AF447
Join Date: Jun 2009
Location: Florida
Posts: 17
Likes: 0
Received 0 Likes
on
0 Posts
Me Myself:
I don't want to get drawn into a prolonged discussion on this, but all I'm saying is that I don't believe that your personal feelings about a declining safety culture are necessarily backed up by the evidence of accident stastics. The point I was making is that by selecting arbitrary start dates and periods for the accident statistics quoted, very different pictures can be painted of the 'change' in accident rate. (Why not start in 1960 and divide into 10 year periods? - again, a completely different 'trend' might emerge; the periods selected were clearly chosen to give the worst possible impression).
It is very likely that in a strict statistical sense, the number of crashes in recent years tells you nothing meaningful about AFs safety record. (As in 'the null hypothesis that AFs safety record has got worse was not proved to a confidence interval of 95%' [or whatever]).
You can have your own anecdotal evidence about a decline in safety culture, and that is what you should probably base your arguments on. After all, that user feedback is what the Flight Safety dept of any airline should focus on, surely? (I mean, they don't work harder only if they've had an accident, do they?)
As I say, I don't have an axe to grind about this - just pointing out that the statistical 'evidence' being touted for a decline in AF safety isn't very
meaningful.
We've wandered away from AF447 a bit. I'm done on this subject, anyway.
This is really far streched mate !! The last accident was in 1969 and nothing happened until 1988 which makes it a good 19 years.
The question that needs to be asked is :
How come we went from the horror show of the 1960's to the pretty good track record of the 70's and 80's.
I can partly answer that one. 2 of these 1960's crashes were the direct result of very poor CRM. I can remember flying with skippers who were F/O's during that period of time and you wouldn't believe the stories they had to tell.
Obviously, things were done during these 19 years and then..........well, the plot was lost.
Fast growth, if a factor, is not acceptable. AF isn't the only airline that grew fast in the last 20 years.
The question that needs to be asked is :
How come we went from the horror show of the 1960's to the pretty good track record of the 70's and 80's.
I can partly answer that one. 2 of these 1960's crashes were the direct result of very poor CRM. I can remember flying with skippers who were F/O's during that period of time and you wouldn't believe the stories they had to tell.
Obviously, things were done during these 19 years and then..........well, the plot was lost.
Fast growth, if a factor, is not acceptable. AF isn't the only airline that grew fast in the last 20 years.
It is very likely that in a strict statistical sense, the number of crashes in recent years tells you nothing meaningful about AFs safety record. (As in 'the null hypothesis that AFs safety record has got worse was not proved to a confidence interval of 95%' [or whatever]).
You can have your own anecdotal evidence about a decline in safety culture, and that is what you should probably base your arguments on. After all, that user feedback is what the Flight Safety dept of any airline should focus on, surely? (I mean, they don't work harder only if they've had an accident, do they?)
As I say, I don't have an axe to grind about this - just pointing out that the statistical 'evidence' being touted for a decline in AF safety isn't very
meaningful.
We've wandered away from AF447 a bit. I'm done on this subject, anyway.
Join Date: Jan 2008
Location: Berlin
Age: 56
Posts: 97
Likes: 0
Received 0 Likes
on
0 Posts
That "Spiegel" Report
Just to make that clear, in my last post I was just paraphrasing what a German language news story said, not my opinion, as I don't have one regarding this particular topic here.
As for the report itself: I wouldn't say the "Spiegel" report shows anti-French bias. It suffers from bad reporting. Pulling a three year old report out of the file and adding it to the mix of seemingly arbitrary and/or completely unrelated numbers for some "perspective" (even if there might not be any) is typical of cheap, sloppy and sensationalist reporting.
As for the report itself: I wouldn't say the "Spiegel" report shows anti-French bias. It suffers from bad reporting. Pulling a three year old report out of the file and adding it to the mix of seemingly arbitrary and/or completely unrelated numbers for some "perspective" (even if there might not be any) is typical of cheap, sloppy and sensationalist reporting.
Join Date: Jan 2008
Location: Pretty far away
Posts: 316
Likes: 0
Received 0 Likes
on
0 Posts
I don't believe that your personal feelings about a declining safety culture are necessarily backed up by the evidence of accident stastics.
Last edited by Me Myself; 11th Jul 2009 at 18:23.
