Mid-Air Collision over Southern Germany (merged)
Join Date: Sep 2000
Location: Luxembourg
Posts: 34
Likes: 0
Received 0 Likes
on
0 Posts
Alpha Leader
You may be right about the level change requirement at trasadingen but my main point was that the TU154 should never have been accepted at FL360 in the first place with crossing traffic at the same level. Swiss ATC should have refused to accept it at that level and requested the German ATC to change its level before handover!!!!!!!
You may be right about the level change requirement at trasadingen but my main point was that the TU154 should never have been accepted at FL360 in the first place with crossing traffic at the same level. Swiss ATC should have refused to accept it at that level and requested the German ATC to change its level before handover!!!!!!!
Join Date: Oct 1998
Location: Manchester, UK
Posts: 35
Likes: 0
Received 0 Likes
on
0 Posts
Conflicting Instructions
Having waded through this entire thread I would add the following observations:-
1) I have flown extensively in and out of Swiss airspace from a multitude of directions and have never suffered any handover problems (other than an initial incorrect freq from the Italians maybe). The swiss have always acted with a high standard of professionalism. Lets give them (as with all the other parties) the benefit of the doubt until the facts appear.
2) I have yet to see evidence, from any official source, that the TU154 was indeed TCAS equipped although I believe that it is a requirement to fly in European airspace. Having said that I have flown an aircraft over the same point within the past two months with a large ADD sticker next to the transponder relating to the unserviceability of the TCAS system. The MEL provides 7 days for it to be rectified.
3) It would seem that this accident may be attributable to, for whatever reason, the instructions to the TU154 to descend coming too late so that, by the time they were translated into action in the cockpit of the TU154, those instructions were in conflict with the instructions from the TCAS. This leads a pilot into a situation where seconds are vital but indecision or even an incorrect response are possible (especially in a quiet sky late at night).
I would argue medium range collision avoidance by controllers can and should involve level changes (ie would have been applicable on first contact with both these aircraft). Once normal seperation has been erroded, and given the likely onset of conflicting Resolution Advisory messages from the TCAS units, that for a controller to offer anything other than a lateral conflict resolution call would be extremely dangerous for the reasons which may have become all too evident in this case.
Simple let TCAS take care of the vertical - limit the controllers inputs to the horizontal - no conflicting instructions can be received.
Sympathy to all involved.
1) I have flown extensively in and out of Swiss airspace from a multitude of directions and have never suffered any handover problems (other than an initial incorrect freq from the Italians maybe). The swiss have always acted with a high standard of professionalism. Lets give them (as with all the other parties) the benefit of the doubt until the facts appear.
2) I have yet to see evidence, from any official source, that the TU154 was indeed TCAS equipped although I believe that it is a requirement to fly in European airspace. Having said that I have flown an aircraft over the same point within the past two months with a large ADD sticker next to the transponder relating to the unserviceability of the TCAS system. The MEL provides 7 days for it to be rectified.
3) It would seem that this accident may be attributable to, for whatever reason, the instructions to the TU154 to descend coming too late so that, by the time they were translated into action in the cockpit of the TU154, those instructions were in conflict with the instructions from the TCAS. This leads a pilot into a situation where seconds are vital but indecision or even an incorrect response are possible (especially in a quiet sky late at night).
I would argue medium range collision avoidance by controllers can and should involve level changes (ie would have been applicable on first contact with both these aircraft). Once normal seperation has been erroded, and given the likely onset of conflicting Resolution Advisory messages from the TCAS units, that for a controller to offer anything other than a lateral conflict resolution call would be extremely dangerous for the reasons which may have become all too evident in this case.
Simple let TCAS take care of the vertical - limit the controllers inputs to the horizontal - no conflicting instructions can be received.
Sympathy to all involved.
Join Date: Oct 2001
Location: uk
Posts: 5
Likes: 0
Received 0 Likes
on
0 Posts
TCAS RA when at max alt
i accept that these a/c were not flying at their max alt, but I was wondering if a 'TCAS expert' could answer this question. If 2 a/c at say 42000 on an intercept course with resulting TA followed by RA would one a/c be given a command to climb? and would this be wise to accept if it took the a/c above its max operating altitude and close to 'coffin corner'.
Or does the system take into account a/c performance when commanding an RA? If flying at high altitudes would TCAS command one a/c to descend and another to maintain its level?
Or does the system take into account a/c performance when commanding an RA? If flying at high altitudes would TCAS command one a/c to descend and another to maintain its level?
Join Date: Mar 2002
Location: India
Posts: 346
Likes: 0
Received 0 Likes
on
0 Posts
Oldjet Jockey:
Your point fully taken. I am not too familiar with (Swiss) civilian ATC (only military ATC), but I would doubt whether you could refuse a handover once the a/c has physically entered your sector.
