Sky-crashGuide report released
Tsamaya sentle
Joined: Apr 2001
Posts: 154
Likes: 0
From: Germany
eyeinthesky,
Read the report carefully: They saw each other.
It is impossible to guess speeds and altitudes under these circumstances, as the report clearly shows (backed up by graphics).
Other avoiding action? Difficult when approaching at a 90 degrees angle. The good reasons for following the RA have been given in this thread. Selecting other/alternative measures means you are undermining RAs which are meant as the FINAL resort!
why were the crews ... not able to look out of the window and see each other in those conditions?
It is impossible to guess speeds and altitudes under these circumstances, as the report clearly shows (backed up by graphics).
Other avoiding action? Difficult when approaching at a 90 degrees angle. The good reasons for following the RA have been given in this thread. Selecting other/alternative measures means you are undermining RAs which are meant as the FINAL resort!
Joined: Apr 2004
Posts: 49
Likes: 0
From: UK
In reading the report it is quite clear that the underlying failure was once again that of management.
I recall reading another report recently where management at an Italian airport were ultimately to blame for an accident.
Wouldn't it be nice if (for once?) senior management was made to take responsibility for their failures. Maybe one or two long prison sentences might just concentrate the mind of these people.
I speak as a senior manager within the transport industry (not aviation) where my actions are subject to very close scrutiny with very serious legal implications when I get it wrong. I am also a regular air passenger.
This does not appear to happen in the air transport industry, or am I wrong in this assumption?
I recall reading another report recently where management at an Italian airport were ultimately to blame for an accident.
Wouldn't it be nice if (for once?) senior management was made to take responsibility for their failures. Maybe one or two long prison sentences might just concentrate the mind of these people.
I speak as a senior manager within the transport industry (not aviation) where my actions are subject to very close scrutiny with very serious legal implications when I get it wrong. I am also a regular air passenger.
This does not appear to happen in the air transport industry, or am I wrong in this assumption?
Joined: Oct 2000
Posts: 301
Likes: 0
From: Somewhere probing
Kalium Chloride - that's a valid point, i.e. about the 50' - 100' being as good as a mile.
Indeed it's all the more relevant here because the controller actually cleared the TU154M down to FL350 - indeed this occured some seconds prior to the RA.
To make sense of what follows, see appendices 1&3 and 2,4,5,6,7,8,9.10
If one examines the superimposed ‘Altitude + V/S’ charts for the B757 & T154 ( in Appendix 6 ) in conjunction with the other FDR extracts versus timeline one sees the following:
From the altitude graph it would appear that, when both aircraft were ostensibly at 36000’, the T154 was actually cruising approx 50’ higher than the B757 - i.e. the former was at approx 36070’ and the latter at 36020’.
Given that both should be using 1013 ( Qff ) then one might put this variance down to the T154 using metres ( converted from feet ) plus some ‘altimeter error’ in either aircraft's altimeter.
21:34:42 – TCAS TA onboard both aircraft announces “Traffic Traffic”
21:34:49 - ATC - tells the T154 “Descend FL350”
21:34:56 – TCAS RA onboard both aircraft tells T154 “Climb” & B757 “Descend” respectively.
Nb. Inspection of the V/S graph shows:
21:34:58 - B757 - commences descent from FL360 ( i.e. within 2" of the RA ).
21:35:02 - T154 - commences descent from FL360 ( i.e. as per the ATC instruction and contradicting the TCAS RA saying that they should "Climb" )
Nb. In the full report ( the ACAS/TCAS section ) it suggest that the T154 crew commenced descent immediately when instructed by ATC ( i.e. at 21:34:49 ), however the graph on appendix 5 ( and others ) seems to contradict this wherein it shows the T154 leaving FL360 no earlier than 21:35:02, i.e. 13" after being instructed to do so by ATC.
From the gradients on the V/S graph it would appear that the initial ROD of the T154 was twice that of the B757 and accordingly at 21:35:04 – whilst having started out at a slightly higher altitude than the B757 - the T154 crosses the level of the also descending B757 ( this occurs at approx 35940’ ) and, from this point on, the T154 is always lower than the B757 ( that is until just prior to impact ).
