply appeared to be the result of someone in the cockpit attempting to minimise the use of the aircraft's systems... consistent with an attempt to turn the plane's communications and other systems off in an attempt to avoid radar detection.

Inmarsat has confirmed the assessment but says it does not know why the aircraft experienced a power failure.

Peter Marosszeky, U of NSW ... the power interruption must have been intended ... would not have caused an entire power failure but would have involved a "conscious" attempt to remove power from selected systems on the plane ... The aircraft has so many backup systems. Any form of power interruption is always backed up by another system. It would have to be a deliberate act to hijack or sabotage the aircraft.

Contrary opinions from other independent informed sources?

[JamesGV]

Propduffer.....

Or... the a/c "held" off the coast, further south of Banda Aceh.

It is the only way you can get "the fit".
And that means it wasn't a "constant" heading South.
And that doesn't fit the ATSB report *which wasn't written for "this" particular purpose anyway...but did state "a constant heading".

Edit (which is speculation) a "hold" would fit "an a/c in distress", rather than any other suggestion.
Although there are other possibilities, however unlikely they may be.

[JamesGV]

"It would have to be a deliberate act to hijack or sabotage the aircraft".

.....Which is speculation as he wasn't there.

(And he readily links the 2.22am MYT with the 2.25am MYT log in, without providing specific proof that this was the case, however likely it was).

[Propduffer]

It is the only way you can get "the fit".I don't know what you mean by that.

By "held" do you mean flew in circles? If so, I've never heard that theory before.

IMO it turned south at or near POVUS or NOPEK, and like everyone else I assume that it held a constant speed and course, or GC route until it ran out of fuel. But like everyone else I can't find a speed or route that intersects all the ping rings at the right times.

[PrivtPilotRadarTech]

The only thing you can turn off that affects radar detection is the transponder, which can simply be turned off. So the power failure is NOT "consistent with an attempt to turn the plane's communications and other systems off in an attempt to avoid radar detection." However, it is consistent with an attempt to avoid communication.

[JamesGV]

It's not a new concept.

https://www.dropbox.com/s/bu7ajeqvhedf37i/MH370%20Flight%20Path.pdf

Edit, that's if there is a route south.
At Banda Aceh a U.K sailor saw a "glowing orange" a/c heading "south". And two other a/c heading "North".

Mention "conspiracy" and they jump on you here !
As an "ex pat" Brit working in the sandpit, I don't believe half the things I am suppose to.
I'm just saying "it either held" (in a hold) or didn't get there at the time specified. (Or "wasn't it")

[Propduffer]

I'm just saying "it either held" (in a hold) or didn't get there at the time specified. (Or "wasn't it"Where is there? If it's 19:41, there is no need for flying in circles. That time works out with no problem, the location is near 2°38'57.58"N 94°42'33.49"E.

If it's the terminal spot you mean, the where is the question. There are a lot of possible solutions for that. As many solutions as there are discernible speeds. But to where?

The dropbox pix is a non sequitur, it shows the flight path over Indonesian territory. We may not know everything, but we know for sure it never overflew Indonesian airspace, especially Indonesian landmass. That flight path puts the plane almost directly overhead of the Indonesian radar at Lhokseumawe.

The flying in circles premise is utter nonsense, there is no indication of that, there is no evidence for it and it makes no sense. The "sighting" you mention was investigated and came to nothing.

[MG23]

You see a "mysterious power outage" doesn't explain anything.
As in.... the sys "logged off" before it "logged on" again.

There's no explicit logoff message in the released logs, as there would be if someone on board had told it to log off. So it lost the satellite signal, or lost power, or temporarily failed in some other way, but about the only thing we know for sure is that it didn't do a clean logoff.

[sky9]

Would you like to look at YWKS as "destination" then have a look at SkyVector.com for the intermediate waypoints and see if you can tie that in with the ping circles. Routes from the north towards YWKS all go through the ATSB amber arc.
If someone can produce the en-route winds for that date in a readable format it would then be possible to modify the reasonable cruise mach no (0.82?) for the ambient winds to generate a G/S.

[JamesGV]

MG23

I stand corrected. This is true, there is no "log off" recorded.
That in itself maybe significant.

[JamesGV]

@Propduffer

We don't know alot at the moment !

We know (it is claimed) that Lhokseumawe DID NOT detect MH370 in the Straits. And we KNOW that DUAMO to PEN is 160nm wide.

Indonesia claim they cover 240nm from Lhokseumawe.
Could be a "pi**ing contest" ?

So where did it go ?


And some pretty pictures to go with it....
http://alert5.com/2014/03/16/could-tni-au-radar-unit-231-shed-light-on-the-fate-mh370/

[nick1austin]

The exact point where the flight path intersected the ping ring was at about 6°32'34.04"N 96°42'17.66"E. this is about 20nm east of where the Malaysians have vaguely given it's position at 18:22. I accept the Inmarsat BTO data as accurate and thus ignore the Malaysian position estimate at 18:22.
There must be some mistake on your calculations because 6°32'34.04"N 96°42'17.66"E isn't on the ping ring.

I can't find anywhere that the Malaysians give coordinates for loss of PSR but using the map contained within AE-2014-054_MH370_SearchAreasReport.pdf it appears to be approximately 6.6°N 96.3°E.

If I project a great arc from just south of Penang Island through that point until it intersects the the ping ring I get 6°52'14"N 95°28'28"E. That works out remarkably well giving a constant ground speed all the way from the turn-around in the South China Sea to the 18:27 ping ring of 560 knots. I therefore can't see any justification for claiming that the 18:22 position is wrong.

[Propduffer]

@nick
We don't seem to be using he same ping ring information.

Here's where mine came from.
https://www.dropbox.com/sh/ezs6imv367w8xqb/AADqEhUIaO__UkCXLEIGDVwQa

It's true that the Malaysians never gave the coordinates for the loss of track. Their tightest definition for that location afik is about 10 miles past MEKAR, your location for 18:22 as stated by them is close enough IMO.

560kts gs is a bit fast for a 777.

[Propduffer]

@James
We know (it is claimed) that Lhokseumawe DID NOT detect MH370 in the Straits. They didn't say that - read it again - what they said was that they didn't track such a target over their territory.

12nm north of Pulau We Island fits that description just fine.

[Propduffer]

@Gysbreght
You state that as a fact, but on what basis?
Sure, the plane must have turned south at some time between 18:29 and 19:41, but when?I've spent a lot of time on this point, I was originally of the opinion that the plane went further west before making the turn.

But after working with the ping rings and taking into account fuel burn limitations I have come to the conclusion that even as far as IGOGU is extremely unlikely. Also there would have been no reason to have gone that far, the normal airline routes past the northern tip of Indonesian Territory are through IGEBO, so I conclude that's about where the turn was made.

This would have occurred about 18:47.

If you have other theories, I'd like to hear them.

[Propduffer]

@sky9
YWKS might work and it is such a cute solution - but the guy in the left seat could have punched in anything he wanted to, so I see nothing compelling about that waypoint.

If someone can produce the en-route winds for that date in a readable formatAn intelligent approximation of this data is very much needed. I would think it would be best done by someone experienced in plotting winds aloft in that part of the globe - which leaves me out.

Is there someone out there..............................?

[nick1austin]

Here's where mine came from.
https://www.dropbox.com/sh/ezs6imv36..._UkCXLEIGDVwQa

It's explained somewhere that the PR_18_27_47.kml file is incorrect. Use Ping_Ring_18_29.kml instead.

[OleOle]

The inmarsat raw data release has a BFO Value of 88 for the 18:39:55 unanswered ground to air phone call. That is an (imho strong) indicator that the aircraft velocity vector at that time already had a southern component, i.e. the heading was somewhat south of due west.

[Propduffer]

@OleOle

The premise of the plane being southbound at 18:39 won't work unless it overflew Indonesian territory, which I consider extremely unlikely.

If we put 9M-MRO's position at 18:29 on the intersection of the ping ring and the FIR boundary (6° 0'5.68"N 96°46'17.43"E) (which is the closest possible position for the plane to have been to clear Indonesian airspace north of the tip) the distance to POVUS (the earliest point of a turn which would clear Indonesian airspace) is about 125nm. That's too great a distance to have been traveled in ten minutes.

[porterhouse]

or was someone on flight deck randomly resetting CBs trying to get things back on and working."randomly" resetting CBs is not part of standard operating procedure, these days pilots are usually not allowed to touch a CB unless they get permission from their maintenance department. Also attempt to "get things back on and working" is really incompatible with the flight of this aircraft. Faced with emergency they would be looking rather quickly for a land/airport and with the magnetic compass working (needs no electricity) they could steer aircraft in the right direction, specially in view of the excellent visibility in the area and their general proximity to land. So here we are - you can easily poke holes in any theory.

[whitav8r]

Here is Richard Cole's list of winds aloft:
TIME Wind Direction WindSpeed(knots) Lat Lon
18:22:00.000 10 4 6.58 96.34
18:25:27.421 355 2 6.79 95.94
18:28:14.904 330 2 6.82 95.84
18:39:55.354 90 11 5.32 95.47
19:41:02.906 70 22 -4.01 93.25
20:41:04.904 85 12 -11.81 91.44
21:41:26.905 10 11 -19.64 89.76
22:41:21.906 275 44 -27.45 88.14
23:14:00.904 265 55 -31.63 87.54
00:10:59.928 245 43 -38.78 84.96
00:19:29.416 240 49 -39.82 86.22

I think the source is:
earth :: an animated map of global wind and weather (http://earth.nullschool.net/#2014/03/07/2100Z/wind/isobaric/250hPa/orthographic=-263.14,-7.41,791/grid=on)

[Propduffer]

Nick, I'm aghast!


"It's explained somewhere that the PR_18_27_47.kml file is incorrect. Use Ping_Ring_18_29.kml instead."This throws everything I've been working on for the last week off by 70 nm.

I thought one of the base assumptions was that the 18:29 ring and the 19:41 ring were 140nm apart, and the rings I've been using matched that pattern.

Also, the ring KML's I used were the most recent on Steel's site.

Where did the information about the incorrect 18:27 kml file come from?
I can't find it and I (we all) have to get this straight.


@OleOle
Read this please. It makes the turn south possible at 18:39.

[whitav8r]

I thought that the last radar contact position @ 18:22 was approximately 6°35'53.39"N, 96°17'45.71"E, so getting at far as 7°20'24.52"N, 95° 7'46.74"E by 18:33 is very practical at 480 knots or so. From there and a little more west, due 194T avoids the NW tip of Indonesia. I'm not saying that's what happened though. I wonder if the radar at Lhokseumawe was operating that night.

[JamesGV]

Makes me wonder that to !

Lhokseumawe latest "in the news" effort managed to track and call down a U.S. transport a/c en route from the Maldives, via Sri Lanka to Singapore.

I can't see how the TNI missed this.



They have a CSF TRS-2215. Displayed range about 330. The TNI say 240. Narrowest point around 160

[OleOle]

@Propduffer

Hdg must not have been due south at 18:39. A southwesterly course like on the leg NILAM -> POVUS could be consistent with the BFO as well.

[JamesGV]

It is amazing to me that MH370 can appear and disappear three times.
And there is not one consistent story to be had !

T/O to IGARI and just beyond.
Malaysian Radar heading inbound to the North of Banda Aceh
Inmarsat from a disputed point to another disputed point.

Not one overlap in all three cases. Amazing.

[Propduffer]

@Nick
The substitution of the 18:29 ring doesn't fit with the other rings.

In Richard Cole's appendix B we find that there is a 255km difference between the 18:29 ring and the 19:41 ring, (137nm.)

Your suggestion gives a distance of about 66nm between these rings so that can't be correct.

My original placement appears correct.

[AAKEE]

I stand corrected. This is true, there is no "log off" recorded.
That in itself maybe significant.

Thats for the sat com. Would be nice to se the ACARS logs from VHF-com.
Early in investigation it was said that Acars was deliberately shut down.
If so, there should have been some kind of log of on those logs ?

[Ulric]

The point that needs to be made here is that any assumption for what happened between 18:29 and 19:41 implies an assumption about whether the plane was travelling east or west when it intersected the 19:41 arc. In turn, this affects the starting point for the long southward track hypothesised to have been flown on auto-pilot. It should be obvious that any constant speed or constant track solution for the later arcs up to 00:19 is rotatable around the satellite location. In other words, the terminal point is very sensitive to assumptions about what happened between 18:29 and 19:41 and the point at which the southerly course was started. The ability to rotate the track around the satellite without disturbing the timings for the southerly leg means that an assumption about the position of either end of the track (the point where it turned south or the terminal location) constrains the possibilities at the other end.

BTW, I have plotted my own ping rings for this and agree with the Australian report that the early rings agree with the submitted Malaysian Radar track.

[Ulric]

There is a difference between a clean shut-down and pulling the plug out.

[Pontius Navigator]

"randomly" resetting CBs is not part of standard operating procedure, these days pilots are usually not allowed to touch a CB unless they get permission from their maintenance department..
I take issue only with this.
I would agree with USUALLY but in unusual circumstances where all else fails then 'random' maybe a last resort.

[Propduffer]

Ulric:
I have plotted my own ping rings for this and agree with the Australian report that the early rings agree with the submitted Malaysian Radar trackWe should discuss this and attempt to reach agreement about the ping rings. No conversation can go forth without agreement here.

You plotted your own rings? This means that you calculated distance for each ring using the the BTO data and included the knowledge of the satellite's exact position at that time; did you also compensate for the Earth's oblong shape?

I'm not up for such mathematical gymnastics so I just copied the latest KML version on Duncan Steel's site. Then after being challenged by Nick above I measured the distance between 18:29 and 19:41 rings and the version I'm using matches the expected distance (137nm is close enough to my expectation of 140nm.)

I further checked this against a Richard Cole graphic. I have previously downloaded a drawing labelled "Track_zpse219149d.jpg" (no url) which has the title "Constant Speed path to the final search area" dated may 29th. And the ring plot that I'm using matches this Cole graphic. Although I don't have a url for that jpg I can direct you to page 39 of Richard's data analysis which has a similar graphic with the same information.

Comments from you would be appreciated.

[Propduffer]

OleOle
A southwesterly course like on the leg NILAM -> POVUS could be consistent with the BFO as well.That works for me.

Although I can't stop myself from thinking that maybe the 18:22 location was really 6° 0'0.02"N 97°30'0.05"E and the plane flew along the FIR boundary to POVUS from there. Could your BFO numbers support that theory?

[olasek]

where all else fails then 'random' maybe a last resort
If my life (and my passengers') were on a line I very doubt I would be resetting CBs without authorization. This often lead to fires. This makes even less sense with redundancy built into 777. I rather fly with some functionality disabled than with smouldering cockpit.

[Ulric]

We should discuss this and attempt to reach agreement about the ping rings. No conversation can go forth without agreement here.

Just to be clear, I took Duncan Steele's calculated radii for the rings and used a kml circle generator to produce the kml files. I have also produced a set of kml waypoints for Malysian, Indonesian and Australian Airspaces from the data at fallingrain which are useful.
http://www.thesamestory.com/kmlcircle/
http://www.fallingrain.com/world/AS/waypoints.html

[Pontius Navigator]

Olasek, by 'random' I implied educated rather than random.

You said it would make nonsense of built-in redundancy - true, but when built-in fails then you're on your own.

Holding in a popped CB is one thing, popping CBs is another.

Pretty clearly this accident was outside any foreseen contingency, and while we are considering probabilities, nothing can be ruled out.

[Propduffer]

Ulric

How did you locate the centerpoints for each of the rings? The satellite wasn't stationary in it's orbit you know.

Would you be so kind as to post your KML file online so that we can all examine it?

[mm_flynn]

Thats for the sat com. Would be nice to se the ACARS logs from VHF-com.
Early in investigation it was said that Acars was deliberately shut down.
If so, there should have been some kind of log of on those logs ?