Join Date: Jun 2009
Location: ARN / STO
Age: 67
Posts: 9
Likes: 0
Received 0 Likes
on
0 Posts
BEA interrim report
On page 47 it is quoted:
The company that operates the satellite used by AF447 has provided the traces of the messages transmitted to the aircraft and seen by the satellite.
Why isnīt the log attached ? Why didnīt BEA take this into their calculation ?
The NOC-log (telemetry-transponder data) from Inmarsat would also include crucial data type strength of transmission signal an timing of packages etc.
As the infamous gap of 31 seconds EXISTS in the report, it should ALSO included the telemetry of all these ACARS. If, for example the signalstrength is fixed all the way with a small variable, we could directly calculate that the a/c kept is fligth level in bad weather, but if signal was degrading all the time we could calculate from what timing the degradation occurs and pinpoint approx at what ACARS-message that the a/c starts to divert.
My impression is that BEA has a copy of the telemetry-read out for this date from Inmarsat Noc, otherwise they should ask for a copy a.s.a.p.
The company that operates the satellite used by AF447 has provided the traces of the messages transmitted to the aircraft and seen by the satellite.
Why isnīt the log attached ? Why didnīt BEA take this into their calculation ?
The NOC-log (telemetry-transponder data) from Inmarsat would also include crucial data type strength of transmission signal an timing of packages etc.
As the infamous gap of 31 seconds EXISTS in the report, it should ALSO included the telemetry of all these ACARS. If, for example the signalstrength is fixed all the way with a small variable, we could directly calculate that the a/c kept is fligth level in bad weather, but if signal was degrading all the time we could calculate from what timing the degradation occurs and pinpoint approx at what ACARS-message that the a/c starts to divert.
My impression is that BEA has a copy of the telemetry-read out for this date from Inmarsat Noc, otherwise they should ask for a copy a.s.a.p.
Join Date: Jun 2009
Location: Europe
Posts: 10
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by ART-DECO
My impression is that BEA has a copy of the telemetry-read out for this date from Inmarsat Noc, otherwise they should ask for a copy a.s.a.p.
Originally Posted by French version p. 50
lécart observé entre le message de 2 h 13 min 14 et celui de 2 h 13 min 45
est dû, au moins en partie, ā une interruption temporaire du lien de communication entre lavion et le satellite,
est dû, au moins en partie, ā une interruption temporaire du lien de communication entre lavion et le satellite,
Yet my assumption proved reasonable, as the French version states on p. 71:
il n'y a pas eu de communication téléphoniques par satellite entre l'avion et le sol.
So they probably had Mobile OnAir installed, which is a satelite linked GSM picocell, allowing for inflight use of a cell phone for calls, SMS and internet via GPRS. Source: http://corporate.airfrance.com/index.php?id=communiques_detail&tx_ttnews[tt_news]=2699&L=1&no_cache=1
But even if there actually were no phone calls (possible, during night), it would be interesting when the last SMS got transmitted or received, when the last internet packet got transmitted, and exactly when INMARSAT saw the picocell on board the very last time. This would allow to differenciate the case of more ACARS messages not being sent because there weren't anymore, and the case that the satellite link broke down.
Last edited by Robin42; 11th Jul 2009 at 19:57. Reason: Messed up the Air France link
Join Date: Jun 2009
Location: ARN / STO
Age: 67
Posts: 9
Likes: 0
Received 0 Likes
on
0 Posts
re: Robin42.
Another detail is the hf communication.
Around the Intol area, Stockholm radio has a class-2 area of reception, and AF OCC used them the same morning but several hours later in a atempt to reach 447. All statements of poor hf is only with Dakar to my knowledge, and Dakar also didnīt have the flightplan, it could pinpoint that Dakar had a comms-breakdown at a real bad timing as well.
Service-area:
Stockholmradio aero
Propagation forecast:
http://www.stockholmradio.se/aero/so..._MAY-JUL09.pdf
Another detail is the hf communication.
Around the Intol area, Stockholm radio has a class-2 area of reception, and AF OCC used them the same morning but several hours later in a atempt to reach 447. All statements of poor hf is only with Dakar to my knowledge, and Dakar also didnīt have the flightplan, it could pinpoint that Dakar had a comms-breakdown at a real bad timing as well.