21:35:02 – ATC - tells T154 to “Expedite Descent” ( possibly due to their delay in having commenced it in the first place ? )
21:35:07 – T154 - crew acknowledge the request to “Expedite Descent”
21:35:10 – B757 - TCAS “Increase Descent”
21:35:19 – B757 - informs ATC of "TCAS Descent”
21:35:24 – T154 - TCAS “Increase Climb”
21:35:27 – T154 - F/O “Climb it says”
21:35:28 – T154 – commences a climb, having ducked below FL350 by approx 120’.
21:35:31 – Both aircraft collide.
Whilst it’s admittedly something of conjecture, had the TU154 levelled at its assigned FL350 ( which is, after all, what the T154 crew were prioritising ) it would appear, according to the graphs, that the B757 would have passed overhead by approx 50’ ( a miss & a mile and all that ). However, as it happened, the TU154's attempt at climbing back to its assigned level made matters worse - the collision occuring at approx 34890'.
A lot of people need to do a lot a soul searching about this whole episode, but wherein lets remember that the TU154 not only failed to comply with the TCAS RA but also failed to level-off at its assigned altitude - there's a lot to be learnt from this, unfortunately.
Indeed it's all the more relevant here because the controller actually cleared the TU154M down to FL350 - indeed this occured some seconds prior to the RA.
To make sense of what follows, see appendices 1&3 and 2,4,5,6,7,8,9.10
If one examines the superimposed ‘Altitude + V/S’ charts for the B757 & T154 ( in Appendix 6 ) in conjunction with the other FDR extracts versus timeline one sees the following:
From the altitude graph it would appear that, when both aircraft were ostensibly at 36000’, the T154 was actually cruising approx 50’ higher than the B757 - i.e. the former was at approx 36070’ and the latter at 36020’.
Given that both should be using 1013 ( Qff ) then one might put this variance down to the T154 using metres ( converted from feet ) plus some ‘altimeter error’ in either aircraft's altimeter.
21:34:42 – TCAS TA onboard both aircraft announces “Traffic Traffic”
21:34:49 - ATC - tells the T154 “Descend FL350”
21:34:56 – TCAS RA onboard both aircraft tells T154 “Climb” & B757 “Descend” respectively.
Nb. Inspection of the V/S graph shows:
21:34:58 - B757 - commences descent from FL360 ( i.e. within 2" of the RA ).
21:35:02 - T154 - commences descent from FL360 ( i.e. as per the ATC instruction and contradicting the TCAS RA saying that they should "Climb" )
Nb. In the full report ( the ACAS/TCAS section ) it suggest that the T154 crew commenced descent immediately when instructed by ATC ( i.e. at 21:34:49 ), however the graph on appendix 5 ( and others ) seems to contradict this wherein it shows the T154 leaving FL360 no earlier than 21:35:02, i.e. 13" after being instructed to do so by ATC.
From the gradients on the V/S graph it would appear that the initial ROD of the T154 was twice that of the B757 and accordingly at 21:35:04 – whilst having started out at a slightly higher altitude than the B757 - the T154 crosses the level of the also descending B757 ( this occurs at approx 35940’ ) and, from this point on, the T154 is always lower than the B757 ( that is until just prior to impact ).
21:35:02 – ATC - tells T154 to “Expedite Descent” ( possibly due to their delay in having commenced it in the first place ? )
21:35:07 – T154 - crew acknowledge the request to “Expedite Descent”
21:35:10 – B757 - TCAS “Increase Descent”
21:35:19 – B757 - informs ATC of "TCAS Descent”
21:35:24 – T154 - TCAS “Increase Climb”
21:35:27 – T154 - F/O “Climb it says”
21:35:28 – T154 – commences a climb, having ducked below FL350 by approx 120’.
21:35:31 – Both aircraft collide.
Whilst it’s admittedly something of conjecture, had the TU154 levelled at its assigned FL350 ( which is, after all, what the T154 crew were prioritising ) it would appear, according to the graphs, that the B757 would have passed overhead by approx 50’ ( a miss & a mile and all that ). However, as it happened, the TU154's attempt at climbing back to its assigned level made matters worse - the collision occuring at approx 34890'.