I am not aware of any substantiated report that any system on MH370 was 'shut down'. All we know was that various systems were no longer responding at certain times. We also know that the satcom was 'off' at least twice by the fact that it logged on twice. 'Off' in this sense means that the aircraft terminal was not in a logged on status (due to loosing the signal, being shut down, having power lost, etc.) but doesn't tell the reason it was 'off'.

we have a last Transponder return time, a last ACARS message time, a Last RTF time, several missed phone calls times, but not a single logoff or power down message. Nor do I believe any of those systems generate such a logoff messages.

As such, I don't believe there is any information to establish if systems were individually powered down at their normal power controls, by pulling local circuit breakers, by pulling busses, because breakers popped due to overload, because the device was damaged (other than the satcom which clearly was not damaged in that it came back twice).

The two most obvious path's of speculation are
1 - some technical emergency happened which motivated the crew to power down most of the aircraft's systems and after a reasonable time (at minimum I believe the time between loss of transponder and logon of the satcom) power up at least the Satcom, but not use it.

Or

2 - someone (crew, crew forced by passenger, passenger) did the same series of events with the objective of making the aircraft disappear.

It does seem difficult to imagine a technical emergency that would a- be responded to in such a way, b - allow the aircraft to fly for many hours, c - prevent any communication (particularly given that the satcom was back on line for most of the time)

[JamesGV]

As with Swiss 111, the SATCOM "logged on", but it could not be determined with any accuracy when it "logged off".

[Ulric]

Satellite is normally assumed to be at the equator and 64.5E. My reading suggested that during the flight of MH370 it's position varied between 0.8 and 1.5N because of it's little elliptical orbit up there. I have centered my rings at 1.2N as a compromise.

I can't remember now where the position information came from but armed with Steele's ring diameters and a circle generator, you can put them anywhere you like. It's a 10 minute job to generate your own rings.

[Propduffer]

@ Ulric

The circle generator is no big deal, if you have GE pro you don't need a circle generator.

But the satellite position information is kind of a big deal, that "little elliptical orbit" covers a lot of miles. Since you don't have the KML file do you have the distance you used for the 18:28 ring? (I know where the satellite was then.)

[Ulric]

I don't think it will help much but here it is
http://www.duncansteel.com/archives/549

If someone hadn't written it, I would have coded my own circle generator - it's easy to do - no need to pay Google for that. The other thing it's useful for is to put range rings around the known and suspected radar installations which might have picked up a trace.

[Ulric]

Swissair 111 had a fire on board which disrupted the systems. We don't know what happened on MH370 but the speculation is that circuit breakers were pulled.

Consider your own PC. If you shut it down normally the event is logged, that is quite different from terminating the power by pulling the plug out (or burning the cables). In that case, you won't get a shutdown log but, depending on how the system is configured, you may get a message on reboot saying that the system was not cleanly shutdown.

[JamesGV]

Ulric

...before we blame a fellow worker.

In Swissair 111, the SATCOM logged on.
There was no "log off" data available from Inmarsat.
The log on was a "renewal".
The time of the "log off" could not be determined.
It was assumed this was due to a "power interruption".

That's the record. I merely am repeating it.

[Vinnie Boombatz]

@Propduffer on 9th Jul 2014, 00:23--

John Zweck, an Australian Ph.D. in math now teaching at UT/Dallas, has posted satellite trajectory data in 1 second steps at:

Aqqa on MH370 (http://www.aqqa.org/mh370.htm)

About 2/3 down the page:

"The satellite position data from STK in 1 second time steps between March 7th 1100Z and March 8th 1100Z are available in the ECEF (fixed) frame and the WGS 84 frame. The LLA version includes latitude, longitude and altitude (geodetic height) above the WGS 84 ellipsoid and the LLR version includes latitude, longitude and radius from the centre of the WGS 84 ellipsoid."

Followed by links to files in several different formats.

Author's home page:

John Zweck (http://www.utdallas.edu/~zweck/)

[Ulric]

I'm sorry, I don't mean to imply blame. It's just that the behaviour of systems is different when they are shut down than when they are unplugged.

[sk999]

When the Signaling Unit Logs for 9M-MRO were released, Inmarsat described the BTO as being the round trip delay "relative to a terminal at a nominal fixed location". There were two different interpretations as to where that terminal sat - one had it at a fixed point on the earth, the other had it at a fixed point from Perth (and thus acting as a single constant offset to the round-trip delay.) These two interpretations led to different predictions for ping elevation angle as a function of time, up to 2 degrees.

The first interpretation seemed to make sense because its predictions matched well with the "7 Data Point" graph that showed ping angles for several times during the flight (aside from a puzzling constant offset in angle.) At the time, this graph seemed to provide the most reliable ping angles.

However, it is now clear that this first interpretation is wrong. Equation 1 in the ATSB report can be rewritten to show that the BTO is, indeed, just the round-trip delay plus a constant bias. Tables 1 and 2 of the ATSB report reinforce this view. Table 2, in particular - BTO Calibration at Kuala Lumpur - shows that the BTO and path length were both increasing at a time when the plane was sitting on the ground. The increase in path length was consistent with the increase in round-trip path between Perth and Kuala Lumpur in Table 1, and was due mainly to the satellite-Perth leg.

As a further check, I recomputed ping angles for both the portion of the flight where we have ADS-B data and for the later portion where one can make use of the ping ring plots in Figure 18 of the report. I get consistent agreement between these ping angles and the newly-interpreted BTO, with a spread of 0.2 degrees in elevation angle. These angles differ from those on the 7 Data Point graph by a peak-peak of 2 degrees - ouch!

Thus, we conclude that the "7 Data Point graph" is flawed. Perhaps it was drawn in a rush by someone for one of the family briefings. Its purpose was to show that the plane did not go to Beijing, for which high accuracy was not needed, but was not intended for anything else.

Duncan Steel's Ping Ring radii in message #549 posted on April 5, 2014 are derived from the 7 Data Point graph and thus are wrong. Unfortunately, Richard Cole's BTO model is also the wrong one, and his Appendix B distances are likewise wrong.

For what it's worth, I was following the wrong interpretion as well, until being confounded by the Kuala Lumpur BTO measurements. Since there is interest, I will give my values for BTO and such in a follow-on message.

[sk999]

Here are the ping ring data. Ang is the elevation angle. Ang and Arc are computed from BTO using a spherical Earth model. (I do convert geodetic to geocentric positions where appropriate.) The "Measured Ang" is either computed based on the known longitude and latitude at the time during the early phase of the flight or read off of Fig. 18 in the ATSB report. For the latter, I actually create my own graphics with ping rings overlaying an Earth map at the appropriate place and compared them with the Google Earth version in the report.

UT BTO Ang Arc "Measured" Source of "Measured"
deg km "Ang"

16:30 14933 46.90 4134 46.9 long = 101.69, lat = 2.72
17:07 15620 45.69 4253 45.6 long = 102.79, lat = 5.23
18:25 12520 53.31 3504 53.2 Fig 18, ATSB report
18:27 12533 53.28 3507
18:28 12490 53.39 3496
19:41 11500 56.17 3226 56.1 Fig 18, ATSB report
20:41 11740 55.41 3298 55.4 Fig 18, ATSB report
21:41 12780 52.44 3589 52.3 Fig 18, ATSB report
22:41 14540 47.86 4039 47.7 Fig 18, ATSB report
24:11 18040 39.87 4834 39.8 Fig 18, ATSB report
24:19 18400 39.10 4911

Long, lat are geocentric. 17:07 position taken from ATSB report.

[Propduffer]

@SK999
I don't know who you are and you only have four posts here, but everything you say checks out and you just saved me the trouble of contacting Richard to see what his take on all this is. So I'll go with your number for the 18:28 ring distance: 3496km = 1887.7nm rounded to 1888 nm. For the Satellite location at that time, my location checks with Vinnie Boombatz's link: 18:28:15 at 1.5691n 64.5286e.

So there we have it, AFIK everything is reconciled. We now have a solid Longitude to plot from at 18:28, and the Latitude can't be very far from MEKAR, the southernmost Latitude possible is 6° N - the FIR boundary, (we can be virtually certain the plane didn't cross into the Indonesian FIR there.) The northernmost Latitude is more open to debate; some people here apparently believe the plane went as far north as IGOGU or IGREX, I see the path most likely being IGEBO ---> POVUS , maybe as far north as NOPEK - but that's a separate subject.

So thanks Ulric, Vinnie and SK999 along with MM43 who started this thread.

[mm_flynn]

Are you sure? While we all know PCs routers etc have log files which almost always log a shutdown request, these are only visible if you have access to the PC (or wreckage in this case). I am pretty certain there is nothing detectable outside the aircraft that lets you know if the RT, transponding or ACARS was powered down at the device or lost power. I don't have any specific details on the Satcom, but all of the information so far suggest the link communications are a logon/authenticate and then periodic status checks, but no expectation of a logoff or shutdown message when the system is powered down normally. So that to gives no clue as to why it stopped responding.

I would certainly expect the FDR to be able to answer many of these questions.


In any event, it is moot as I don't think there is any sensible scenario where someone is carfully following a power down protocol.

[sysconfig]

That sounds like you are done and everyone can move along now nothing to see anymore..thanks for your help

You were quite certain based on one set of KLM that you had a done deal.., and asked for agreements by everyone. Until you were given the set corrected and now you are are sure as God made little green apples that it is a virtual..certainty...

almost a fact..is not the same even as almost the truth..But thank you for your help and take care too.

This man says in item 8 there is a significant degree of inaccuracy..in the whole ping matter to begin with.
Malaysian MH370: SATCOMS 101 (Part One) | Air Traffic Management | Air Traffic Management - ATM and CMS Industry online, the latest air traffic control industry, CAA, ANSP, SESAR and NEXTGEN news, events, supplier directory and magazine (http://www.airtrafficmanagement.net/2014/03/mh370-satcoms-101/)
I am just curious if there is any information how defective inmarsat equipment has behaved in the past..spurious bursts ..as they do have a return defective equipment policy and secondly can a defect like that collapse a planes electronics ..


Thank you

[BuzzBox]

JamesGV:
As a way of "resolving" this "log off and Log on" issue which is believed to have occurred at "some time prior to 1825 UTC" and "at 1825 UTC", you are saying with your time on T7's it is possible to "force" a "log off" without de-powering the system...and without leaving the flight deck.

Assuming you did this. The system has power but is "off line", how would you "log on" again ? Would this be something that would be "automatic" ? Or a "manual" function ?

Sorry, I neglected to answer your question. Once logged off, the flight crew would need to log on again through one of the flight deck CDUs. There are two types of log on available - auto, where the system will log on to the best available satellite; or manual (constrained), where the system can be constrained to log on to a particular satellite. I could be wrong, but I think that once the system is manually logged off, it will not log on again until one of those options is selected.

Note: I am only trying to explain how the system works on the flight deck. I am not inferring what may or may not have happened in the case of MH370.

[ItsMeFromEarth]

I fly the B777 and SAT COM LOSS is very common, it dates back from B777 rolled out.
After a while datalink will be lost too, then com is re established, datalink follows soon after. This happens several times during any flight. I am amazed nobody mentioned it.

[JamesGV]

An issue is... there is no notification if ACARS "dumps".
Post ETOPS after oceanic entry, if you don't update, you won't know.

And SATCOM can (and does) go "offline".
The next log on is a "renewal" not "initialisation".

If VHF/HF is required and not attained (as can be in an incident), what do you report ? You simply can't.

Is there an AUTO INIT issue on the B777 (all series), especially earlier series?

[Propduffer]

@sysconfig
In the same vein of sarcasm I thank you for your comment.
That sounds like you are done and everyone can move along now nothing to see anymore..thanks for your helpBut it should be pointed out to you that the above exchange was needed in order for everyone to "move on" with the same point of reference.

Up until this discussion, there had been no way for any but the most diligent and up to date sat engineer types to know the location of the plane at 18:28, because of a mistake that had been made early on. If you will read sk999's post on the previous page you will learn that Duncan Steel's Ping Ring radii are wrong, as is, Richard Cole's BTO model, and I dare say that without the above conversation mm43's graphic on this page would have carried forth the same error.

I'm glad to see that you made it as far as two posts, but it seems that it's time for you to get a new handle (probably once again.)

[Hyperveloce]

:ok: very nice map mm43, as usual. what is the average ground speed of this great circle trajectory ?

I tend to agree with OleOle when he suggests that the 18:40 BFO value of the C-channel Rx (during the Call Progress - Test ) around 90 Hz is strongly indicative of a south bearing at 18:40. It may allow the A/C to fly around the Indonesian air space, crossing it only in a tangential way (flying around Banda Aceh). I am running a new MC simulation assuming the south turn is one minute before 18:40 and taking in account the 88 Hz BFO value at 18:40.

[mm43]

...what is the average ground speed of this great circle trajectory ?

The following are the basic parameters compiled using the Vincenty Ellipsoid formulae:-

Start of track = 6.9752N 93.1072E, 172.4814T 131.144NM from VOCX
End of track = 29.8542S 98.4167E
Initial HDG = 172.3201T
Final HDG = 171.2092T
Dist = 4114660m 2221.739NM
Elapsed Time = 4.6405HRS - 4H 38M 26S
Ground Speed = 478.766KTS

Sorry about the "accuracy";)

Someone else asked for positions crossing each ring. I'll calculate those when I get some time.

As I've previously said, the graphic is just speculation based on the ability to derive a GC track at around 0.84 Mach. With regard to the southward turn at 1830, I assume that any turn or climb taking place at that time could provide the apparent BFO change.

[Propduffer]

I have created a Google Earth KMZ file with accurate rings and track plots at 20 kt intervals from 490kts to 410 kts and projected where 390 and 370 kts would fall on the 00:11 ring. ( I didn't include the 00:19 ring because I felt it would add more clutter than information.)

I chose a different route than mm43, I plotted what is probably the shortest route possible; I made the route through the strait pass south of MEKAR and from the 18:28 ring the route went direct to IGEBO and POVUS.

I included stuff from east of the 18:28 line, but the purpose of making this kml file available is for the 19:28 line and the southern route.

The KMZ file is at: https://www.dropbox.com/s/jz4d1j674o8g2a3/MH370.kmz

Pictures of the Google Earth presentation are at:
http://i1349.photobucket.com/albums/p753/sfojimbo/7-8POVUSTurn_zpsa1a0ba64.jpg

http://i1349.photobucket.com/albums/p753/sfojimbo/7-8SouthernRoutes_zps31397ae3.jpg

[OleOle]

. With regard to the southward turn at 1830, I assume that any turn or climb taking place at that time could provide the apparent BFO change.

Yep, the measurement has the problem of all the BFOs: It could be strongly influenced by vertical speed. In this case, if it the low BFO had been caused by vertical speed, it would have been caused by a vertical speed away from the sat, i.e. a descent.

PS: OTOH looking at appendix 1 of the raw data again: The BFO constantly lies between 86 and 90 in the one minute interval between 18:39:55 and 18:40:56. That is indicative of either a constant descent during this minute or a southerly course. To me the latter has higher probability

[Propduffer]

@Ole
It wasn't descending at that stage of the flight, but I believe it was headed SSW.

I have looked at the possibility of the flight having hugged the FIR boundary again and I now see that that looks very unlikely - if it did, it would have passed within 5 nm of Pulau We Island, which would have put it in Indonesian sovereign territory.

[OleOle]

The main problem I see with the 18:40 BFO is, that it occurs in the C-Channel and not in the R-Channel as all other BFO measurements. Yet the BFO for the 23:14 ground to air call also is in the C-Channel and aligns nicely with the "neighboring" R-Channel measurements, so a big difference between R- and C-Channel BFO doesn't seem to exist. How deterministic this behavior is for this particular terminal we don't know (inmarsat probably does).