Service-area:
Stockholmradio aero
Propagation forecast:
http://www.stockholmradio.se/aero/so..._MAY-JUL09.pdf
Join Date: Mar 2001
Location: us
Posts: 694
Likes: 0
Received 0 Likes
on
0 Posts
FLY400, thanks for the clarification and the link. As I read the description, the first step is that the crew of the plane must first logon to the system, and after logon is successful, then contact is initiated by the ground station, which can query the plane periodically without further involvement of the crew.
If so, my sense is that the crew tried to logon with Dakar three times, and each time the logon failed because Dakar had no flight plan filed for AF447. I would not think the system would allow Dakar to initiate these three contacts if Dakar had no AF447 flight plan within its system. (Though I would expect that the system at Dakar would record logon failures, and that is where the record originates.)
If so, my sense is that the crew tried to logon with Dakar three times, and each time the logon failed because Dakar had no flight plan filed for AF447. I would not think the system would allow Dakar to initiate these three contacts if Dakar had no AF447 flight plan within its system. (Though I would expect that the system at Dakar would record logon failures, and that is where the record originates.)
Join Date: Jan 2008
Location: Pretty far away
Posts: 316
Likes: 0
Received 0 Likes
on
0 Posts
But even if there actually were no phone calls (possible, during night), it would be interesting when the last SMS got transmitted or received
Join Date: Jun 2009
Location: ARN / STO
Age: 67
Posts: 9
Likes: 0
Received 0 Likes
on
0 Posts
Bea interrim report #2:
Could someone here explain why the 3rd ADS-C attempt 2:01 didnīt give a GPS-position, and why BEA has removed all 3 attempts from the ACARS-list?
Clearly the 02:01Z was via ACARS as they were out of VHF-range.
Copied from BEA:s intrerim-report:
1.9.2 Coordination between the control centres
At 1 h 46, the DAKAR controller asked the ATLANTICO controller for further information regarding flight AF447 since he had no flight plan. The ATLANTICO controller provided the following elements: A332, from SBGL to LFPG, SELCAL: CPHQ.
The DAKAR OCEANIC Regional Control Centre created the flight plan and activated it. The result of this was to generate a virtual flight following the planned trajectory in the DAKAR FIR between TASIL and POMAT. There was no radio contact between AF447 and DAKAR, nor any ADS-C connection. The flight remained virtual.
1.16.2.1 ATC messages
No ATC messages were received or transmitted by F-GZCP. Only three attempts were made to connect up to the Dakar centre ADS-C system and were recorded on 1st June at 1 h 33, 1 h 35 and 2 h 01. The three requests were refused with a FAK4 code, meaning that the control system had detected the absence of a flight plan for this aircraft or that there was a mismatch between the flight plan filed for this registration number, the flight number and the reported position.
Clearly the 02:01Z was via ACARS as they were out of VHF-range.
Copied from BEA:s intrerim-report:
1.9.2 Coordination between the control centres
At 1 h 46, the DAKAR controller asked the ATLANTICO controller for further information regarding flight AF447 since he had no flight plan. The ATLANTICO controller provided the following elements: A332, from SBGL to LFPG, SELCAL: CPHQ.
The DAKAR OCEANIC Regional Control Centre created the flight plan and activated it. The result of this was to generate a virtual flight following the planned trajectory in the DAKAR FIR between TASIL and POMAT. There was no radio contact between AF447 and DAKAR, nor any ADS-C connection. The flight remained virtual.
1.16.2.1 ATC messages
No ATC messages were received or transmitted by F-GZCP. Only three attempts were made to connect up to the Dakar centre ADS-C system and were recorded on 1st June at 1 h 33, 1 h 35 and 2 h 01. The three requests were refused with a FAK4 code, meaning that the control system had detected the absence of a flight plan for this aircraft or that there was a mismatch between the flight plan filed for this registration number, the flight number and the reported position.
Join Date: Jul 2009
Location: EPWA
Age: 65
Posts: 86
Likes: 0
Received 0 Likes
on
0 Posts
pilots and security statistics
BryceM: .... After all, that user feedback is what the Flight Safety dept of any airline should focus on, surely? (I mean, they don't work harder only if they've had an accident, do they?)