A lot of people need to do a lot a soul searching about this whole episode, but wherein lets remember that the TU154 not only failed to comply with the TCAS RA but also failed to level-off at its assigned altitude - there's a lot to be learnt from this, unfortunately.
Last edited by Devils Advocate; 24th May 2004 at 12:35.
Joined: Aug 1999
Posts: 1,050
Likes: 4
From: England
Just to reiterate a point I made in previous threads about this accident:
Please remember to differentiate between:
(a) Not complying with an RA.
and
(b) Manouevering in the opposite sense to an RA.
There are circumstances where not complying an RA may be necessary. This is recognised by manufacturers and authorities and hence final authority to follow or disregard an RA rests with flight crew.
However, manouevres in the opposite sense are forbidden.
Excluding the circumstances preceeding the TCAS event itself, this accident was NOT caused by the failure of the T154 to follow its RAs.
It was caused by the T154 manoeuvering opposite its RAs, not just once, which the strengthened 'Increase Descent' RA issued by the B757 would probably have dealt with, but twice.
Anyone that thinks I am splitting hairs is missing the point. The golden rule of surviving a TCAS event is 'Don't Manoeuvre Opposite'.
Please remember to differentiate between:
(a) Not complying with an RA.
and
(b) Manouevering in the opposite sense to an RA.
There are circumstances where not complying an RA may be necessary. This is recognised by manufacturers and authorities and hence final authority to follow or disregard an RA rests with flight crew.
However, manouevres in the opposite sense are forbidden.
Excluding the circumstances preceeding the TCAS event itself, this accident was NOT caused by the failure of the T154 to follow its RAs.
It was caused by the T154 manoeuvering opposite its RAs, not just once, which the strengthened 'Increase Descent' RA issued by the B757 would probably have dealt with, but twice.
Anyone that thinks I am splitting hairs is missing the point. The golden rule of surviving a TCAS event is 'Don't Manoeuvre Opposite'.
Uneasy Pleistocene Leftover
Joined: Feb 2003
Posts: 320
Likes: 0
From: Gone, but not forgotten apparently?! All forums marked "Private"...
Capt Pit Bull, while most of us will appreciate what you mean, just how would this translate into all the non-English mother-tongue administration rules?
By (a), I take it you mean ignoring a RA as opposed to (b) doing the exact opposite ie ascending instead of descending. Just how could and would these interpretations translate?
Those who write the rules should KEEP THEM SIMPLE and UNAMBIGIOUS so that there IS NO DOUBT!
Please remember to differentiate between:
(a) Not complying with an RA.
and
(b) Manouevering in the opposite sense to an RA.
(a) Not complying with an RA.
and
(b) Manouevering in the opposite sense to an RA.
Those who write the rules should KEEP THEM SIMPLE and UNAMBIGIOUS so that there IS NO DOUBT!
Pegase Driver
Joined: May 1997
Aviation Qualifications: ATCO
Posts: 4,452
Likes: 1,164
From: Europe
The ( quite good ) report goes in great details to explain certain things but pass very fast on others.
Some of them I would like to have seen mentionned :
The name RA for "Resolution Advisory " was choosen by the FAA to limit the liability of the US manufacturer of the Software ( The Mitre Corporation ) Only in the US was it known implicitely that although for legal reason we call it an advisory, it is in fact a command .
Unfortunately when you translate the manuals and the procedures in other languages, Advisories remain advisories.
TCAS was made by the US for the US... (on one of its first report Eurocontrol mentionned that TCAS II logic behaved badly in European types of en-route airspace...)
To follow an RA you need to be 2 in the cockpit : one that look outside to identify the intruder, the other that look at the display and follow the needle to keep it in the green.
Attempting to do both at the same time results in large deviations...
The 757 pilot was alone for a great part of the RA..( I do not believe this played a role in this case , but it show us that Murphy' s law is always valid...)
The unsheduled late inbound to to Friedrichshafen was, from an ATC point of view the most important link of the chain of event that led to this collision.
The paramount importance of this fact is missing
This forced the controller to look into two different scopes a few meters appart, and monitor alone two different active fequencies whit degraded telephone system and faulty back ups. This led to the late detection of the conflict and the late decent clearance to the TU.( which was the action most controllers would have taken since the TU was pllanned and coordinated with the adjacent centre at 350 and the DHL had just reached his cruising altitude ) .