Your point fully taken. I am not too familiar with (Swiss) civilian ATC (only military ATC), but I would doubt whether you could refuse a handover once the a/c has physically entered your sector.
Last edited by Alpha Leader; 3rd Jul 2002 at 11:50.
Join Date: Mar 2002
Location: India
Posts: 346
Likes: 0
Received 0 Likes
on
0 Posts
Breach of duty at Skyguide conceded
Immediately prior to the mid-air collision over Lake Constance there was – contrary to earlier reports – a breach of duty at Skyguide, the Swiss ATC company. One of the two duty controllers had left his post contrary to standing orders.
The second controller of Swiss ATC company Skyguide who had taken a break at the time of the mid-air collision over Lake Constance had left his workplace at Zurich Airport control tower without authorization. This account of events was conceded by Skyguide manager Anton Maag during an interview on Wednesday morning.
Internal regulations stipulate that during maintenance work on the in-house conflict warning system installed at Skyguide, there must be two controllers on duty at all times. “The violation of this directive will have internal consequences for our entire operations” said Maag. “We must ensure that current regulations are followed more accurately” he added.
In this interview, he did not exclude that the absence of the second controller might have contributed to the collision of the two aircraft. This, however, would have to be a matter for the ongoing investigation, he added.
The above based on AP at 1237 MEZ
The second controller of Swiss ATC company Skyguide who had taken a break at the time of the mid-air collision over Lake Constance had left his workplace at Zurich Airport control tower without authorization. This account of events was conceded by Skyguide manager Anton Maag during an interview on Wednesday morning.
Internal regulations stipulate that during maintenance work on the in-house conflict warning system installed at Skyguide, there must be two controllers on duty at all times. “The violation of this directive will have internal consequences for our entire operations” said Maag. “We must ensure that current regulations are followed more accurately” he added.
In this interview, he did not exclude that the absence of the second controller might have contributed to the collision of the two aircraft. This, however, would have to be a matter for the ongoing investigation, he added.
The above based on AP at 1237 MEZ
Join Date: Mar 2002
Location: LONDON
Posts: 314
Likes: 0
Received 0 Likes
on
0 Posts
Even if the STCA was turned off, isn't that only a last safety net?
I really feel for the Russians. Sure, they could probably have prevented the collision and the STCA would probably have done too, but how was the confliction set up in the first place?
It is a REGULAR occurance for traffic to miss r/t calls, sometimes it takes 3 or 4 attempts to get a response. At Heathrow yesterday a shuttle was out of contact for several minutes; they were about to scramble the Tornados. A few years ago traffic off BNN and LAM on conflicting tracks very nearly took each other out because BOTH were off the r/t for several minutes. If we are expecting things not to go wrong then we are being naive in the extreme and if we are building systems on that basis so that we can move more traffic then we are in for more of this.
I feel sick.
Point 4
I really feel for the Russians. Sure, they could probably have prevented the collision and the STCA would probably have done too, but how was the confliction set up in the first place?
It is a REGULAR occurance for traffic to miss r/t calls, sometimes it takes 3 or 4 attempts to get a response. At Heathrow yesterday a shuttle was out of contact for several minutes; they were about to scramble the Tornados. A few years ago traffic off BNN and LAM on conflicting tracks very nearly took each other out because BOTH were off the r/t for several minutes. If we are expecting things not to go wrong then we are being naive in the extreme and if we are building systems on that basis so that we can move more traffic then we are in for more of this.
I feel sick.
Point 4
Join Date: Jul 2001
Location: Here and There
Posts: 167
Likes: 0
Received 0 Likes
on
0 Posts
Boulderman,
your eveluation is correct. The TCAS in deed does include the performance-factors of the aircraft, and in 42000 feet (assuming that is close to the max. altitude under given conditions -A/C weight, Temp., etc.), the TCAS would not issue a climb-command on RA, but as you correcly stated, to maintain the present altitude, while the other A/C is commanded to descend.
Of course, a Learjet operating at 42.000 feet is likely to receive a climb command on RA, while a 747 will certainly be commanded to either maintain altitude or to descent by the TCAS.
Cheers
your eveluation is correct. The TCAS in deed does include the performance-factors of the aircraft, and in 42000 feet (assuming that is close to the max. altitude under given conditions -A/C weight, Temp., etc.), the TCAS would not issue a climb-command on RA, but as you correcly stated, to maintain the present altitude, while the other A/C is commanded to descend.
Of course, a Learjet operating at 42.000 feet is likely to receive a climb command on RA, while a 747 will certainly be commanded to either maintain altitude or to descent by the TCAS.