[AAKEE]

I am not aware of any substantiated report that any system on MH370 was 'shut down'. All we know was that various systems were no longer responding at certain times. We also know that the satcom was 'off' at least twice by the fact that it logged on twice. 'Off' in this sense means that the aircraft terminal was not in a logged on status (due to loosing the signal, being shut down, having power lost, etc.) but doesn't tell the reason it was 'off'.

we have a last Transponder return time, a last ACARS message time, a Last RTF time, several missed phone calls times, but not a single logoff or power down message. Nor do I believe any of those systems generate such a logoff messages.

Early efter disappearance it was said that it was shut down. I'm finding it hard to know the status of how the investigation looks at this today tough...

Final words from Flight MH370 came after systems shutdown | The Japan Times (http://www.japantimes.co.jp/news/2014/03/17/asia-pacific/final-words-from-flight-mh370-came-after-systems-shutdown/#.U7-5OvNvkuU)

Authorities have said someone on board the plane first disabled one of its communications systems — the Aircraft and Communications Addressing and Reporting System, or ACARS — about 40 minutes after takeoff. The ACARS equipment sends information about the jet’s engines and other data to the airline.
Around 14 minutes later, the transponder that identifies the plane to commercial radar systems was also shut down. The fact that both systems went dark separately offered strong evidence that the plane’s disappearance was deliberate.

[OleOle]

... that the BFO value at 22:41 indicates that the airplane is heading almost due east at that time ...

Hi Gysbreght,

I tried to double check your above conclusion but come to disagree.

To my interpretation in your graph the red line (calibration) represents

DeltaFdown + deltaf_sat + deltaf_AFC + deltaf_bias

from equation (2) page 55. The difference between red and blue line then represents the remaining

DeltaFup + deltaf_comp

At 22:41 this difference between red and blue line happens to be almost zero.

In richardC10's nomenclature DeltaFup + deltaf_comp is:

DeltaFup + deltaf_comp = D1aircraft+D2aircraft + D2Satellite

with
D1aircraft+D2aircraft : error remaining after imperfect compensation (miscompensation)
D2Satellite ............ : Doppler component caused by sat's motion towards aircraft position

That means D1aircraft+D2aircraft and D2Satellite are of equal size but different sign at 22:41. For an assumed position of 15S 98E at 22:41 I calculate D2Satellite as 3.9 m/s or ~21Hz.

Then the miscompensation also was ~21Hz which leads me to a guestimate of 320+-100 kts for the southward component of the speed.

[Shadoko]

Blog - Cautious optimism in search for MH370 (http://www.atsb.gov.au/infocus/posts/2014/cautious-optimism-in-search-for-mh370.aspx)

What is really new is that now you can have some answers from "them"! Very kind!

[tawnybird]

Whilst we don't know how the 'investigation' looks at this today, we do know that we can't definitively say that ACARS was disabled (neutral term) before the last transmission.

We only know that ACARS did not make it's expected transmission at 17:37. This would place the time of disablement between 17:07 and 17:37.

There is some thought that the disabled transponder event should have triggered an ACARS message (can anyone confirm or refute this?). If this is true, then we would be further able to deduce that ACARS was disabled between 17:07-17:21. However, this would still not definitively place it before the 17:19 last transmission.

[AAKEE]

Is that right ?

While being on the non civil side of aviation, I don't know all about ACARS. But from what I read, there supposed to be a ACARS over VHF also ? These logs we didnt se?

For the satcom/Inmarsat logs, it seems like ACARS did report stuff on a quite regulary basis, not more than 16 minutes between statusreports ?

[Rev1.5]

The fact that both systems went dark separately offered strong evidence that the plane’s disappearance was deliberate.

Or there was a catastrophic event that was gradually knocking out system after system?

[olasek]

"Catastrophic" events usually don't take 14 minutes ....
Probably it is difficult to come up with a realistic failure scenario that would explain all the events how they unfolded in sequence. This is area strictly for experts in the aircraft systems.
By the way, this is old news, it is based on a March article, I think much more relevant to "human tampering" is the article quoted in #11270.

[mm_flynn]

The fact that both systems went dark separately offered strong evidence that the plane’s disappearance was deliberate.

Or there was a catastrophic event that was gradually knocking out system after system?

We don't know the systems went dark separately. We do know.

There was an ACARS air to ground message at 1:07
There was an RT transmission finishing about 1:19:30
There were transponder responses up to 1:21:13
There was no check in call (expected about 1:20)
The next expected ACARS message ( around 1:23-1:37) was not recieved
There was no response to a ground to air ACARS message at 2:03 and 2:05
There was a satcom login by the aircraft at 2:25
There was a ground to air phone call initiated and connected to the aircraft, but not answered at 2:39
There than followed hourly short status message (1 packet each direction)
at 7:13 there was another ground to air phone call initiated and connected to the aircraft, but not answered
finally the remaining pings and the last logon

To me, this would look like everything stopped at around 1:20 and then just the satcom terminal itself came back on line at 2:25.

[Shadoko]

Thank you all for these explanations.

Gysbreght: I don't understand yours:
"Therefore, if the difference between red and blue line is almost zero, the calibrated BFO is almost zero, and hence the north-south component of the airplane groundspeed is almost zero, i.e. the track is almost due east or west."
IMHO, no Doppler between a/c and sat happens when a/c and sat have parallel pathes and same speed, or concentric pathes and speeds proportional to the pathes radii.

[Ian W]

Whilst we don't know how the 'investigation' looks at this today, we do know that we can't definitively say that ACARS was disabled (neutral term) before the last transmission.

We only know that ACARS did not make it's expected transmission at 17:37. This would place the time of disablement between 17:07 and 17:37.



I do not know if the ACARS logs have been made available. But the ARINC 620 datalink specification has a 'logon/logoff' protocol. If the ACARS 'failed' or had its CB pulled, then there would be no logoff protocol messages. If the ACARS was tidily switched off then it would have followed the log off protocol.

The early statements implied that it had logged off, meaning someone who knew how to find the ACARS menu had switched ACARS off.

[Ulric]

which would have put it in Indonesian sovereign territory

Is that a constraint? I don't remember seeing any firm data either way?

[Ulric]

The Inmarsat document has a definition here
http://www.inmarsat.com/wp-content/uploads/2014/03/Inmarsat-Differential-Doppler-Study.pdf

[OleOle]

Hi Gysbreght,

I tried to understand your equation from post #21. Unfortunately I failed.

But I can try to better explain my understanding :

I think up to the term DeltaFup + deltaf_comp we have a common understanding.

Of course it's kind of artificial to split DeltaFup into (D2aircraft + D2Satellite), yet I don't see any reasons not to do so. In contrary it's very helpful because the terminal applies deltaf_comp to compensate for D2aircraft only. The terminal doesn't know about D2Satellite because it supposes the sat to be stationary. Actually splitting into these two components is the natural thing to do, if you take the earth at rest as frame of reference.

With DeltaFup = (D2aircraft + D2Satellite) we can write:

DeltaFup + deltaf_comp = (D2aircraft + D2Satellite) + deltaf_comp = (D2aircraft + deltaf_comp) + D2Satellite

For a given aircraft position D2Satellite can be calculated from the angle between line of sight and sat velocity vector. The calculation is the same as for D3 aka DeltaFdown, only that instead of the coordinates of Perth the coordinates of the aircraft position are taken (and of course the L-Band Frequency).

Now deltaf_comp almost cancels out with D2aircraft, but just almost so, because the terminal's assumption of the sat position is not accurate (it supposes the sat position to be above the equator). That causes the terminal to calculate an inaccurate line of sight and thus an inaccurate line of sight speed.

(D2aircraft + deltaf_comp) is what I called "miscompensation" before. It depends on the offset of the sat from its geostationary position and on the aircraft velocity vector (mainly the component of the aircraft velocity that is parallel to the offset of the sat from its geostationary position i.e: its north/south component).

[Shadoko]

Gysbrecht: sorry, we don't speak about the same value. My remark was about "pure" Doppler between a/c and sat. I had not read you with enough attention...

Ole-Ole: how do you find a "D2Satellite" Doppler (Doppler between sat and a fixed point where the a/c is in my understanding) of 43Hz at 22:41 between sat and a fixed point 15S 98E?
In my simple model, I have a 3.96 m/s LoS speed for these data. Which give (to me!) -21.7Hz (for 1643MHz).

[tawnybird]

Ian,

Unfortunately the ACARS logs are 'under wraps'. Aside from the fuel load on board at 17:07, the absence or presence of a log off is also vital data in regards to the sequential timeline. This is under the purview of the criminal investigation, which is still open and ongoing.

I tend to believe that those 'early' reports vis a vis ACARS retain credibility.

If ACARS was tidily switched off, we then are almost certainly faced with an hostile act, either by crew or interloper. This is not pleasant to contemplate, but contemplate we must.

Along these lines (and I am in no way casting blame), the redundant FL350 transmissions AND the failure to read back the Ho Chi Min frequency loom much larger as true outliers, IMO.

[OleOle]

Ole-Ole: how do you find a "D2Satellite" Doppler (Doppler between sat and a fixed point where the a/c is in my understanding) of 43Hz at 22:41 between sat and a fixed point 15S 98E?
In my simple model, I have a 3.96 m/s LoS speed for these data. Which give (to me!) -21.7Hz (for 1643MHz).

Dyscalculia :O

Already fixed it. And the speed guestimate too : 36000/800 * 3.9 m/s ~= 160 m/s = 320 kts

36000 : Distance aircraft -> sat in km
800 : Sat offset from geostationary position in km

[Propduffer]

Ulric
Is that a constraint? I don't remember seeing any firm data either way?The Indonesians have stated that they didn't see it pass "over their territory." I have no reason to question that statement.

This also indicates to me that the person flying the plane wouldn't have wanted to enter their FIR unannounced.

[formulaben]

"Catastrophic" events usually don't take 14 minutes ....

Swiss Air 111.

[JamesGV]

South African 295

[Ian W]

Swiss Air 111.

Swiss Air 111 was able to communicate until the very last minutes. Had they not delayed to dump fuel on a relatively standard approach procedure and instead made an immediate emergency descent and landing they may even have survived.

To recap, under the MH370 catastrophe scenario:

MH370 is supposed to have had a catastrophic failure
AND
the crew not be able to use any of the aircraft systems to report the failure
AND
it took 10 - 15 minutes for these communications systems to fail
AND
be able to fly apparently under control back across Malaysia and around Indonesia then South with power to SATCOM and a working FMC for the next 6 hours

While everyone accepts the aphorism: Aviate, Navigate, Communicate - it has been my experience that even in extremely fast moving emergencies flight crew do tell their current controller or anyone on emergency frequency that they have a problem. It only takes one button to be pressed. It is possible to have catastrophic failures that prevent communications but they would appear to be ruled out by the subsequent conditions above.

[mm_flynn]

I do not know if the ACARS logs have been made available. But the ARINC 620 datalink specification has a 'logon/logoff' protocol. If the ACARS 'failed' or had its CB pulled, then there would be no logoff protocol messages. If the ACARS was tidily switched off then it would have followed the log off protocol.

The early statements implied that it had logged off, meaning someone who knew how to find the ACARS menu had switched ACARS off.

Ian,

Are you certain there is actually a logoff ACARS message (as compared to 'logging off' in the airplane as a planned way of causing the system to stop sending messages). I can't find any indication that such a message type exists. Equally, I can't see a 'logon' ACARS message (as compared to the satcom logon). The closest message type to logon seems to be Link Test.

Also, the 1:07 message, based on the Inmarsat log, looks to be 'large' (30 signalling units). What appears to be the ACK is only three signalling units. I would expect a logoff message to be similar in size to the ACK. That coupled with the fact that no official documents has said 'the last ACARS message at 1:07 was a log off message' leads me to believe the early comments of ACARS being 'turned off' were imprecise comments that should technically have been 'ACARS stopped transmitting expected messages or responding to interrogations some time after the last routine message at 1:07'.

One can easily understand how a briefer in the heat of the moment would use version one vs version two.

[JamesGV]

Ian W

With Swiss111 there was an INMARSAT "log on" (renewal).
No "log off" was registered. No prior "log off" could be determined.

The report suggests there was a "power interruption".

[Alloyboobtube]

If it was an electrical fire could the crew have removed all electrics including batteries and even climbing to stall the RAT.
Fire stopped successfully but during re-initialisation of systems didn't notice loss of cabin pressure ,aircraft remains on last set heading.

[Ian W]

Ian,

Are you certain there is actually a logoff ACARS message (as compared to 'logging off' in the airplane as a planned way of causing the system to stop sending messages). I can't find any indication that such a message type exists. Equally, I can't see a 'logon' ACARS message (as compared to the satcom logon). The closest message type to logon seems to be Link Test.

Also, the 1:07 message, based on the Inmarsat log, looks to be 'large' (30 signalling units). What appears to be the ACK is only three signalling units. I would expect a logoff message to be similar in size to the ACK. That coupled with the fact that no official documents has said 'the last ACARS message at 1:07 was a log off message' leads me to believe the early comments of ACARS being 'turned off' were imprecise comments that should technically have been 'ACARS stopped transmitting expected messages or responding to interrogations some time after the last routine message at 1:07'.

One can easily understand how a briefer in the heat of the moment would use version one vs version two.

ACARS was only licensed and activated on VHF by the Malaysians it was not linked to INMARSAT SATCOM or to Boeing or Rolls Royce health monitoring. So the ACARS logon/logoff would be an ARINC/SITA protocol over VDL which I believe is ARINC 622 'ATS Data Link Applications over ACARS, or 618 Air/Ground Character-Oriented Protocol Specification.

[porterhouse]

,aircraft remains on last set heading.
Which it didn't happen in this case, aircraft did not remain on last set heading.

[mm_flynn]

ACARS was only licensed and activated on VHF by the Malaysians it was not linked to INMARSAT SATCOM or to Boeing or Rolls Royce health monitoring. So the ACARS logon/logoff would be an ARINC/SITA protocol over VDL which I believe is ARINC 622 'ATS Data Link Applications over ACARS, or 618 Air/Ground Character-Oriented Protocol Specification.
Ian,

two points.
1 - I can't see described in any of the protocols a logoff message (nor for that matter a log on) at the ACARS level (i.e. a message that would indicate that the ACARS system was logging off, rather than that the underlying transport layer was logging off). Equally, I am certain that the VHF transport layer does not log on or logoff as it is just open radio spectrum.

People have speculated the authorities have some indication the ACARS was 'logged off in a conscious deliberate way'. I can see no basis in the technology being used that would allow that possibly to be concluded with respect to the ACARS system. As such, I believe it is certain to have been a turn of phrase used by someone in an early briefing rather than anything relevant.

2 - People have represented that Malaysian Airlines did not use the satcom for ACARS, yet the official preliminary report declares it did, and that the 1:07 ACARS message was transmitted via satcom link, and finally, there is a satcom message originating from the aircraft at just before 1:07 and finishing just after 1:07.
It was later established that the transmissions from the Aircraft Communication and Reporting System (ACARS) through satellite communication system occurred at regular intervals starting before MH 370 departed Kuala Lumpur, Malaysia at time 12:56:08 MYT and with the last communication occurred at 01:07:49 MYT.

I had originally assumed the attribution of link layer (satcom vs VDL) was an error. However, at this remove, if it was an error, it would have been corrected. Also, it exactly matches the timing of the Inmarsat final packet acknowledgement from the aircraft (to the exact second).

I therefore believe, subject to some conclusive evidence of MA non-use of Satcom for ACARS, that MH370 was using Satcom as its ACARS link layer. And therefore, the reason we have no VDL ACARS logs for MH370 is because it wasn't using VDL, not because they exist but have not been released.

[portmanteau]

mm_flynn.

12:56:08 myt is not just before 01:07:49 myt. it is some 12 hours before in the early afternoon. 00:56:08 myt would be just before. note that aircraft took off at 00:41 myt.

[mm_flynn]

The 12:56 time was not implied as shortly before 1:07. The Preliminary Report clearly implies a number of routine messages for an appreciable time prior to the last message prior to the loss of transponder contact. The quote is directly from http://www.dca.gov.my/MH370/Preliminary%20Report.pdf.