.
Various Management levels put the controller into that position that night. This later also cost him his life.
Finally to "Die Skyguide" : first post and already full of hate..
For info in most ATC centres at night before that day, SMOP was either planned by Management ( still is in some places, especially in TWR-APP )or at least tolerated.
Some of them I would like to have seen mentionned :
The name RA for "Resolution Advisory " was choosen by the FAA to limit the liability of the US manufacturer of the Software ( The Mitre Corporation ) Only in the US was it known implicitely that although for legal reason we call it an advisory, it is in fact a command .
Unfortunately when you translate the manuals and the procedures in other languages, Advisories remain advisories.
TCAS was made by the US for the US... (on one of its first report Eurocontrol mentionned that TCAS II logic behaved badly in European types of en-route airspace...)
To follow an RA you need to be 2 in the cockpit : one that look outside to identify the intruder, the other that look at the display and follow the needle to keep it in the green.
Attempting to do both at the same time results in large deviations...
The 757 pilot was alone for a great part of the RA..( I do not believe this played a role in this case , but it show us that Murphy' s law is always valid...)
The unsheduled late inbound to to Friedrichshafen was, from an ATC point of view the most important link of the chain of event that led to this collision.
The paramount importance of this fact is missing
This forced the controller to look into two different scopes a few meters appart, and monitor alone two different active fequencies whit degraded telephone system and faulty back ups. This led to the late detection of the conflict and the late decent clearance to the TU.( which was the action most controllers would have taken since the TU was pllanned and coordinated with the adjacent centre at 350 and the DHL had just reached his cruising altitude ) .
.
Various Management levels put the controller into that position that night. This later also cost him his life.
Finally to "Die Skyguide" : first post and already full of hate..
For info in most ATC centres at night before that day, SMOP was either planned by Management ( still is in some places, especially in TWR-APP )or at least tolerated.
Joined: Apr 2001
Posts: 1,839
Likes: 0
From: Adrift upon the tides of fate
Upon reflection, one of the things that I took away from reading this is how compliant we are all becoming in the erosion of safety in the industry.
IMHO, a lot of stuff goes on now that just wouldn't have in times past. If you want to get anywhere, you just have to put up with it, or be labled as 'belligerent' or 'change resistant'. Manning levels, fuel policies etc are all being squeezed as the dollar keeps becoming more and more important.
Maybe I'm just getting old........
IMHO, a lot of stuff goes on now that just wouldn't have in times past. If you want to get anywhere, you just have to put up with it, or be labled as 'belligerent' or 'change resistant'. Manning levels, fuel policies etc are all being squeezed as the dollar keeps becoming more and more important.
Maybe I'm just getting old........
Joined: Mar 2000
Posts: 301
Likes: 0
From: Obvious
Extract from "Air Safety Week" issue dated 31 May 04
Big sky – little sky
This tragic case should end the “big sky” notion that the air traffic controller’s ultimate savior will be the vastness of the airspace and the smallness of the airplanes. Satellite navigation accuracy, RVSM and automated, computerized flight controls have combined to kill the big sky theory. Dense clusters of airplanes headed in both directions and separated by half the 2,000-ft. vertical separation of yesteryear can make the sky along preferred tracks very small. Constructing a mid-air collision is now as straightforward as putting two trains on one set of tracks. In this accident, the controller tried to close the switches after the locomotives had passed by on their intersecting tracks. Technical refinement of navigational tracking accuracy and height-holding precision has increased the efficiency of preferred airspace use, but at an increase in the risk of mid-air collisions. Controllers now need to become sensitized to the “small sky” theory. For pilots, there is a reason why the “see and avoid” principle of collision avoidance may not be sufficient.
The ‘strange attractor’
For pilots, sighting the threat can lead to a greater hazard that might be called the “strange attractor.” This term describes a scenario where two pilots at night without a defined horizon become fixated on keeping each other’s lights in sight. In this mid-air collision, note that the TU154 changed its heading from
264º to 274º about two seconds after sighting the B757. The result can be a mutually maintained constant relative bearing on each other, which can greatly increase the risk of collision or near-miss. The control inputs by both pilots in the last few seconds tends to bear out the sudden arousal of imminent collision as
each “target” blossomed in size.