Cheers
Join Date: Jun 2000
Location: uk
Posts: 4
Likes: 0
Received 0 Likes
on
0 Posts
Just an observation but on the front cover of the flight safety publication issue 47 the aircraft in the back ground appears to be Reg. 00 DL-K.
I would only add that it is a very sad event and shall await the formal report to make any conclusions
I would only add that it is a very sad event and shall await the formal report to make any conclusions
Join Date: Mar 2002
Location: India
Posts: 346
Likes: 0
Received 0 Likes
on
0 Posts
German aviation expert takes swipe at Skyguide
From an interview with Volker Thomalla, Editor-in-Chief of "Flug Revue", published today via Süddeutsche Zeitung/AFP/dpa:
Swiss AT controllers did not work "to their optimum capabilites" in the minutes immediately leading up to the fatal mid-air collision, according to Volker Thomalla, Editor-in-Chief of "Flug Revue". Their instructions to the TU-154 crew had been issued at very short notice, he said. There was no apparent reason for the controller to issue his warning a mere 50 seconds prior to the projected collision point, but he should have done so considerably earlier. On the other hand, AT controllers should normally be able to rely on cockpit crews reacting promptly.
He also added that Swiss ATC "did not have a very good year in 2001", with some 55 near misses recorded. This had to be seen against a background of very heavy traffic on intersecting air routes, as North-South and East-West routes crossed over their air space. A unified European air space as well as a single standard would contribute towards greater safety, he added.
Swiss AT controllers did not work "to their optimum capabilites" in the minutes immediately leading up to the fatal mid-air collision, according to Volker Thomalla, Editor-in-Chief of "Flug Revue". Their instructions to the TU-154 crew had been issued at very short notice, he said. There was no apparent reason for the controller to issue his warning a mere 50 seconds prior to the projected collision point, but he should have done so considerably earlier. On the other hand, AT controllers should normally be able to rely on cockpit crews reacting promptly.
He also added that Swiss ATC "did not have a very good year in 2001", with some 55 near misses recorded. This had to be seen against a background of very heavy traffic on intersecting air routes, as North-South and East-West routes crossed over their air space. A unified European air space as well as a single standard would contribute towards greater safety, he added.
Join Date: Jul 2002
Location: UK
Posts: 4
Likes: 0
Received 0 Likes
on
0 Posts
It seems as though the early warning systems had been switched off by Swiss ATC to test the software, shortly before the crash:
(From: ananova.com, bits snipped )
An automatic warning system alerting controllers that two planes were on a collision course was switched off when they crashed above Germany.
A Swiss air traffic control spokesman said the warning system was switched off for maintenance, and it was "a purely theoretical question" whether the system alone could have prevented the disaster.
"Many signs point to an exceptionally unlucky combination of circumstances," Patrick Herr said.
The spokesman said the warning system was undergoing software checks. The routine maintenance is carried out at quiet air traffic times, he said.
Story filed: 11:24 Wednesday 3rd July 2002
http://www.ananova.com/news/story/sm_621141.html
(From: ananova.com, bits snipped )
An automatic warning system alerting controllers that two planes were on a collision course was switched off when they crashed above Germany.
A Swiss air traffic control spokesman said the warning system was switched off for maintenance, and it was "a purely theoretical question" whether the system alone could have prevented the disaster.
"Many signs point to an exceptionally unlucky combination of circumstances," Patrick Herr said.
The spokesman said the warning system was undergoing software checks. The routine maintenance is carried out at quiet air traffic times, he said.
Story filed: 11:24 Wednesday 3rd July 2002
http://www.ananova.com/news/story/sm_621141.html
Join Date: Sep 2000
Location: Dubai
Posts: 74
Likes: 0
Received 0 Likes
on
0 Posts
Even as a controller, I have to say that it saddens me to see that there are still those who are referring to the scenario as though the T154 pilot could have done much about it.
To be honest, I see both sets of pilot's as though they were rabbits stuck in the headlights of an oncoming truck.
The ATC system appears to have put the aircraft in a very dire situation, and now claims that it is their fault because they ducked instead of diving.
Quite honestly, unless there is some (as yet un-forthcoming) reason why the ATC's attention was not on his Radar Display, providing simple Standard Separation (timeously), I can see no excuse for ATC.
It may well be that over-reliance on MTCD has given the ATC staff a false sense of security.
It is also known that (in ATC) most incidents occur when the traffic levels are low and the ATC is not paying 100% attention to the task at hand.
To be honest, I see both sets of pilot's as though they were rabbits stuck in the headlights of an oncoming truck.
The ATC system appears to have put the aircraft in a very dire situation, and now claims that it is their fault because they ducked instead of diving.
Quite honestly, unless there is some (as yet un-forthcoming) reason why the ATC's attention was not on his Radar Display, providing simple Standard Separation (timeously), I can see no excuse for ATC.