I was focused on the 'last ACARS message time' which matches exactly to the second with the last Satcom message (prior to the pings, missed calls, and two logons post IGARI). And which, I believe, establishes clearly that the Satcom message is from the ACARS system and that MH370 was using Satcom and not VDL to transmit its ACARS messages.

I have no idea to what the 12:58 time refers. It may be when the aircraft was first powered up that day (i.e. an earlier flight) or it may be some time near the Satcom initial login of 00:13 but documented incorrectly, or it could be some entirely different event that the preliminary investigators considered relevant in this context.

[Ornis]

http://www.dca.gov.my/MH370/Preliminary%20Report.pdf Search and Rescue (SAR). It was later established that the transmissions from (ACARS) through satellite ...

Safety recommendations. While the aircraft had the necessary communication equipment to provide information on its location, the last ACARS message ...

It is recommended that the ICAO examine the safety benefits of introducing a standard for real time tracking

This doesn't quite make sense to me: asking for something DCA claims existed. And why would Acars be via satellite where VHF was available?

[expat400]

Ian W

"Swiss Air 111 was able to communicate until the very last minutes. Had they not delayed to dump fuel on a relatively standard approach procedure and instead made an immediate emergency descent and landing they may even have survived."

This is not correct. Read the report.

"Based on these factors, it is evident that even if the pilots had attempted a minimum-time emergency diversion starting at 0114:18, it would have been impossible for the pilots to continue maintaining control of the aircraft for the amount of time necessary to reach the airport and complete a safe landing."

[BuzzBox]

Ornis:
This doesn't quite make sense to me: asking for something DCA claims existed. And why would Acars be via satellite where VHF was available?

They're asking ICAO to examine a 'standard' for the real time tracking of aircraft. That is something that does not currently exist. There are systems available that can track aircraft outside radar coverage in real time, such as ADS-B/C, however only a few ATC centres have the capability and not all aircraft are equipped. Airline aircraft routinely transmit position information and engineering data via ACARS, but that information is not used to 'track' the aircraft in real time.

In the case of MH370, the aircraft had the capability to transmit its position via ADS-B and ACARS. However, that all stopped when the transponder and ACARS both stopped transmitting. I guess an ICAO standard would have to require an independent system that can't be tampered with or disabled by anyone onboard the aircraft.

Regarding the second part of your post, it all comes down to cost and availability. In some areas the airline's provider (SITA or ARINC) may not have the necessary VHF coverage, and in others (e.g. Japan) it is actually cheaper to route ACARS messages via satcom rather than VHF.

[Hyperveloce]

I wonder if such a bunch of trajectories would draw the attention of the Indonesian airspace surveillance chain (the location of the unidentified A/C matters versus a RoI, but also its radial speed ?), the best fit trajectory is slightly (a few NM) off Banda Aceh:

https://drive.google.com/file/d/0B3srJsCh7WAtY3BXOFRQaWJfS28/edit?usp=sharing

The 18:40 C-channel BFO 88 Hz value is not an outlier for the predicted BFO enveloppe of my MC simulation and it does not degrade the BFO error optimization (as well as the 3rd handshake around 18:28), the following result (30 000 runs) is for a south turn occuring 6 min after 18:28, early trajectory not constrained by the ADS-B and radar tracks:

https://drive.google.com/file/d/0B3srJsCh7WAtc2FtcXdHcks5cTA/edit?usp=sharing

The inclusion of this new BFO value does not alter the most probable area (between 28 and 32°S) along the last ping ring, the best fit trajectories last south leg remains at a relatively low true ground speed (around 410 kts) ending round 30°S ...and curvy (not a a great circle or a rhumb line):

https://drive.google.com/file/d/0B3srJsCh7WAtaGVncnZyWUhQVW8/edit?usp=sharing

It does not modify the mean BFO error for each handshakes (as a function of time and true ground speed):

https://drive.google.com/file/d/0B3srJsCh7WAtSUNWeVFEU2VLbm8/edit?usp=sharing

The inclusion of a new BFO data point or of a delay (from 18:28) for the south turn does not seem to impact the simulation end result, what really does seem to have an impact is whether I allow the 1st part (till 18:28) of the trajectories to fluctuate & drift (resulting in a wider variety of trajectories and of injection points for the last south leg) or if I constrain them to the ADS-B & radar tracks (which seems to shift the most probable area more than 1° toward the south).

[Propduffer]

I wonder if such a bunch of trajectories would draw the attention of the Indonesian airspace surveillance chainIMO anyone who thinks 9M-MRO overflew Indonesia is so focused on their mathematics that they've lost sight of reality.

[Ulric]

IMO anyone who thinks 9M-MRO overflew Indonesia is so focused on their mathematics that they've lost sight of reality.“When you have eliminated the impossible, whatever remains, however improbable, must be the truth.”

Arthur Conan Doyle

[Hyperveloce]

IMO anyone who thinks 9M-MRO overflew Indonesia is so focused on their mathematics that they've lost sight of reality.

Are you suggesting that the official investigation team has lost sight of reality when it produces such trajectories ?

http://www.airtrafficmanagement.net/wp-content/uploads/2014/05/MH370-Radar-and-Satellite-Tracks-20140501.jpg
BTW, in another life I have been sitting at a surveillance radar console (onboard an anti aerial frigate) and both the position and the speed vector orientation (versus the protected area) of an A/C would draw my attention.

[Propduffer]

Are you suggesting that the official investigation team has lost sight of reality when it produces such trajectories ?The Official Investigation team doesn't release information in GE format.

Are you saying that graphic is from the Australian Government?

[Propduffer]

“When you have eliminated the impossible, whatever remains, however improbable, must be the truth.”You must have left part of your post out. That saying doesn't have any meaning in the present context that I can see.

[Hyperveloce]

The Official Investigation team doesn't release information in GE format.

Are you saying that graphic is from the Australian Government?

No, I am saying that these graphics were released as part of the MH370 preliminary report by the office of the chief inspector of air accident, Malaysian MoT:
MH370: Transport Ministry releases preliminary report, cockpit-air traffic control recordings - Nation | The Star Online (http://www.thestar.com.my/News/Nation/2014/05/01/mh370-prelim-report-cockpit-tower-recordings-released/)
Had the MoT lost sight of reality when it suggested such "possible routes" ?
Maybe that routes skimming around the Indonesian airspace are conceivable ?
Which kind of route (compatible with both the BTO and the BFO) would you suggest between 18:28 and 19:41 ?

[Propduffer]

Had the MoT lost sight of reality when it suggested such "possible routes" ?The Malaysian Transport Ministry has issued little but disinformation since March 8th, so it isn't a matter of them losing sight of reality, it's more a matter of them throwing a smoke screen over reality. Do you really take their "information" seriously?
If they would just release the 18:07 ACARS information, along with their actual radar track, the search for the resting place of 9M-MRO would probably become unnecessary.

Which kind of route (compatible with both the BTO and the BFO) would you suggest between 18:28 and 19:41 ?I have posted a link to my KML file (post #210) which contains a detailed plot that I consider a possible if not probable approximation of 9M-MRO's flight path past Pulau We Island.

I believe that my proposed flight path is supported by OleOle's BFO hypothesis, and it is not the only solution bypassing Indonesian territory that fits the current understanding of BFO information. There is nothing to support the flight passing over Indonesian territory, yet there is strong evidence (Indonesian statements) to the contrary.

[BuzzBox]

There is nothing to support the flight passing over Indonesian territory, yet there is strong evidence (Indonesian statements) to the contrary.

I'm playing devil's advocate here - what if the Indonesians aren't telling the whole truth? Governments in that part of the world are a little touchy about national security and air defence. Would the Indonesians admit to the world that an unidentified aircraft flew over their territory undetected?

You seem to have unswerving faith in the Indonesian statements, yet discredit the information released by the Malaysians as a 'smokescreen'.

[Ulric]

I don't see any reason for treating one set of statements differently to another. In God we trust. All others must bring data.

http://www.lockheedmartin.co.uk/us/news/press-releases/2012/november/lockheed-martin-and-pt-cmi-teknologi-complete-radar-production-r.html

[Ian W]

Put another way - bandwidth.

In some areas the data link from the VHF ground station may have a very low bandwidth with limited capacity and slow speed. The satellite link will have the necessary high-speed links from ground station to the main base system. All that is necessary is to ensure that the satellite bandwidth is sufficient for the maximum traffic levels.

While it would be very expensive to increase satellite bandwidth it would involve fewer agencies and be more easily achieved on an international scale than for national agencies.

Both INMARSAT and Iridium (with their Next constellation) were in the process of increasing their available bandwidth before MH370. Both will be providing broadband links to their customers in the next few years.

The customers will be the ones having to upgrade their existing SATCOM systems to obtain the increased bandwidth.

However, most transoceanic aircraft are already capable of ADS-C over SATCOM that will work on current low bandwidth systems, and all they need to alter is the reporting rate. As was recommended by the French BEA after the AF447 crash.

[Pontius Navigator]

Ian, I am a tyro there but surely the aircraft narrow or slow connection is not a problem for the satellite companies. It must mean their uplink is slow but not a problem for the satellite.

[Ornis]

[mm flynn] 2 - People have represented that Malaysian Airlines did not use the satcom for ACARS, yet the official preliminary report declares it did,

Did MAS subscribe to a satellite service for acars and was data sent to MAS? Boeing, RR?

Thank you for sharing your knowledge.

[oldoberon]

my understanding was they subscribed to RR but VHF only.

The satcom would be on for the satphone so it would look for a logon, how you stop the RR acars going via sat when you don't pay for it I have no idea, but suspect something like lack of a data key would lead the satellite to ignore it.

[AreOut]

anyone comment on this?

(the post from duncansteel.com) :

Byan C.
2014/07/03 at 08:22

I woke up wondering if Duncan thinks I’m some kind of crackpot, so I thought I should send you a clearer explanation. If R represents satellite to plane distance, s, p, r reresents the satellite, plane, and reference satellite respectively. Numbers 1 and 2 represents two points in time and x, y, z are the coordinates. Then,
(x-xs)^2 + (y-ys)^2 + (z-zs)^2 = Rs^2
(x-xr)^2 + (y-yr)^2 + (z-zr)^2 = Rr^2
x^2 + y^2 + z^2 = r^2 where r = earth radius and recognizing that
xs^2 + ys^2 + zs^2 = (r+H)^2 where H is the satellite height above the earth. The same applies to the reference satellite.
Converting BTO data to velocities yield:
V-Vsp-Vps = Vpr
Assuming a constant satellite velocity the Vsp + Vps = Rs2 – Rs1
Similarly, Vpr = Rr2-Rr1 or Rr2 = Rr1 + Vpr
This is where I got into trouble. Since Rr2 depends on Rr1 if I don’t check the methodology using the course defined by the BTO data, I don’t get a good fit to the BTO data at 19:40. After solving, I plug the plane velocities derived from (x2-x1)/(t2-t1) to a program the solves based on the distance traveled between 2 ping rings. This routine calculate the BTO values based on the actual Vsp at time t. Comparisions indicate that my solution using average Vsp values is yielding reasonable results. To calculate values using the actual Vsp value at time t involves solving 10 equations with 10 unknowns. The good news is that the solution would reduce to a quadratic equation. It is interesting that the solution for the equations based on distance traveled yields 2 roots with different latitudes. The solution to the equations based on velocities yields 2 roots with different longitudes. 1 to the west of the satellite and 1 to the east, effectively resulting in 1 solution. As a check, I plugged in Ra and Rs values for the northern and southern paths, it duplicates those paths.

Imagine my shock after I got this debugged (I keep reversing signs in my vector math) and I get a result that indicates the plane took a large circular path to the south heading eventually back to KLIA. I almost fell off my chair. I was not expecting this. I have to admit that I have not been that excited about anything in a long time, hence my need to hurriedly share this with someone.

I’m sharing this at this time because I believe the actual solution will be difficult to achieve. I could sove the 10 equations, but developing reasonable flight paths is beyond my ability.

I indicated that I thought we should keep this private because if there is a 3rd flight path, this will cause hugh controversy and we need to be certain before presenting such a result. Perhaps, my results are just a fluke of the data, but since I achieved these results with absolutely no manipulation of the raw data, I feel like it needs to be pursued further.

[AreOut]

pardon, forgot to copy his autocorrection

"Byan C.
2014/07/03 at 22:34

I noticed a few typos in my last post. The most significant being that the differences in sat distances should have been divided by (t2-t1). On further reflection and trying a few different scenarios on my constant speed course calculator I’ve deduced the following.
If the radar data does not absolutely preclude a more north-western starting point(this matches the BFO) data better, then you need to proceed south for the next 4 ping rings at about 450 kts. The curious thing is that you can set a slightly north easterly course fo the last ping ring and it does not seem to affect the BTO values. I now think that the more northerly starting point and the interesting fact that you can turn to the northeast on the last ping ring is why my results obtained from not assuming a plane sppeed appears to suggest a route roughly returning to KLIA."

BFO/BTO was also a typo

[Ian W]

I don't believe that understanding is correct. All available information says that the ACARS system used SatCom. I would be really surprised, having made that decision, MAS then decided not to send the engine data via the same link.
.

From way back in the thread.
Malaysian Airways did not subscribe to Rolls Royce or Boeing health monitoring. ACARS was set up to only operate over VHF and the only INMARSAT subscription that the airline had was for satellite phone.

ACARS was set up to make routine reports at 30 minute intervals.

There was no connection of any sort between ACARS and SATCOM this was disabled as the company did not want to pay the subscriptions.

[AreOut]

yeah he is a bit clumsy in explaining (probably not the lecturer but still good at math) but certainly makes point which has not been disproven yet

I have also mailed ATSB and got the same response, I doubt they even look at that given they probably get tons of crazy conspiracy mails on a daily basis, sorting everything out would take significant amount of time.

[Shadoko]

From "MH370 Data Communication Logs.pdf"
Page 36:
17:07 - Last Acknowledged Ground to Air DATA-2 ACARS Message. Link lost at sometime between here and 18:03:41.
7/03/2014 17:07:48.907 IOR-R1200-0-36ED IOR 305 4 R-Channel RX 0x62 - Acknowledge User Data (R-channel)
18:03 - No Response to Ground to Air DATA-2 ACARS Data. Link lost at sometime between 17:07:48 and here.
Page 37:
7/03/2014 18:04:29.413 IOR-P10500-0-3859 IOR 305 10 P-Channel TX 0x61 - Request for Acknowledge (RQA) User Data
18:04 - Last of five requests for Acknowledge to the Air DATA-2 ACARS Data at 18:03
18:05:11 - Data-2 Ground to Air transmission, automatic ACARS retransmission after expiry of 90 sec timer. Terminal is not responding.



IMHO, this indicates that ACARS transited by SATCOM. At least some of them. From the comments, these ACARS are about "AIR DATA-2". From memory, these data contains ??? and fuel remaining at 17:07 (all the contents had not been published).

[BOAC]

All I can see is the ground station REQUESTING the ACARS data. Posts 8869,8872 and 8873 (plus WSJ) state Malaysian did not send.

[Shadoko]

From "MH370_Considerations on defining_FactSheet.pdf" (ATSB 26 May):
The final ACARS transmission was at 17:07 UTC and provided location reports from the initial stage of the flight as well as a recording of the aircraft fuel remaining.
And from ""MH370 Data Communication Logs.pdf":
All the transmissions (RX and TX) between 7/03/2014 17:06:49.406 and 7/03/2014 17:07:46.905 (pages 30 to 36).

The fact that channels are assigned and user data acknowledged implies that the communication(s) were established. The SATCOM use and the known ACARS timing published are in same minute. So, my understanding is the data transmitted via SATCOM were, at least part of them, ACARS.
And voice by SATCOM use C channel.
But I can be wrong!