The two crews had been "visual" for almost 30 seconds. That is a long time in aerospace terms. Pilots disregarding a coordinated TCAS RA while attempting visual avoidance do so at their peril.
This tragic case should end the “big sky” notion that the air traffic controller’s ultimate savior will be the vastness of the airspace and the smallness of the airplanes. Satellite navigation accuracy, RVSM and automated, computerized flight controls have combined to kill the big sky theory. Dense clusters of airplanes headed in both directions and separated by half the 2,000-ft. vertical separation of yesteryear can make the sky along preferred tracks very small. Constructing a mid-air collision is now as straightforward as putting two trains on one set of tracks. In this accident, the controller tried to close the switches after the locomotives had passed by on their intersecting tracks. Technical refinement of navigational tracking accuracy and height-holding precision has increased the efficiency of preferred airspace use, but at an increase in the risk of mid-air collisions. Controllers now need to become sensitized to the “small sky” theory. For pilots, there is a reason why the “see and avoid” principle of collision avoidance may not be sufficient.
The ‘strange attractor’
For pilots, sighting the threat can lead to a greater hazard that might be called the “strange attractor.” This term describes a scenario where two pilots at night without a defined horizon become fixated on keeping each other’s lights in sight. In this mid-air collision, note that the TU154 changed its heading from
264º to 274º about two seconds after sighting the B757. The result can be a mutually maintained constant relative bearing on each other, which can greatly increase the risk of collision or near-miss. The control inputs by both pilots in the last few seconds tends to bear out the sudden arousal of imminent collision as
each “target” blossomed in size.
The two crews had been "visual" for almost 30 seconds. That is a long time in aerospace terms. Pilots disregarding a coordinated TCAS RA while attempting visual avoidance do so at their peril.
Joined: Jun 2003
Posts: 69
Likes: 1
From: BRU
Suggestion for better TCAS logic
I am not a pilot, but have followed this mid-air accident closely. I have the following suggestion for the TCAS logic, and would like comments on this.
In my opinion, the flaw is that the TCAS software assumes that the other party will follow the RA. This could be improved upon. I suggest that if TCAS realises that the other aircraft is not following the RA, but rather moving opposite to it (for whatever reason):
- assume that the other aircraft will not follow any RA in this situation
- issue a reversing sense RA to the non-complying aircraft so that the RA is consistent with its actual vertical speed
- issue a corresponding RA to the complying aircraft to achieve maximum separation
This logic would have avoided the Überlingen mid-air as follows:
(1) First RAs issued (descend to B757, climb to Tu154)
(2) after several seconds both TCAS computers realise that the Tu154 is not complying, as it is in fact descending.
(3) The computers should now re-calculate the forecast for the altitude at the potential collision location of the noncomplying aircraft, the Tu154, assuming that the current vertical speed of the Tu154 is maintained. The result of this calculation would have been a projected altitude of somewhere between 34,500 and 35,500 feet.
(4) The computer now re-calculates the optimal RA for the B757, which would have likely been a reversal sense RA ("Climb, climb now").
(5) Issue both RAs: "Climb, climb now" to the B757 (possibly followed up with a "Increase climb, increase climb") and a "Descend, descend now" to the Tu154 (possibly followed up with a "Increase descend, increase descend").
(6) Now there would have been no more confusion for the Tu154 crew, and while the B757 pilots might have been sweating while following the reversal sense RA, a safe vertical separation would have been achieved.
OK, comments please. Note that I do agree with many other additional safety measures suggested, and this only one further suggestion. It would have helped in this case, but might it lead to a higher probability of accidents in other cases?
Profit Max.
In my opinion, the flaw is that the TCAS software assumes that the other party will follow the RA. This could be improved upon. I suggest that if TCAS realises that the other aircraft is not following the RA, but rather moving opposite to it (for whatever reason):
- assume that the other aircraft will not follow any RA in this situation
- issue a reversing sense RA to the non-complying aircraft so that the RA is consistent with its actual vertical speed
- issue a corresponding RA to the complying aircraft to achieve maximum separation
This logic would have avoided the Überlingen mid-air as follows:
(1) First RAs issued (descend to B757, climb to Tu154)
(2) after several seconds both TCAS computers realise that the Tu154 is not complying, as it is in fact descending.