BryceM: ... just pointing out that the statistical 'evidence' being touted for a decline in AF safety isn't very meaningful.
I am observing this thread for the last 4 weeks. I have learned a lot on intricacies of the planes and procedures. Many hypothesies were proposed and some seem to look more probable than the others. Yet without more hard facts from FDR/CVR we are left in the dark.
At the same time the thread has reached a point where it started scratching the surface of the delicate subject of the pilots being humans spending part of their time flying the planes and the other part being a part of the big commercial corporation. These two lives do mingle and this may influence safety.
And without any prejudice what or who has played an important role in critical sequence this may be important factor towards understanding what has caused AF477 down.
For all the AB automation it is the PF who is called for when there is a problem. And for any foreseeable future passengers will prefer to fly with qualified pilots onboard
Wojtek
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes
on
0 Posts
@Art-Deco
1.. They are not ACARS messages. Those ACARS messages were transmitted via SATCOM to Air France Operations/Maintenance. The crew also rec'd METAR / TAF info ex Paris by the same means.
2.. The ADS-CPDLC uses SATCOM also, but the initial contact with an ATS is made by the crew. Once log on is established the ATC select the info they require etc..
Here are DAKAR's requirements to establish ADS-CPDLC:-
NOTE: AF447 was not ADS-CPDLC with Brazil ATC and the 20 minute Log On requirement was in effect.
It is obvious that DAKAR ATC failed to make provision in their ADS-C system for an a/c they had accepted on a Virtual Flight Plan - hence the Log On failures. The interaction between ATLANTICO and DAKAR (or lack of it) may ultimately have a significant bearing on where the finger points.
mm43
Could someone here explain why the 3rd ADS-C attempt 2:01 didnīt give a GPS-position, and why BEA has removed all 3 attempts from the ACARS-list?
Clearly the 02:01Z was via ACARS as they were out of VHF-range.
Clearly the 02:01Z was via ACARS as they were out of VHF-range.
2.. The ADS-CPDLC uses SATCOM also, but the initial contact with an ATS is made by the crew. Once log on is established the ATC select the info they require etc..
Here are DAKAR's requirements to establish ADS-CPDLC:-
Connection procedure in the DAKAR FIR (DAKAR land and DAKAR ocean)
The first connection with the system is made by the crew. For flights entering the DAKAR control region from an FIR not equipped with CPDLC, the DAKAR control centre demands the connection at least twenty minutes before entry into the DAKAR FIR. For flights from a FIR equipped with CPDLC, the first connection must occur five minutes before entry into the DAKAR FIR.
The first connection with the system is made by the crew. For flights entering the DAKAR control region from an FIR not equipped with CPDLC, the DAKAR control centre demands the connection at least twenty minutes before entry into the DAKAR FIR. For flights from a FIR equipped with CPDLC, the first connection must occur five minutes before entry into the DAKAR FIR.
It is obvious that DAKAR ATC failed to make provision in their ADS-C system for an a/c they had accepted on a Virtual Flight Plan - hence the Log On failures. The interaction between ATLANTICO and DAKAR (or lack of it) may ultimately have a significant bearing on where the finger points.
mm43
Last edited by mm43; 12th Jul 2009 at 23:27. Reason: added sig
With the entry of a number of new posters, the thread is now chasing it's own tail with all these comments about ACARS, SATCOM and personal opinions about cultural differences and flight safety and micro-interpretations regarding tiny nuances of the BEA report and the safety records of various airlines. These issues were either dealt with a thousand posts ago, or are irrelevant and inconclusive. Given that those who were contributing good, creative stuff that got us somewhere have gone silent, I suspect for most these fly****-in-pepper mini-me points are mind-numbingly trivial. Can we not do better than this?
Join Date: Jun 2009
Location: I am where I am and that's all where I am.
Posts: 660
Likes: 0
Received 0 Likes
on
0 Posts
WhyIsThereAir
That raises another can of worms. It is reported that the plane was over MTOW as it left the gate and was just under it as it lifted off. Does that suggest that the aircraft was overloaded and the crew did its best to handle the situation despite that fact? The fuel situation may well have colored the crew's election to plow on roughly straight line into the mess.
I hope the hole in my conjecture does not mean the pilot lacks the authority to cut some of the load from the plane to get more fuel aboard. That sounds absurd.