[Shadoko]

About the "circle path": I am afraid it is the result of oversimplification of the sat and the airplane relative trajectories.
Why using a constant speed for the sat? Inmarsat 3F1 data are available.
In my very simple model, I don't use spherical coordinates but Lat, Long and Altitude, because it is easier (for me) to "see" the data. It is not very accurate because "my" Earth is spherical, but just for "see" what can be deduced from the "faint" known data, I think it is enough.
I suspect the circling trajectory obtained by Byan C is from not taking in consideration the fact that all airplanes flying along a "straight" line are, in fact, flying on a circle around the Earth. If the simplification include flying in a (flat) plane, the only way to follow the BFO values is to "curb" the trajectory in the (flat) plane...

[Ulric]

Yes, I am not at all sure about the "circle path" because it seems to imply such variations of speed and heading. I am also unclear about what failure conditions could cause an aircraft to fly such a path.

[nick1austin]

Two lots of ACARS appears in the Inmarsat logs. You'll notice that the first message of each request is different.

The first request is:
17:06:49.406 ... R-Channel RX 0x22 - Access Request (R/T-Channel)
The RX indicates the direction of transmission which means that the aircraft had triggered the sending of ACARS data.

This is followed by a minutes worth of RX and TX messages until the final acknowledge at 17:07:48.907. I interpret this as the period when the ACARS data being sent and accepted. This means that ACARS-over-satellite must have been enabled.

The second request:
18:03:41.405 ... P-Channel TX 0x71 - User Data (ISU) - RLS
In this case the difference is that this is a TX. This means that the request came from the ground. My assumption is that someone (probably ATC) phoned either Boeing or Rolls-Royce to ask if they knew where the aircraft was and this request was triggered by Boeing/RR.

You'll notice that there are no RX messages following the second request so for some reason the aircraft was unable to respond. There are many reasons for a non-reply. It's possible that ACARS was turned off or unpowered. However ACARS-over-satellite needs needs a functioning SDU so it's also possible that this was unpowered.

[AreOut]

I'm quite sure there was no failure before 2:22 MYT, after that maybe but it's only speculation based on satcom restart (which doesn't automatically imply failure).

the thing is there are many paths that could fit BFO&BTO values within margin of error, especially since we don't have a clue if any of speed/altitude/heading were constant at all

P.S. http://www.duncansteel.com/ group has made the new statement

this figure is interesting to me

http://www.duncansteel.com/wp-content/uploads/2014/07/Interim-Report-Figure-1a.jpg

anyone tell me why they assume constant heading from 19:41 onwards for a flight that drastically changed its heading 3 times in first ~2 hours?

[BuzzBox]

My assumption is that someone (probably ATC) phoned either Boeing or Rolls-Royce to ask if they knew where the aircraft was and this request was triggered by Boeing/RR.

I think ATC would be more likely to contact the airline in the first instance, rather than the aircraft or engine manufacturer. The ground transmission may have been a text message from the airline to the crew asking them to confirm their position.

[mm_flynn]

From way back in the thread.
Malaysian Airways did not subscribe to Rolls Royce or Boeing health monitoring. ACARS was set up to only operate over VHF and the only INMARSAT subscription that the airline had was for satellite phone.

ACARS was set up to make routine reports at 30 minute intervals.

There was no connection of any sort between ACARS and SATCOM this was disabled as the company did not want to pay the subscriptions.

Ian,

You have said this several times. Yet it (the ACARS over satcom part! not necessarily the RR or Boeing monitoring) is wholly inconsistent with all of the official statements by the Malaysian CAA. Do you have an actual source or is it a repetition of a prior assertion in the thread?

[BOAC]

mm - if there is 'Acars Data' floating around as you say, where is it and what does it contain?

[mm_flynn]

The Malaysian CAA hasn't released the content of the data stream and I seriously doubt it is relevant. there will have been all of the normal messages and then the 1:07 packet will all most surely be a top of climb report or a routine monitoring report. But the CAA has declared that it was received, they declare it was received via satcom and Inmarsat logs show a set of data packets at exactly the same time.

The ACARS data is irrelevant, I am not in any way suggesting there is data post the 1:20ish loss of comms and transponder. The relevant point is, the earlier statements in the thread that MAS did not use Satcom for their ACARS are clearly not correct. It may well be true that MAS did not subscribe to RR or Boeing monitoring service - but that is irrelevant to the loss of this aircraft.

[BOAC]

It would still be useful to know what data was being sent from Acars via SAT as there should be no reason why the data should stop at loss of comms - that I can see. Eg if any nav data................?

[Ian W]

ITU to assist in real-time monitoring of flight data

ITU Membership calls for development of standards for aviation cloud

Geneva, 25 June 2014 – ITU has established a new Focus Group on Aviation Applications of Cloud Computing for Flight Data Monitoring. The group will study the requirements for the telecommunication standards to enable an ‘aviation cloud’ for real-time monitoring of flight data, including those for the protection, security and ownership of flight data and the technical mechanisms and policies to govern access to these data.
The formation of the Focus Group comes in response to the call from the Minister of Communications and Multimedia, Malaysia, Mr Ahmad Shabery Cheek in March 2014 urging ITU to develop leading edge standards to facilitate the transmission of flight data in real-time. Subsequently, an “Expert Dialogue on Real-time Monitoring of Flight Data, including the Black Box – the Need for International Standards in the Age of Cloud Computing and Big Data (http://www.itu.int/en/ITU-T/Workshops-and-Seminars/ccsg/expdial/Pages/default.aspx)”, held in Kuala Lumpur, 26-27 May 2014 with the participation of airlines, aviation bodies, avionics and ICT companies, service providers, civil aviation authorities, the International Civil Aviation Organization (ICAO) and other international organizations, issued a communiqué (http://www.itu.int/en/ITU-T/Workshops-and-Seminars/ccsg/expdial/Documents/communique.pdf) outlining a roadmap for the way forward.
Participation in the Focus Group will be open to all interests, including non-members of ITU, and it will work in close collaboration with ICAO, ICT solution providers, aircraft manufacturers, airlines and other standardization expert groups.
ITU Secretary-General Hamadoun I. Touré said that the call from Minister Shabery for an international effort to find solutions to monitor flight data in real time has been given top priority by ITU and its Membership.
“ITU has a long history of developing international telecommunication and ICT standards, policies and regulations and is offering to bring this competence to assist aviation,” said Malcolm Johnson, Director of ITU’s Telecommunication Standardization Bureau. “I applaud our membership for the urgency with which it is addressing this issue by responding so quickly to Malaysia’s call for ITU action.”
The new Focus Group will study advances in cloud computing and data analytics to develop use cases for the application of state-of-the-art data analytics and data mining techniques in real-time. It will develop technical reports to provide the foundation for standards-based aviation clouds. In close collaboration with ICAO, the envisioned reports will address questions surrounding the type of data to be transmitted and the periodicity and reliability of its transmission, as well as the mechanisms to enable data security and privacy and the prevention of data misuse. The reports will be the basis for the development of telecommunications standards providing security and providing interoperable and secure aviation cloud systems.
Note to the editor: ITU-T Focus Groups (http://www.itu.int/en/ITU-T/focusgroups/Pages/default.aspx) are formed in response to immediate ICT standardization demands, charged with laying the foundation for subsequent standardization work in membership-driven ITU-T Study Groups (http://www.itu.int/en/ITU-T/studygroups/2013-2016/Pages/default.aspx). Focus Groups are open to organizations outside ITU’s membership and they are afforded greater flexibility in their chosen deliverables and working methods.
For more information, please contact:
Sanjay Acharya
Chief, Media Relations and Public Information, ITU


See ITU to assist in real-time monitoring of flight data (http://www.itu.int/net/pressoffice/press_releases/2014/37.aspx#.U6wdz7VRSkM)

[Ian W]

I am wondering what other people think about the involvement of the ITU when the ICAO, RTCA, EASA, ARINC/SITA INMARSAT/Iridium, are already heavily involved and have experience of aviation requirements and knowledge of existing systems, communications and standards.

ITU may also be less aware of the niceties of CVR issues and FOQA data.

[MG23]

I'm guessing this will somehow involve X.25, sixteen layers of different protocols on top of each other, and about a billion configuration options where any misconfigured option will prevent you from talking to anyone else.

[Ulric]

I agree. I think it is highly unlikely that a ping coincidentally occurs at the furthest westerly extent of the track.

[Ornis]

From the website: About ITU. ITU is the leading United Nations agency for information and communication technology. For nearly 150 years, ITU has coordinated the shared global use of the radio spectrum, promoted international cooperation in assigning satellite orbits ... ITU is committed to connecting the world.

Nearly 150 years doing what? ITU was founded in Paris in 1865 as the International Telegraph Union.

Sounds to me like more empire building from the UN, using an "excuse-me" from Malaysia.

Surely private communication companies can come up with straightforward solutions acceptable to ICAO and airlines that meet ITU approval (if needed).

[AreOut]

me too, I'd put it some ~200 miles lower

[FE Hoppy]

The ITU have responsibility for anything the transmits in any spectrum. ICAO the UN, BT or the BBC all work to ITU standards. So the ITU will define a new standard for a new data transmission regime

[cribbagepeg]

ITU is a UN body funded by countries hosting the major players in their sphere of influence, with lobbying by those players aimed at preserving their monopolies or at least dominant positions within their respective industries. In this case, I suspect the satcom carriers and perhaps a/c nav equipment providers would be the major forces. Follow the money.

Satcom LINK management, as outlined very early in this very large body of comments, uses "pings" to set up and maintain said link between the a/c and ground. A link to an a/c may be used to carry communications under subscription agreements between that a/c's operators and the satcom provider.

The takeaway summary: satcom link messages exist without ACARS data messages, the latter being subject to subscription payments. Please don't conflate the two.

[IanWorthington]

I'm guessing this will somehow involve X.25, sixteen layers of different protocols on top of each other, and about a billion configuration options where any misconfigured option will prevent you from talking to anyone else.

Could be worse, they could get SITA involved.

[FoolsGold]

ACARS is, I believe, a maintenance information collection and transmission system. The next stop will have manuals open to the correct pages and spare parts close at hand and appropriate test equipment since they have received information as to what systems have reported error states en route. Reports go to the cockpit after a one minute suspense window that groups them by priority.

A "Ping" is merely an electronic handshake at the network level and does not mean anything at all about intending to transmit information or expecting to receive information; it is simply confirmation of the existence of a network connection.

Now for some reason the Carrier did NOT pay to have its ACARS data sent via satellite so the question for me is: Why would anyone try to disrupt the ACARS system if it was known that the Carrier did not pay to have ACARS data transmitted. Does such an attempt to disrupt the ACARS system indicate lack of familiarity with the carrier's operations?

[BuzzBox]

Two points:

First, ACARS is not simply a 'maintenance information collection and transmission system'. Broadly speaking, it is a communications and reporting system that is used to transmit all sorts of information to and from the aircraft. That information includes maintenance and systems data that is sent automatically by the aircraft to the airline and airframe/engine manufacturers. The system is also used by the crew and airline personnel to send operational information to and from the aircraft, including operational messages, weather reports, NOTAMS, flight plans and ATC clearances, etc.

Second, although there has been speculation here on PPRuNe that MH did not pay to have ACARS data sent by satcom, I don't believe that has been verified. Indeed, the Malaysian Ministry of Transport's MH370 Preliminary Report states:

'It was later established that the transmissions from the Aircraft Communication and Reporting System (ACARS) through satellite communication system occurred at regular intervals starting before MH 370 departed Kuala Lumpur, Malaysia at time 12:56:08 MYT and with the last communication occurred at 01:07:49 MYT'.

That report seems to indicate that ACARS transmissions were sent via satcom.

Position information is sent automatically by ACARS together with the engineering data. The crew would not necessarily know whether that information was being sent by VHF or satcom. If MH370 was the result of a deliberate act to make the aircraft 'disappear', then I would think the culprit(s) would make every effort to disable any systems that could transmit its position.

[porterhouse]

then I would think the culprit(s) would make every effor
making "every effort" doesn't necessarily equate with succeeding.

[Hyperveloce]

From the communication log, I get the following BTO values
TimeBTO=[0 12 13 25 37 118 191 251 311 371 461]; % min from 16:30
BTO =[14920 14940 14920 15200 15600 12500 11500 11740 12780 14540 18040]; % µs
and applying the BTO formula (1) page 54 of the ATSB report, I get the following ping ring radiuses:
PingRad =[2243 2241 2238 2267 2309 1904 1758 1798 1953 2194 2621]; % NM
and the following elevation angles:
ElevAng =[46.72 46.76 46.81 46.27 45.48 53.10 55.87 55.12 52.17 47.63 39.70]; % degrees
(I have also generated kml files for these ping rings to compare visually with those of the figure 18 of the ASTB report)

Do you get similar values ? (radiuses and elev. angles). They seems to differ from Duncan Steel's.

[jimjim1]

BOAC mentioned:-
It would still be useful to know what data was being sent from Acars via SAT

Perhaps ...

I seem to recall reading on pprune that the engines were leased from RR (power by the hour?) and that RR used ACARS for engine data. Perhaps RR paid for satellite acars for their data?

There would likely be engine start messages, who knows? Once in the cruise no more messages due to stable engine operation?

I forget now, is there some evidence that the satellite acars was turned off or disconnected leaving the bare satellite terminal doing its ping thing? I vaguely recall that there is. Presumably fuel exhaustion would generate some engine acars messages but none were received.

So many questions, so few answers.

[matkat]

My knowledge of engine health monitoring is a few years out of date but IIRC the engines were reporting continuously from start to shutdown, would expect RR to have a similar system however I fail to see what good this would be as they only transmitted engine useage/health monitoring parameters but no positional data.

[Ian W]

Before the hamster wheel spins up again - this was all discussed earlier in the thread and discounted.

The engine health monitoring to Rolls Royce and the aircraft health monitoring to Boeing via ACARS were both inactivated with no subscription paid. There was no subscription for any ACARS activity over SATCOM only over VHF and that was routine 30 minute reporting the last of which was at 1.07. The only reason that the SATCOM was operative at all was the SATCOM phone facility was still available.

[roulishollandais]

mm - if there is 'Acars Data' floating around as you say, where is it and what does it contain?The problems with ACARS and analog new type of communication - RR engines' management messages- is that they are said proprietary. That fact put their messages out of the ICAO operators' rules. i.e. AF447 ACARS told in one page and real time what happened helping a high probability analyse. That analyse has been confirmed two years later by the Flight Data Recorder. But the Crew missed that analyse from ACARS available in the cockpit at the same time because certified procedures did not use them. Congratulations to the BEA who published ACARS in the report.

[mm_flynn]

Before the hamster wheel spins up again - this was all discussed earlier in the thread and discounted.

The engine health monitoring to Rolls Royce and the aircraft health monitoring to Boeing via ACARS were both inactivated with no subscription paid.

That is probably true, but I have not seen anything more substantive than 'someone on PPrune said it' and the lack of any info from these companies as the basis. For the comment,



There was no subscription for any ACARS activity over SATCOM only over VHF and that was routine 30 minute reporting the last of which was at 1.07. The only reason that the SATCOM was operative at all was the SATCOM phone facility was still available.

You have said this part over and over. Your statement is flatly contradicted by the preliminary report, this has been mentioned on every spin of the hampster wheel with a request to provide some insight as to why your assertion might be true and, on every occasion you have chosen not to provide that insight.

[Wind_Tunnel]

Using p.5 of the ATSB's June 26 report, it is easy to replicate their working endurance model, March 17-27 (in, say, Google Earth):

Start at NW tip of Indonesia
Assume nil wind
Pick an arbitrary speed, s (say, 490 KGS)
Draw straight lines between Inmarsat arcs of length (time between arcs)/s
After 7th Inmarsat arc, keep going until you hit S1-S3 SE border
Repeat for several other speeds (say, 20kt intervals, down to 330KGS)

You have now precisely replicated the ATSB's working endurance model, March 17-27. (SE border of S1/S2/S3 has been publicly confirmed to be the ATSB's March 17-27 working endurance limit.)