(3) The computers should now re-calculate the forecast for the altitude at the potential collision location of the noncomplying aircraft, the Tu154, assuming that the current vertical speed of the Tu154 is maintained. The result of this calculation would have been a projected altitude of somewhere between 34,500 and 35,500 feet.
(4) The computer now re-calculates the optimal RA for the B757, which would have likely been a reversal sense RA ("Climb, climb now").
(5) Issue both RAs: "Climb, climb now" to the B757 (possibly followed up with a "Increase climb, increase climb") and a "Descend, descend now" to the Tu154 (possibly followed up with a "Increase descend, increase descend").
(6) Now there would have been no more confusion for the Tu154 crew, and while the B757 pilots might have been sweating while following the reversal sense RA, a safe vertical separation would have been achieved.
OK, comments please. Note that I do agree with many other additional safety measures suggested, and this only one further suggestion. It would have helped in this case, but might it lead to a higher probability of accidents in other cases?
Profit Max.
Tsamaya sentle
Joined: Apr 2001
Posts: 154
Likes: 0
From: Germany
Profit Max,
we are always talking here of the RA as a (i.e. THE!) last resort. As such we are talking of a matter of seconds. If a once given advisory was changed this would probably cause more confusion than help - especially when we continue toying with ideas and think of a situation where the crew who had not complied in the first place (here: the TU154) suddenly make up their mind and start following the (original) RA (not an unlikely scenario, as we know now that a short discussion about the RA had taken place in the TU154 cockpit). This would then create yet another RA, and so on, rendering TCAS unnecessary/useless.
we are always talking here of the RA as a (i.e. THE!) last resort. As such we are talking of a matter of seconds. If a once given advisory was changed this would probably cause more confusion than help - especially when we continue toying with ideas and think of a situation where the crew who had not complied in the first place (here: the TU154) suddenly make up their mind and start following the (original) RA (not an unlikely scenario, as we know now that a short discussion about the RA had taken place in the TU154 cockpit). This would then create yet another RA, and so on, rendering TCAS unnecessary/useless.
Pegase Driver
Joined: May 1997
Aviation Qualifications: ATCO
Posts: 4,452
Likes: 1,164
From: Europe
Profit Max, what you described already exists and was planned to work with version 7. ( sense reversal RA ) It did not work in Ueberlingen for a lot of reasons which have been in the meantime identified. A change proposal (CP )to rectify the CAS logic was issued by Eurocontrol , evn prior Ueberlingen, it is called CP112.
The problem is that the FAA who owns the software has disbanded the TCAS programme and there is nobody to work on the software change.
The problem is that the FAA who owns the software has disbanded the TCAS programme and there is nobody to work on the software change.
Joined: Jan 2001
Posts: 2,044
Likes: 0
From: UK
I think we need to be careful of adding "more technology" to solve a "problem", when the current technology should have prevented it, but other (Human) factors prevented it.
The "Human Factors" here were poor worldwide management in ensuring that all Flt Crew and Manuals were at one over obeying the RA rather than ATC. That was quickly corrected... so the problem should not recur. (NB of course TCAS works perfectly well if one aircraft ignores the RA).
As has also been stated, the RA is a last resort. Why did we get to the last resort? Again, Human failings - largely Management again with the staffing levels and disabled technology at ATC. All this combined with one small error and the tragic results.
Let's concentrate on correcting the identified error's (largely done?), rather than inventing more technology to cope if this particular set of c*ck-ups occurred again.
NoD
The "Human Factors" here were poor worldwide management in ensuring that all Flt Crew and Manuals were at one over obeying the RA rather than ATC. That was quickly corrected... so the problem should not recur. (NB of course TCAS works perfectly well if one aircraft ignores the RA).
As has also been stated, the RA is a last resort. Why did we get to the last resort? Again, Human failings - largely Management again with the staffing levels and disabled technology at ATC. All this combined with one small error and the tragic results.
Let's concentrate on correcting the identified error's (largely done?), rather than inventing more technology to cope if this particular set of c*ck-ups occurred again.
NoD