JD-EE
I read the report of the IB crew as saying they had decided to take on the additional 2000 kg of fuel. AF of course took on about half that.
I hope the hole in my conjecture does not mean the pilot lacks the authority to cut some of the load from the plane to get more fuel aboard. That sounds absurd.
JD-EE
Join Date: Jun 2009
Location: SPAIN
Age: 65
Posts: 50
Likes: 0
Received 0 Likes
on
0 Posts
If we was able to return backward
Hi,
BEA said that "nothing" so far can link the disaster to failure of the Pitot probes.
If we was able to return backward .. and have not the pitot problem (a wizzard keeping them working) ... can the AF447 disaster occur so certainly ?
Bye.
BEA said that "nothing" so far can link the disaster to failure of the Pitot probes.
If we was able to return backward .. and have not the pitot problem (a wizzard keeping them working) ... can the AF447 disaster occur so certainly ?
Bye.
Join Date: Jul 2009
Location: EPWA
Age: 65
Posts: 86
Likes: 0
Received 0 Likes
on
0 Posts
PJ2: ...With the entry of a number of new posters, the thread is now chasing it's own tail with all these comments about [...] flight safety and [...] the safety records of various airlines. These issues were either dealt with a thousand posts ago, or are irrelevant and inconclusive.
Sir, i do have highest respect for your extensive knowledge on the subject yet i can not concur with your statement that comments on flying safety are ' irrelevant and inconclusive '. One can talk about conclusive when discussing technical points which can be proven by referencing to other identical or similar planes/solutions. Yet there are some inconclusive subjects which are as important as the technical ones.
In principle planes crash because of :
- plane (technical) problems - many possibilities have been discussed earlier, also with your, highly respectable, input. Seems that the F-GZCP plane was fit for the flight and whatever technical problems encountered the plane itself was not a direct cause for the crash. Still there are open issues about AI approach to flight automation which were thoroughly discussed with many improvements proposed
- environment (weather) problems - discussed earlier in the thread with excellent Tim's & al input. The weather was bad yet typical for the time and area. Planes do fly the same route every day.
- pilots error - just a possibility but as it is extremely delicate matter to be discussed among pilots this part was not so carefully discussed. As tough it is, still it has to be analysed because it MAY reveal aspects that MAY have lead to the tragic crash. Especially that those aspects are sometimes referred to as soft or indirect ('economy' training being already discussed, corporate culture discussed a little) - still having crucial influence on many things. The water also can drill the stone if given enough time. Last few days has brought some light on this subject and even more input is needed to grasp the problem. Even the captain is reporting to someone... If flying crew made mistakes lets identify them and teach others to avoid them. If roots of the mistakes are in the corporate politics lets identify them and imply change to the corporate ethics/policies before some other pilot fill find himself and his plane in problems
PJ2: ...Can we not do better than this?
Join Date: Jun 2009
Location: US
Posts: 497
Likes: 0
Received 0 Likes
on
0 Posts
With no new data rehashing old information seems pretty meaningless at this point. Hopefully the black boxes will be found or this will be all we will ever know about what happened.
WojtekSz
We tried this out in the Techical forum and it seemed to work best there.
If you place this kind of discussion within a specific accident thread it begins to smell like the blame game before the wreckage is even recovered and they should have seen it coming the word (they) being a fill-in-the-blank group
One can talk about conclusive when discussing technical points which can be proven by referencing to other identical or similar planes/solutions. Yet there are some inconclusive subjects which are as important as the technical ones.
If you place this kind of discussion within a specific accident thread it begins to smell like the blame game before the wreckage is even recovered and they should have seen it coming the word (they) being a fill-in-the-blank group
Join Date: Jan 2008
Location: southwest
Age: 78
Posts: 287
Likes: 0
Received 0 Likes
on
0 Posts
Additional fuel
If AF447 took extra fuel it would at least give a clue that they were prepared to deviate around the weather.
If a flight didn't take more fuel it tells us nothing. The extra fuel required to pass 50nm to one side and then rejoin the original track is almost negligable: just a fraction of the en-route contingency already on board.
If a flight didn't take more fuel it tells us nothing. The extra fuel required to pass 50nm to one side and then rejoin the original track is almost negligable: just a fraction of the en-route contingency already on board.