The feasible range (S1/S2/S3) is the set of points bounded by the 6th Inmarsat arc, and the endurance line you've just reproduced.

Here's where it gets good:

Perform the update to this model the ATSB claims to have performed on March 27 - take away starting fuel. How do you do this? Easy: just shorten each flight path by the exact same PROPORTION. For example, make every flight path 5% shorter. Or 3%, or 7% - we don't know exactly how MUCH fuel they took away, so maybe try several different reductions. The key point is that each path is still going the same speed, and thus must still hit each arc at the exact same spot - the lack of fuel simply shortens the length. By the SAME X% FOR EACH PATH. Because that's precisely how a set reduction in available fuel would AFFECT each path's range.

You will find that, as you take fuel away, the S3 search zone becomes infeasible much faster than does S2. In fact, there is absolutely NO amount of fuel that could be taken away that could rule S2 out, yet S3 in.

So here's my question: why, then, did the ATSB move the search from S2 1,100km NE into S3, after taking AWAY fuel?

Illustrations (overview & close-up):
https://drive.google.com/file/d/0B-r3yuaF2p72aTU0Q1FxTzJSbHM/edit?usp=sharing
https://drive.google.com/file/d/0B-r3yuaF2p72OWR0NVpmZW1wRG8/edit?usp=sharing

Something doesn't add up. At all.

[Ulric]

I think that to draw any sort of sensible conclusion about the track terminal, you must make an assumption about the starting point of the track since displacement of one, results in displacement of the other. Your analysis is fine if one supposes that the tip of Indonesia is in fact, the starting point of the southern track. This is far from clear from the known data.

[roninmission]

I know the contract for survey of new search area has been let. Has the contract for actual search been let?

[Hyperveloce]

Wind-Tunnel,
I am a bit bothered too with the data. Here are my last (Montecarlo simulation) results:
The last south legs of all the simulated trajectories are at constant speed (and altitude) and the south turn occurs randomly between 18:28 and 18:40: it is assumed that the C-channel BFO measure of 88 Hz at 18:41 is a valid one (it is included in the observed BFO profile the trajectories are trying to mimick) and suggests a south bearing at 18:41. The constant (ground) speeds span from 330 kts to 530 kts, the crash latitudes from 22°S to 41°S:
https://drive.google.com/file/d/0B3srJsCh7WAtY2ppSzQ2aTdOS1k/edit?usp=sharing
https://drive.google.com/file/d/0B3srJsCh7WAtQnZqVGVrWjBsYjA/edit?usp=sharing
The first part of all the trajectories comply with the ADS-B measures (they are not randomized on this first leg), and roughtly with the Butterworth radar track above the Malacca strait (supposedly linked to the MH370 ). When the south turn is triggered, a constant speed reference trajectory is built (intersecting two kinds of small circles on the earth spheroid: the ping rings and the positions reachable from one given location at a given constant speed) from the A/C position and randomized:
https://drive.google.com/file/d/0B3srJsCh7WAtcnNDTzg0R05ueFk/edit?usp=sharing
The trajectories best fitting to the observed BTO/BFO values are around 460 kts (true gound speed) and end near 36°S, not near 30°S (ATSB):
https://drive.google.com/file/d/0B3srJsCh7WAtTTdHQVhvV3U2NU0/edit?usp=sharing
https://drive.google.com/file/d/0B3srJsCh7WAtNUZac3V2ZmlMVnM/edit?usp=sharing
the BFO errors versus the crash latitude:
https://drive.google.com/file/d/0B3srJsCh7WAtcGJydGZvYWxjcUE/edit?usp=sharing
the BFO errors versus handshake instants and ground speed:
https://drive.google.com/file/d/0B3srJsCh7WAtQkVvekQ5akV0TW8/edit?usp=sharing
the BFO and the BTO enveloppes of the 50 000 flights:
https://drive.google.com/file/d/0B3srJsCh7WAtMGo2WmV1V09jZHc/edit?usp=sharing
But we do not know clearly what was the injection/turning point toward the last south leg in the ATSB report.
Jeff
PS) Rerunning the simulation (16h run time) to confirm these results.

[Ulric]

So, are we assuming that the 19:40 ping line was traversed in an easterly or westerly direction? Assuming the plane was travelling south east initially makes sense but then forces a much more southerly heading for the timing to be right for the 20:40 ping. Assuming the 19:40 ping occurred with the plane travelling west requires "the hook to the north" between 18:28 and 19:40 referred to in the Australian report but then allows a more plausible track through the 20:40 arc and results in a constant speed track which intersects the Northern end of the proposed search areas.

In the midst of all this, there is a glaring anomaly which doesn't go away unless and until you have accurate fuel figures. They have not been released AFAIK.

When all is said and done, you cannot make a good enough prediction of the terminal position unless you have certainty about the direction of travel at 19:40 and the fuel load. Even then, you must indulge in some educated guesswork to narrow down the possible tracks.

In the report, we have some clues which I think are important. The report states the likelihood that the southerly track was flown on autopilot and therefore, I am inclined to believe that any solution which would require any heading or speed changes becomes highly unlikely. This in turn, implies (to me at least) that the 19:40 direction was west because that fits in with the 20:40 ping and a plausible speed and heading.

[AreOut]

"The report states the likelihood that the southerly track was flown on autopilot"

I have seen it but never found out reason why.

[Ulric]

There is much information which is accessible to the investigation team but, not to us. We must await their conclusions.

[Wind_Tunnel]

Ulric:

Your analysis is fine if one supposes that the tip of Indonesia is in fact,
the starting point of the southern track. This is far from clear from the known
data.


Ulric, I am not trying to build the ATSB's current model (and best estimate). I am trying to show you what the ATSB model (and best estimate) would have looked like in the March 17-27 era. Their report explicitly states (on p.5) that their working model at that time assumed a turn at the NW tip of Indonesia.

The reason replicating those (now obsolete) assumptions is important is because I go on to prove that the model update they (claim to have) made on March 27 could not possibly have supported the DECISION they made on March 28 - which was to move the search dramatically NE.

[Ulric]

I think that is a reasonable thing to do. Any projection of course is based on some set of assumptions and I think the report is laced with clues about what those assumptions were. You will appreciate that there are possible scenarios implied by the data which do not sit very well with public statements made by officials. We all realise (at least I assume we must by now) where these anomalies lie and that the data necessary to resolve them is not yet available to us.

Patience is a virtue in this case.

[Wind_Tunnel]

Still think you're missing the point, Ulric. ATSB issued a release March 28 saying, "plane had less fuel, so we're moving search 1,100km NE". It is a matter of simple geometry to prove that, if the starting point doesn't change (Appendix A suggests it didn't), and the Inmarsat arcs didn't change (they have always been taken as gospel), then their statement is false - the decision is counter-indicated by what would have HAD to have been their model's reaction to less available fuel.

When the ATSB issues key statements (this fuel analysis underpinned the critical decision precipitating an 8 week waste of time and money up at s20 lat) which are provably false - and when this falsehood can be demonstrated by the ATSB's own data (which suggests they ought reasonably to have KNOWN they were false) - I for one feel that patience is one of the last things called for. On this, you and I may have to agree to disagree.

[Ulric]

I think the assumptions made by the investigating team are only provably false if you have certainty about some of the variables and in particular, the point at which the SSE course was assumed. Once you have fixed this point, speed and heading are covariant and the range of terminal points becomes constrained in exactly the way you describe. If you take away the certainty about the starting point, and I think we must, the range of available terminal points matches those assumed by the investigation team.

The message is clear - information about the starting point is not available to the investigators. I believe that should make us circumspect about accusing the investigation team of incompetence.

[Blake777]

Hyperveloce/Wind Tunnel

Excuse me if I am missing something but if the aircraft was on autopilot for the last southern leg (an assumption by ATSB) what effect would magnetic variation have on your calculations or is this allowed for? I am trying to get my head around the reasons the ATSB may have made that assumption other than "best fit model".

[Ulric]

This might help
https://upload.wikimedia.org/wikipedia/commons/d/d4/World_Magnetic_Declination_2010.pdf

[sladen]

unless and until you have accurate fuel figures.

A "1641 492" fuel on board update (49,200 kg at 16:41) was included in the 16:41:58 ACARS OOOI off event:


www.paul.sladen.org/download/aaib/sladen-20140703-briefing-note.pdf (http://www.paul.sladen.org/download/aaib/sladen-20140703-briefing-note.pdf) (4MB, 3 July 2014)


The above may also be of interest for "Ian W", as it provides an example of an ACARS message, being forwarded via Satcom.

[Ulric]

Thank you very much for that. It is one of the outstanding pieces in the puzzle!

[Wind_Tunnel]

Ulric,

Your last post claimed to rebut mine by pointing out that the turn point was not known. Good grief - where do I claim that it is? All I'm trying to do is accurately build the ATSB's first two performance models - for this project, then, "correct" means "what they were using as an assumption", not "where the plane actually turned". Please stay with me on that concept.

Why do I have confidence that my starting point - NW tip of Indonesia - is an accurate depiction of what the ATSB was assuming in late March?

1) That's what they said they used. On page 5 of their report. That was their assumed turn point for both March 17-27 (per p.5) and March 28-April 1 (per Appendix A) performance limits.

2) I have independently replicated their endurance line using a systematic endurance vs speed study from a paper by Delgado/Prats. When I calibrate one point of their model to the ATSB's endurance line, I find the Delgado/Prats model matches the ATSB line almost perfectly. The green line in my illustration is, in fact, the Delgado/Prats result. If I had a materially wrong starting turn point, the two curves (ATSB, Delgado/Prats) would have diverged. They match.

[Ulric]

All I'm saying is that if the start point is not known, the end point is not known because one is dependent on the other. The investigation team presumably have access to more information than we do and at some point they changed their minds about what they thought happened.

I don't have a problem with that but when you come along and start to claim that their working hypothesis back in March was "provably false", those are strong words and I have to respond by saying that it is probably irrelevant at this point in time.

[Wind_Tunnel]

Ulric: your first two statements are completely true, and completely irrelevant to my point, for reasons already explained. They CERTAINLY have changed their mind about where the turn occurred - but not by March 28th.

Yes, "provably false" are strong words. Which is why I presented strong proof. If you don't understand it, just ask: I'm here to help; if you refuse to TRY to understand it, that's not my problem.

I have proven the ATSB went against (what they have described as) their own analysis in sending the search NE on March 28. They either...

1) stink at performance analysis,
2) lied (in release AND report) about why they REALLY moved NE, or
3) intentionally misdirected the search

I HOPE it's "only" 2) - and that the March 28 shift was for LEGITIMATE reasons. But if so, they should have told us what those legitimate reasons WERE; teams of spectacularly astute independent experts are working very hard trying to help find the plane; but they are working blind because of 2).

HIGHLY relevant, I'm afraid.

[mixture]

teams of spectacularly astute independent experts are working very hard trying to help find the plane; but they are working blind because

Well they can't be particularly astute "experts" if they are working blind without access to the information that only those directly involved have..... sounds more like armchair investigators. :cool:

[Ulric]

I HOPE it's "only" 2) - and that the March 28 shift was for LEGITIMATE reasons. But if so, they should have told us what those legitimate reasons WERE; teams of spectacularly astute independent experts are working very hard trying to help find the plane; but they are working blind because of 2).

HIGHLY relevant, I'm afraid.

Indeed, it is highly relevant that the investigation team seems to have made a set of assumptions which is at odds with their public statements. There are however, good reasons for these discrepancies.

There are two bodies of information here which are qualitatively different. One is the set of public announcements made by various bodies which may be coloured by perceptions and motives that are not transparent. The other is the data. If one treats these two sets of information as separate, it becomes clear that there are differences in the implied track of the aircraft depending on how much weight one gives to each set. It also becomes clear why the investigators may be making assumptions which look inconsistent and are unable to be forthright about why they have done this.

I don't think this implies any impropriety on the part of the investigation team since I believe that the decisions they have made are sensible based on the data alone.

[phiggsbroadband]

Just one question for those on this forum with technical knowledge....


If the auto-pilot was set to a heading of 180 degrees, would the plane fly to the Magnetic, or Geological, South Pole? And would the flight path be a straight line, or deviated by the differing winds aloft?

[porterhouse]

the auto-pilot was set to a heading of 180 degrees, would the plane fly to the Magnetic, or Geological, South Pole?Heading is always magnetic unless your are very close to the Poles.

And would the flight path be a straight line, or deviated by the differing winds aloft? Yes, the aircraft would be subject to winds aloft so its path would not be a straight line.

[BOAC]

Heading is always magnetic unless your are very close to the Poles. - no. It is selectable at any latitude to either magnetic or true. Sometimes Grid at high latitudes.

[porterhouse]

I stand corrected - it is selectable.

[Ulric]

On a constant magnetic heading the course flown is a Rhumb line.
Rhumb line - Wikipedia, the free encyclopedia (http://en.wikipedia.org/wiki/Rhumb_line)

[phiggsbroadband]

Thanks for the replies.... So if any Southerly heading was set, the plane would eventually reach the South Pole (Mag or Geo, dependant upon setting), flying a not particularly straight line.

Only if 180 degrees was set would the plane fly a great circle route, other times it would be the Rhumb Lines.... Thanks again...

[WingNut60]

Shooting from the hip here, and practically speaking I think that you would be correct.


But I suspect that, the rhumb line projection will never, in theory, end exactly at the pole.


Come on you mathemeticians, shoot me down, if you will.

[hamster3null]

As the Wikipedia link says, the rhumb line reaches the pole in a finite distance. It is relatively intuitive. When you are moving along the rhumb line, you have to travel fixed distance for every degree of latitude (60 nautical miles if you're going due north or due south, 60/|cos(bearing)| nm for other bearings). Therefore, to go from the equator to the end of the line, you go 60*90/|cos(bearing)| nm.

[WingNut60]

Right. I think that my tired old brain can visualise that now.


I'm not sure how much relevance there is in this to MH370, but an excellent reference can be found at Calculate distance and bearing between two Latitude/Longitude points using haversine formula in JavaScript (http://www.movable-type.co.uk/scripts/latlong.html)

[susier]

http://www.jacc.gov.au/media/releases/2014/july/mr054.aspx




Two ships, Zhu Kezhen and Fugro Equator, continue to work in the southern Indian Ocean, surveying the sea floor in preparation for the deep-sea search for missing Malaysia Airlines flight MH370.
In June, an expert satellite working group identified a search zone of up to 60,000 square kilometres, which represents the highest priority for future search efforts. As with previous search areas, it is located along the seventh arc—a thin but long line that includes all the possible points where the last known communication between the aircraft and a communications satellite could have taken place.
Before the deep-sea search can commence however, it is necessary to map the sea floor in this remote region of the Indian Ocean, which until now has been poorly charted. The aim is to identify significant features on the sea floor, which may present a hazard for the deep water vehicles that will be used for the search.
Since 24 May 2014, the Zhu Kezhen, a Chinese PLA-Navy vessel has been conducting survey operations. The bathymetric data it is collecting will assist in characterising the sea floor topography. As of 30 July 2014, it has sounded over 25,000 square kilometres along the seventh arc.
Fugro Equator, an Australian-contracted specialised survey vessel, has also been conducting bathymetric survey work. As of 30 July 2014, over 43,000 square kilometres have been sounded by Fugro Equator.
Analysis and mapping of this data is continuing.
On 6 July, the Government of Malaysia announced that its hydrographic survey vessel, the KD MUTIARA, would join the Zhu Kezhen and the Fugro Equator in August to continue the bathymetric survey work.
It is expected that the bathymetric survey work will be completed by September. The deep-water search is expected to commence in September following the appointment of a prime contractor through a request for tender process.
The JACC will provide updates on the activities of the bathymetric survey, which can be found at www.jacc.gov.au/media/maps/index.aspx (http://www.jacc.gov.au/media/maps/index.aspx)

[Downwind Lander]

During the search for MH370, there was something sad about sailors on aboard RN ships, with binoculars, searching for wreckage by looking out to sea. How far can they see with any reliability? Is this next to futile?

A UK Freedom of Information request asked if the Royal Navy have any towed observation platforms, a device a bit like a parascender/hang glider, which, apparently, can reach almost to 1000 feet, and towed behind the ship. This seems gloriously low tech, very cheap and potentially useful. Having said that they did not, this is what they went on to say:


"UK Armed Forces rely on advanced sensors that provide a highly effective detection capability. For example, surveillance radar systems, such as the SCANTER 4100, which are currently in service with the Royal Navy, offer a combined surface and air surveillance capability to a range of 110 nautical miles from the ship. The radar is capable of detecting small objects in all weather conditions which gives it a clear advantage over the approach you propose. In addition, highly capable sonar systems, such as the 2087 used by the Royal Navy's Type 23 frigates provide an effective means of acoustically detecting submerged objects".

Can anyone comment on this, bearing in mind that nobody is talking of "either or", but rather of "all" useful ideas. Who knows when the world will be on another similar search?

[Bare Plane]

MH370: Australia to announce next phase of search - CNN.com (http://edition.cnn.com/2014/08/05/world/asia/australia-malaysia-airlines-mh370/index.html?hpt=hp_t3)

Wish them luck.

[mmurray]

Transcript of press conference here

Joint Agency Coordination Centre (http://www.jacc.gov.au)

[Going Boeing]

Dutch firm to conduct MH370 underwater search

A Dutch firm will conduct a deep-water search in the Indian Ocean for missing Malaysia Airlines Flight MH370 from September, Australia said on Wednesday, expressing "cautious optimism" that the plane will be found.

Fugro Survey will use its state-of-the-art vessel MV Fugro Equator and the Fugro Discovery to search a 60,000 square-kilometre (23,000 square-mile) zone in the southern Indian Ocean.

"The underwater search will aim to locate the aircraft and any evidence, such as debris and flight recorders, to assist the Malaysian investigation team on the disappearance of MH370," Australian Deputy Prime Minister Warren Truss said.

"The Malaysian government have also offered equipment which will participate in that search and of course, there's a Chinese vessel already operating in that area in relation to the survey."

A huge air and sea surface search has failed to find any sign of MH370, which went missing on March 8 with 239 people on board. Authorities believe the doomed airliner veered mysteriously from its Kuala Lumpur-Beijing route before crashing in the remote Indian Ocean.

The Fugro Equator is currently working with Chinese PLA-Navy ship Zhu Kezhen to map the seabed in the search area, based on the missing plane's last satellite communication around 1,800 kilometres (1,100 miles) west of Perth.

The Fugro Discovery is en route to Perth from Britain, with the deep-water search expected to take up to a year to complete.

Both vessels have towed deep-water vehicles and crew with search expertise, Australia's Joint Agency Coordination Centre said. The sea floor search will use sonar equipment and video cameras to locate and identify any debris.

Truss said he remained "cautiously optimistic that we will locate the missing aircraft within the priority search area". He added that the search "will obviously be a challenging one".

Martin Dolan, commissioner of the Australian Transport Safety Bureau, the agency leading the search, said the underwater mapping survey due to end in mid-September was discovering "some surprises".

"We haven't completed the mapping, so we are still discovering detailed features that we had no knowledge of -- underwater volcanos and various other things," Dolan said.

Truss said the discoveries "demonstrated why this mapping was so necessary".

"The ocean is not just simply flat and featureless," he said of the underwater terrain in the search zone, which is about half the size of peninsular Malaysia. About 60 percent of the area has been mapped so far.

"There's quite a lot of geological features there that will be a challenge in the search, that we needed to be identified to make it actually possible to undertake the sonar work that is going to occur from now on," Truss added.

The underwater search contract could cost Aus$52 million (US$48 million) if it stretches up to a year, he said.

by Usman SHARIFI © 2014 AFP

[The Old Fat One]

During the search for MH370, there was something sad about sailors on aboard RN ships, with binoculars, searching for wreckage by looking out to sea. How far can they see with any reliability? Is this next to futile?

DL, I had 27 years carrying out SAR missions (airborne) and much of the time all those expensive sensors are useless, and you are reduced to visual search. Acoustics is useless unless there is an acoustic source, with a usable range.
Radio homers are useless if nothing is emitting. Radar is useless unless something detectable is sticking out the ocean.

Further problems abound (in all oceans and scenarios)...

The sea is full (and I mean full) of junk. All of which will have to be investigated if it is detected within the search area.

In deep ocean, range is massive issue (in terms of fuel and time on task for aircraft and reaction time for ships)

Surface wreckage drifts (50 miles plus a day), expanding the datum.

Sea state has a huge and very variable effect on the efficacy of visual search.

the list goes on but you get my point.

In the case of MH370 the deck was stacked against a quick find from the get go...no datum whatosever, long way from land, no sign of detectable sized wreckage or high visibility survival equipment, very short range and short life acoustic beacon...

So yeah, futile sums it up completely. Still have to try though, don't they?

Better sensors are not the answer (the sensors do what the taxpayer pays them to). The answer is better tracking in the sky and better beacons on the boxes.

[Wind_Tunnel]

Just documenting latitudes and distances for two key flight paths from the fan I generated last month in replicating & validating the ATSB’s original endurance line (SE border of S1/S2/S3, Fig.3, p.5, June 26 ATSB report):


(wide map: http://bit.ly/WIb2Ng close-up of search zone: http://bit.ly/1nI3V0Q)


Start point (NW tip of Sumatra):
18:36:03 5°59′ N


A 460 knot path would hit Inmarsat arcs 2-thru-6 at these SOUTH latitudes:
19:41:03 2°14′
20:41:05 9°50′
21:41:27 17°32′
22:41:22 25°11′
0:11:00 36°35′
…and end in the middle of the March 17-27 search zone
…with (according to the width of S2 at that point and heading) 329 nmi to spare
…for a total flight distance of 5.58hrs x 460 kts = 2,568 + 330 = 2,897 nmi



if new fuel analysis shortens range 11% = 319 nmi:
…still feasible (by March ATSB standards), with 10 nmi to spare


A 390 knot path would hit Inmarsat arcs 2-thru-6 at these SOUTH latitudes:
19:41:03 0°59′
20:41:05 7°30′
21:41:27 14°3′
22:41:22 20°27′
0:11:00 29°48′
…and end in the middle of the Mar.28-Apr.1 search zone
…with (according to the width of S3 at that point and heading) 248 nmi to spare
…for a total flight distance of 5.58hrs x 390 kts = 2,177 + 248 = 2,425 nmi



if new fuel analysis shortens range 11% = 267 nmi:
…NOT feasible (by March ATSB standards) by 19 nmi


If you drop fuel by less: original search location still feasible
If you drop fuel by more: new search location still infeasible




This simple, transparent demonstration proves mathematically – without a single parameter that does not come from the ATSB’s own reports – that their “drop in starting fuel” argument could not POSSIBLY have been a good reason to move the search 1,100km NE on March 28.


So why did they?

[Alloyboobtube]

You have a government pleading poor yet they are paying for another countries expenses. Which will be huge as It will not be found.

[Shadoko]

So why did they?

IMHO, an erroneous detection by the HMS Tireless (remember the "self satisfied" statement when she returned to England?) and then, using the pinger locator of the Navy towed by Ocean Shield at the limits of its capabilities, a "false" signal from the black box pingers.
Anyway, it was (IMHO) the right thing to do if there was the smallest chance to find the plane there.
Perhaps this can't be told for "military" reasons (Tireless capacities) and thus the Australian can't tell that, nor show the "signal" acquisition charts (Navy capacities). Nothing abnormal from my point of view this is not made public. And they don't lie: the search zone was moved after "more study".

[Wind_Tunnel]

Shadoko: thanks for replying. If you agree with me that reduced available fuel could not possibly have been the reason the ATSB moved the search 1,100km NE on March 28, then wouldn't you also agree with me that the statement...


"On 27 March (D20), the JIT advised they now had more confidence in the increased speeds provided by primary radar near Malaysia. This increased the aircraft fuel burn and the most probable track moved north to the S3 area." (source: "MH370 - Definition of Underwater Search Areas", June 26, page 6)


...is deceitful?


P.S. I'm only pointing out the logical flaw in the FIRST humongous shift NE that occurred that week. Four days later, on April 1, they moved the search a FURTHER 1,500km (from yellow to red, on the chart I provided above) - up into 20s latitude - where the acoustic pings then pinned the search for eight weeks. Your quote may be referring to the rationale for that SECOND move (which is unrelated to my argument).

[Shadoko]

Your quote may be referring to the rationale for that SECOND move (which is unrelated to my argument).

You are absolutly right, I may have read you with poor attention: sorry for that.

[autoflight]

I expect T7, like FBW airbus, can be in heading or track. Assumptions about heading, but then describe track of the aircraft seems to indicate carelessness of a few posters.

[Chris Scott]

In late April I commented it was rather odd that no identifiable wreckage had been found on shore. That the a/c could have found its way to the ocean floor completely intact, even in the unlikely event of a relatively successful ditching, seemed implausible. We're not talking about the Hudson River.

At risk of being a nuisance, did anyone come up with an explanation for that?

[Ulric]

Chris Scott

There have been discussions about this elsewhere. The questions are about things like how rough the sea might have been - weather reports say calm. There are also questions about how much control the automatic flight envelope protection systems might have had over the final trajectory.

It is one thing to say that it is unlikely that the aircraft remained substantially intact but quite another to say it is impossible.

Subsequent to the touchdown there are questions about ocean winds and currents and locations where debris might eventually wash ashore. I have read suggestions by people who know more than I do that wreckage may drift for many months - even years - before washing up. The possible locations are affected by the range of possible impact points but may be as far apart as South Africa and Tasmania. Since the impact point is unknown and all of the searches to date may have been distant from it, I don't find it surprising that no wreckage has been identified. The Ocean is vast and the possibility remains that very little floating wreckage might be present.

[MG23]

The ATSB won't just pick search areas at random. Any area they pick will be based on advice from the groups studying satellite communications data and aircraft performance.

[olasek]

We are thus left with only unlikely - possibly highly unlikely - potential explanations.
There is a difference between human behavior and what is likely and/or unlikely and purely technical things. Human behavior as we know from history of aviation can be so aberrant that anything however unlikely can't be discounted but on the other hand purely technical aspects - here what is likely and unlikely remains pretty much constant and I don't think this particular accident will surprise us a lot in technical aspects (if we ever solve it in its entirety) but human behavior can astonish us again.

[onetrack]

I've just spent 3 weeks on Cocos-Keeling Islands and I was stunned at the absolutely vast amounts of flotsam and jetsam littering the shores of the islands.

I was always under the impression (now sadly corrected) that the Indian Ocean was one of the worlds cleanest oceans, and I was very pleased to live on the Eastern edge of it, and revel in its cleanliness.
The truth is, the Indian Ocean is a watery garbage pit, on a par with every other world ocean.

Much of the flotsam and jetsam has Asian origins, and much of it is reputed to be remnants of the Aceh and Fukushima tsunamis.
I believe a lot of it is even more recent and is a result of purely a lazy disposal attitude, on a large percentage of the population of the above nations.

Surviving MH370 wreckage (and there is bound to be some) would be buried in the overwhelming volume of general rubbish that populates the Indian Ocean.

AFAIC, the greatest failure on the authorities part is not immediately issuing a reward notice to the populace of every nation and every island and continent bordering the Indian Ocean, for a monetary reward for any possible aviation-related flotsam or jetsam found in these areas.

Even just one small wreckage item from MH370 would add to the tiny levels of available information and enable researchers to try and calculate the track of the wreckage.

[Ulric]

I remember spending a day driving around Grand Cayman many years ago exploring all the little beaches and coves. One thing that struck me at the time was the extent to which the waves seem to sort the rubbish into objects of approximately equal size and density.

One small beach we found was obviously the place where lost shoes wash up. Large amounts of rubbish but almost all shoes. That leads me to believe that different components of any wreckage are likely to wash ashore eventually at different places or times depending on size.

[Wind_Tunnel]

I was invited to speculate. Here's where logic takes us (not very far, I'm afraid, but it's a start):

The idea that a debris field was created in the SIO - and yet the world's collective search efforts have yet to identify a single ounce of it - is well beyond the bounds of common sense.

I concede that the search area is vast and inaccessible, and that, over time, debris sinks, degrades, and comingles with other junk. But all of those things take time; and many items (lifejackets, eg.) retain their visibility, flotation, and evidenciary value for years, so all that would have been needed to positively identify something was a sufficiently vast and hi-def satellite search.

I distinctly remember watching a televised news item about new technology that allowed satellites to read the headline on a newspaper held by someone on the ground. They used "newspaper" instead of "iPad", because the item I watched was broadcast in the 1970's. So I'm going to go out on a limb and suggest that current satellite technology is sufficient to scan quickly for possibilities, and zoom in to identify.

So here are the 3 possible avenues taken by the investigation leadership over the past 5 months:

#1 they decided to withhold this search technology (i.e. chose not to find anything)
#2 they applied the technology, found debris, kept it secret, yet still haven't found MH370
#3 they fully applied the technology, but didn't find debris

#1 requires either irrational search leadership, or the known non-existence of debris.
#2 requires irrational search leadership.
#3 requires the non-existence of debris.

Result: if the search leadership is rational, then there is no debris.

The southern-track-until-fuel-exhaustion hypothesis requires debris (a powerless 777 cannot magically "self-Sully" onto a rolling, roiling ocean).

Result: if the search leadership is rational, then the southern-track-until-fuel-exhaustion hypothesis is invalid.

[Pace]

Wind Tunnel

I don't know why I bother writing anything in this thread as no matter how rational or polite it gets deleted! I must be on auto delete :E ( That includes the MH17 thread :{ but finding any floating fragments whether life jackets or otherwise from the wreckage will not after all this time give an indication of the crash location.
You are not dealing with just currents and tides but also varying winds over a considerable period of time.

if the Ocean was relatively shallow there might be some small chance of finding static wreckage on the Ocean bed but with the considerable depth and pressure that chance is negligible bar an absolute miracle

[Ian W]

Invited to speculate - that's got to be 50% of this thread... :8

However, look at what happened to the Asiana 777 which hit a sea wall then pirouetted on its nose down the runway then belly flopped back onto the ground. What bits fell off that? The tail cone due to the impact on the sea wall, some undercarriage parts and the engines. The fuselage stayed intact.

So what would happen with a relatively slow and level impact with the water, perhaps following an invented RNAV STAR to an imaginary airport 50 feet above the ocean? I would say the same - the tail may have remained in place without a hard strike but engines may have been lost. But the fuselage could have stayed intact. Engines would have sunk immediately, followed by the fuselage after a few hours. Given 2 - 3 weeks delay what would be still floating? Even if the wings and tail had broken away they may well have sunk. But after floating for 2 weeks or so would be nowhere near the fuselage which itself would not be near the engines which would have sunk close on vertically.

Beach-combing for debris may not be very productive.

[Chris Scott]

Pace,
You and I both...
Quote:
"...finding any floating fragments whether life jackets or otherwise from the wreckage will not after all this time give an indication of the crash location."

No, except that if authentic they would indicate the a/c came down in the ocean.

[The Old Fat One]

The idea that a debris field was created in the SIO - and yet the world's collective search efforts have yet to identify a single ounce of it - is well beyond the bounds of common sense.

Nope, your post is well beyond the bounds of common sense. SAR professionals such as myself have tried to explain why it is entirely possible (nay, probable) for a debris field to remain undetected...but armchair experts simply know better.

Hey ho, there are none so blind as those that cannot see.

[slats11]

Its not easy to spot wreckage from the air. Hours of staring at white waves wondering if one of them is something more than a wave. Grey clouds, grey seas, and grey aluminium.

If you want to find wreckage at sea, paint the planes fluoro yellow. White and grey are almost camouflage in poor conditions.

And thats before you consider how large the search area is, and the delay before the search started in the Indian ocean.

Its not always easy to even spot ships from the air. There are numerous WW2 examples of planes missing ships (and even convoys) when they must have passed almost overhead. Pilots from the Ark Royal launched torpedoes at the Sheffield when they believed they were attacking the Bismarck.

Im not at all surprised at the lack of wreckage. Maybe if the fuselage had been breached and lots of bright things spilled out. But a few control surfaces or even a couple of wings would be hard to find, even if still afloat at the time people started looking.

[rh200]

Please correct me if its wrong, but my understanding is the FOV of a satellite camera will decrease in proportion to its resolution.

Another words you can see a big area but not small things. Or you can read a headline, but only on a small area (for those that can read headlines). Hence to see the small stuff you would have to know where to look.

The ocean and search area is :mad: huge, combined with the fact they didn't get to the "most likely spot until significant time had past means dispersal was even worse.

[MartinM]

Swissair 111 had an undetected fire on board which was initially underestimated. When the fire was detected, by following the checklist, and the pilots did exactly follow the procedure, they were finally unable to control the aircraft and died.

I strongly suggest to listen to the ATC/Cockpit voice recordings. They pilots decided, due to the smell of smoke which was not found, to head to Halifax. But they had to dump fuel. By that time, they did not know how catastrophic the fire was under the hood.

Remember that the MD11 was not having all the sophisticated systems as a B777.

MH370 had a small little SATCOM antenna on the side of the aircraft. The ACARS system is connected to two redundant SafetyBuses systems which have a 28V battery powered, aircraft independant power source. Generators, Bat or RAT are not needed to have the system alive.

There is no CB in the cockpit, like in a Boeing 737. You cannot pull it out just like this.

So, for me ot logical what event would stop ACARS to send information, except a big booom, like with TWA800.

[Shadoko]

A new Bathymetric Progress Map of the search area just released:
Progress report (http://www.atsb.gov.au/mh370/progress-report.aspx)

And also a revised version of the report about the search area, with new info:
MH370 - Definition of Underwater Search Areas (http://www.atsb.gov.au/mh370/mh370-definition-of-underwater-search-areas.aspx) :

Since the public release of the report 'MH370 – Definition of Underwater Search Areas' on 26 June 2014, the ATSB has received a number of queries about some of the technical details contained in the report. The queries have been made directly to the ATSB or through the Chief Commissioner’s blog, InFocus, on the ATSB website.
As a result of the queries, the ATSB is today releasing an updated version of the report to clarify a number of technical aspects. The changes to the report are detailed in the Addendum on the inside cover.

Some of the "queries" are in the comments of this page: Blog - Cautious optimism in search for MH370 (http://www.atsb.gov.au/infocus/posts/2014/cautious-optimism-in-search-for-mh370.aspx)

[Shadoko]

Differences viewed in the "revised version of the report" after a quick look (Old >>> New):
p.02: Jaworkski >>> Jaworski
p.20: Note 16: Complete table of handshake satellite data values provided at appendix X >>> Nothing (there was no appendix X... and it is not in this revised version!). So all subsequent notes are numbered minus 1.
p.32: reference to figure 31 >>> reference to figure 32. It was a typo.
p.56: Table 4 >>> Table 3. There was a mis-numbering.
p.57: Four paragraphes added. From "The reason for the partial compensation..." to "This
combined factor was used to determine the ‘Satellite and EAFC Effect’ in the BFO calculations." Have to be read!
p.58: Tables 5 and 6 modified with one line added which explains why the "Downlink Doppler" was changing with a/c latitude (Table 6):
"Uplink Doppler" and "Downlink Doppler" >>> "Aircraft Doppler (uplink)", "Satellite Doppler (uplink)" and "Satellite Doppler (downlink)"

It appears from this revised version that the "Downlink Doppler" was the sum of the "pure" downlink Doppler (which have to be the same for different position of the a/c) and the Doppler from the velocity of the sat to the a/c (the Doppler from the velocity of the a/c to the sat being computed separately). This answers the question of "many"!

[slats11]

End of flight scenario
Note: Given the imprecise nature of the SATCOM data, it was necessary to make some assumptions regarding pilot control inputs in order to define a search area of a practical size. These assumptions were only made for the purposes of defining a search area and there is no suggestion that the investigation authority will make similar assumptions.
The limited evidence available for MH370 was compared with the accident classes listed previously.
In the case of MH370, there were multiple redundant communications systems fitted to the aircraft (3 x VHF radios, 2 x HF radios, SATCOM system, 2 x ATC transponders). However, no radio communications were received from the aircraft after 1719.29, 7 hours prior to the last SATCOM handshake at 00:19. Analysis of the SATCOM data also showed that there were probably no large changes to the aircraft’s track after approximately 1915, about 5 hours prior to the last SATCOM handshake.
Given these observations, the final stages of the unresponsive crew/ hypoxia event type appeared to best fit the available evidence for the final period of MH370’s flight when it was heading in a generally southerly direction:


Interesting turn of phrase in first paragraph. Obviously it is not the role of ATSB to speculate on possible earlier events. The role of ATSB is locating the plane if possible. The Malaysians are responsible for investigating the psssinlr cause.

Diplomatic protocol would also preclude ATSB from speculating along lines that could embarrass Malaysia.

So ATSB is appropriately not speculating on cause. Yet they had to make some assumptions in order to define the area. Hence the most "neutral" possible hypothesis.

Hopefully the model is not too constrained such that it does not encompass alternative explanations. In particular, the document makes some comments about possible gliding distance if there were control inputs. At 90 odd km, the search areaay not be "wide" enough to cover a controlled glide.

[andrasz]

the unresponsive crew/ hypoxia event type appeared to best fit the available evidence

To paraphrase, a conscious but unresponsive and unconscious crew event scenario is indistinguishable, and it had to be one or the other

[slats11]

Yes. But likely quite different profiles after fuel exhaustion. Is 90 odd km sufficient to encompass a glide for max range? I don't think so.

There will have been careful consideration of diplomacy in this document. Also carefully worded disclaimers to minimise any criticism in case of an alternative ultimate explanation. "Hey that was not our responsibility and was not relevant to our role."

It is certain there was very careful consideration given to every comment that even hints at cause in a document of this nature.

[sandos]

I am wondering about one thing in the definion of the search area report: Balancing this was the consideration that, by the time of the final SATCOM log on message, the autopilot could have been disengaged for approximately 3 minutes and 40 seconds and the aircraft would have been descending during that periodIf the AP had been off for that long, how could the satellite system log on? Doesn't it require a directional antenna? Wouldn't a spiral dive already have developed after 3 minutes, making any directional antenna use impossible? Or is the tracking that quick?

[billslugg]

The 777 does not have a steerable antenna for the satcom. There is a low gain and a high gain antenna located on top of the fuselage which look above the horizon.

[mm43]

ATSB - Update 18 August 2014

Since the public release of the report 'MH370 – Definition of Underwater Search Areas' on 26 June 2014, the ATSB has received a number of queries about some of the technical details contained in the report. The queries have been made directly to the ATSB or through the Chief Commissioner’s blog, InFocus, on the ATSB website.

As a result of the queries, the ATSB is today releasing an updated version of the report to clarify a number of technical aspects. The changes to the report are detailed in the Addendum on the inside cover.

MH370 - Definition of Underwater Search Areas -PDF: 7.33MB- (updated 18 August 2014) (http://www.atsb.gov.au/media/5243942/ae-2014-054_mh370_-_definition_of_underwater_search_areas_18aug2014.pdf)

[AstraMike]

Hmmm... I would have assumed it was a phased array conformal antenna

[slats11]

Something that does not seem to be in the ATSB report us an analysis of civilian shipping location and course at the time of the 7th partial handshake. The information we do have is the exact time of this transmission, and the location of this arc. The location of ships in the vicinity of this arc at that time is surely available.

The ocean is huge and shipping is fairly low in density relative to other waters. However there is a reasonable amount of shipping between Australia and South Africa - the sea lane of most relevance to MH370. Many of the tracking services are only accessible via paid subscription. However one historical graphic I was able to find shows 1920 vessels in the Indian ocean for 6th April 2012, with a small minority of these in the general area of interest.

Given we have an absence of positive information (i.e. something seen), have we made all possible use of negative information. Might it help define the most likely areas to search if we use reports of what was NOT seen.

Sure any ships in the area could easily have missed something. There are countless stories from shipwrecked people ultimately rescued of not being seen by a passing ship.

However everything about the proposed new search area is based on analysis of various probabilities and compromises - which is logical given we can't search everywhere.

So could we assign (admittedly somewhat arbitrary) values for the probability of a ship seeing something had it been in the right area e.g. 777 spiralling into ocean, or a fresh debris field. Looking at the solar terminator at the time of the final handshake, it was daylight in the proposed search area. That increases the probability something would have been seen by a ship in the right place.

Can we use this information to calculate adjusted probabilities (i.e. MH370 less likely in this area as a ship in that area didn't see anything, and correspondingly other areas become more likely).

[billslugg]

"Hmmm... I would have assumed it was a phased array conformal antenna" - Astramike

You are correct Sir, and I am wrong. There is a non steerable low gain antenna on the roof of the T7 but there is also a phased array high gain antenna.

The phased array antenna is an Esterline CMC Electronics Inc (http://www.airframer.com/direct_detail.html?company=111456)(Canada) model CMA-2102SB (http://www.esterline.com/avionicssystems/en-us/productsservices/aviation/communication.aspx) (source airframer.com)

It is 60" long, 18" wide and 4.5" high, takes 69 watts, can switch the beam in 50 microseconds and offers a minimum of 9dBiC gain over 100% of the hemisphere above the horizon. Looking straight up it gives 17dBiC gain.

dBiC is decibels of gain of the antenna relative to an isotropic antenna (no gain) dealing with a circularly polarized EM field.

The gains offered by this antenna range from 2 to 7 relative to an omnidirectional antenna.

[CargoFlyer11]

MH370: Malaysian Firm Says Chinese Hackers Took Classified Data From Investigators? Computers (http://www.ibtimes.com/mh370-malaysian-firm-says-chinese-hackers-took-classified-data-investigators-computers-1663688)

Chinese hackers targeted the computers of high-ranking officials at agencies involved in the investigation of Malaysia Airlines Flight MH370, just a day after it went missing, and stole classified information, a Malaysian newspaper said Wednesday.

[WingNut60]

The West Australian newspaper is reporting today :


Fresh information suggested the jet "may have turned south" earlier than thought, Deputy Prime Minister Warren Truss said.
The detail came to light following "further refinement" of satellite data and as investigators attempted to map the plane's position during a failed attempt to contact it earlier in its flight path.


"After MH370 disappeared from the radar, Malaysia Airlines ground staff sought to make contact using a satellite phone. That was unsuccessful," he said.
"But the detailed research that's being done now has been able to... trace that phone call and help position the aircraft and the direction it was travelling."
The minister said investigators still believed MH370 was somewhere on the search zone's seventh arc, where it emitted a final satellite "handshake".
"It remains on the seventh arc - that is, there is a very, very strong view that this aircraft will be resting on the seventh arc," he said.

[sky9]

A further report in the Guardian today
MH370: satellite phone call revealed as Australia gives update on search | World news | theguardian.com (http://www.theguardian.com/world/2014/aug/28/mh370-satellite-phone-call-revealed-as-australia-gives-update-on-search#start-of-comments)

I am still surprised that there has been no comment in the ATSB report on any attempt to establish whether any passenger or crew mobile phones were left on by mistake.

[Shadoko]

"A further report in the Guardian today
MH370: satellite phone call revealed as Australia gives update on search..."
The article probably comes from this Communiqué:
MH370 Tripartite Meeting (http://www.jacc.gov.au/media/communiques/2014/august/com002.aspx)
Hope we have a full transcript, because the fact about "the revelation of a phone call" is not written in the Communiqué.

[KTVaughan]

I wonder if this lends more weight to the passenger sailboat that thought they saw the plane..............

[Chronus]

Whilst public attention is focused on the search effort, no reports are released on the criminal investigation which continues under wraps. Is it not high time the authorities release some information as to progress on this aspect of the search for answers.

[mm43]

The Inmarsat Communication Log provides the following details:-

18:39 - Ground Initiated to Air Telephony Call - Zero Duration (Not Answered)

18:39:52.907 | IOR-P10500-0-386B | IOR | 301 [BURUM NL] | 10 | P-Channel TX | 0x20 - Access Request/ Call Announcement Telephone/ Circuit-Mode Data

18:39:52.907 | IOR-P10500-0-386B | IOR | 301 [BURUM NL] | 10 | P-Channel TX | 0x33 - C-Channel Assignment (Regularity)

18:39:55.354 | IOR-3730-21000 .... | IOR | 305 [PERTH AU] | 06 | C-Channel RX | 0x30 - Call Progress - Test | 88 [BFO]

18:40:56.354 | IOR-3730-21000 .... | IOR | 305 [PERTH AU] | 06 | C-Channel RX | 0x30 - Call Progress - Channel Release | 90 [BFO]Remarks in [brackets] added by me, and a duplication of the BURUM data in the log is not reproduced here.

The unanswered call was routed via BURUM in the Netherlands. While the most relevant thing is that PERTH checked the call progress and subsequent release, which resulted in two BFO's.

[smiling monkey]

The Inmarsat Communication Log provides the following details:-
Remarks in [brackets] added by me, and a duplication of the BURUM data in the log is not reproduced here.

The unanswered call was routed via BURUM in the Netherlands. While the most relevant thing is that PERTH checked the call progress and subsequent release, which resulted in two BFO's.

That sounds great ! Now would you mind explaining the signigicance of two BFO's or even one BFO for that matter.

[hamster3null]

I'm not sure how you get 441 kt (I get ~360), but yes, that's more or less what it means. BFO 88 Hz is in line with later BFO points and it's a good indication that the turn south was completed by 18:40. I don't know if that's the point they are trying to make. This data was available from the beginning and I would've assumed that it was included in original estimates.

[no-hoper]

Back to the satcom system.There is no single high gain antenna on top center.

http://www.upload-pictures.de/bild.php/52482,highgainJ9U9R.jpg

Photos: Boeing 777-2H6/ER Aircraft Pictures | Airliners.net (http://www.airliners.net/photo/Malaysia-Airlines/Boeing-777-2H6-ER/1641369/L/&sid=7746813e9d274d5cefafd5f6a22567a1)

[slats11]

Its a bit hard to follow why the BFO analysis from the 1839 satellite call changes things now. They presumably had this information before. But the media release certainly suggests this data is being used to refine the most likely search area.

Anyway if we accept the location of the last primary contact at 1822, this fits pretty well with the first "ring" at 1825. Unlikely the plane turned south between 1822 and 1825 as this would have taken it over land.

Then a turn to the south by 1840 - not long after passing the "tip" of Sumatra.

Its hard to identify a constant course that would hit all the rings at the right time. Best fit appears to be close to due south - although the distance covered between the 2141 and 2241 rings seems a bit large.


Assuming the last ring = fuel exhaustion and crash, we know the plane flew for a defined time from last known point. We can estimate an interval of most likely speeds from the fuel load and fuel consumption. So we can estimate an interval of most likely distance covered from last known point. An earlier turn south doesn't change the known duration, nor the estimated velocity or distance covered. These stay constant.

So an earlier turn south presumably puts the crash a bit further south west along the 7th arc.

[cwatters]

Something that does not seem to be in the ATSB report us an analysis of civilian shipping location and course at the time of the 7th partial handshake. The information we do have is the exact time of this transmission, and the location of this arc. The location of ships in the vicinity of this arc at that time is surely available.

I might be totally wrong but... Although ships are required to keep adequate lookout I don't think this means they are required to have a man physically staring out to sea all the time. I believe the legal term "lookout" refers to an overall "information collecting process" that can include staring at radar. I can well imagine that particularly at night that's all they do.

So unless the aircraft appeared and was recorded on radar as it impacted I doubt anyone saw anything. Might be different if it was on fire though?