It may well be that over-reliance on MTCD has given the ATC staff a false sense of security.
It is also known that (in ATC) most incidents occur when the traffic levels are low and the ATC is not paying 100% attention to the task at hand.
Join Date: Feb 2000
Location: Ayrshire, Scotland
Posts: 474
Likes: 0
Received 0 Likes
on
0 Posts
With 150% hindsight and making no judgements whatsoever for all professionals involved going about their duty ….
So far the situation looks like –
Geneva ATC had their ATC conflict alert off for maintenance.
The sector should have been manned by 2 ATCOs and only one was there. The second ATCO had actually left the building.
Geneva accepted the Russian aircraft at FL360 on a track which directly conflicted with the DHL in the cruise at FL360 and under their control. Normal separation for crossing tracks at the same level would be at least 10 minutes. (If aircraft were on direct routes the advance warning of confliction on Flight Progress strips may have been less obvious or non existent)
Swiss ATC said initially they instructed the Russian to descend 2 minutes before collision. That has now been revised to 50 seconds. ( A time now probably measured to the second from RT tapes)
So far it is not clear what phraseology was used for the avoiding action descent.
It appears no avoiding action was given to the DHL when a hard right turn would probably have been an appropriate option.
The Russian crew did not acknowledge the first call but did acknowledge the second and was descending.
At the same time, if his TCAS was serviceable, it would have been annunciating the RA “climb, climb” as a consequence of the “descend, descend” RA to the DHL.
It is likely then that the Russian was receiving conflicting resolution from ATC and TCAS.
Spoken to in his non native tongue, with less than 50 seconds notice he assimilated two avoiding action instructions, disengaged the autopilot and took the ATC avoiding action and managed a descent of about 500ft.
The simultaneous and similar TCAS and ATC avoiding action kept the aircraft in conflict until collision.
One big lesson pilots may take from this scenario even if not 100% accurate - never, never, never choose not to follow a TCAS RA no matter who tells you or what you think is correct.
Hopefully the industry will not allocate blame but talk about cause.
So far the situation looks like –
Geneva ATC had their ATC conflict alert off for maintenance.
The sector should have been manned by 2 ATCOs and only one was there. The second ATCO had actually left the building.
Geneva accepted the Russian aircraft at FL360 on a track which directly conflicted with the DHL in the cruise at FL360 and under their control. Normal separation for crossing tracks at the same level would be at least 10 minutes. (If aircraft were on direct routes the advance warning of confliction on Flight Progress strips may have been less obvious or non existent)
Swiss ATC said initially they instructed the Russian to descend 2 minutes before collision. That has now been revised to 50 seconds. ( A time now probably measured to the second from RT tapes)
So far it is not clear what phraseology was used for the avoiding action descent.
It appears no avoiding action was given to the DHL when a hard right turn would probably have been an appropriate option.
The Russian crew did not acknowledge the first call but did acknowledge the second and was descending.
At the same time, if his TCAS was serviceable, it would have been annunciating the RA “climb, climb” as a consequence of the “descend, descend” RA to the DHL.
It is likely then that the Russian was receiving conflicting resolution from ATC and TCAS.
Spoken to in his non native tongue, with less than 50 seconds notice he assimilated two avoiding action instructions, disengaged the autopilot and took the ATC avoiding action and managed a descent of about 500ft.
The simultaneous and similar TCAS and ATC avoiding action kept the aircraft in conflict until collision.
One big lesson pilots may take from this scenario even if not 100% accurate - never, never, never choose not to follow a TCAS RA no matter who tells you or what you think is correct.
Hopefully the industry will not allocate blame but talk about cause.
Join Date: Sep 2000
Location: Luxembourg
Posts: 34
Likes: 0
Received 0 Likes
on
0 Posts
Alpha Leader
I understand your point but the sequence for hand over is that first there is a co-ordination between the two ATC units in which the handing over control unit offers the traffic to the receiving unit under certain conditions the main one of which is the flight level of the aircraft concerned. If this is accepted by the receiving controller the hand over takes place. If it is not acceptable due to other traffic already in the receiving controller's airspace he would refuse the offered co-ordinartion and request a different level or a delay perhaps rarely with one or two orbits at the same level to ensure separation and to resolve the confliction. This should all take place BEFORE handover and not afterwards.
Suplementary questions would be: was the aircraft at the c0-0rdinated level? or was there a correct co-ordination with a level bust?
Only time will tell but these are questions that must be asked
I understand your point but the sequence for hand over is that first there is a co-ordination between the two ATC units in which the handing over control unit offers the traffic to the receiving unit under certain conditions the main one of which is the flight level of the aircraft concerned. If this is accepted by the receiving controller the hand over takes place. If it is not acceptable due to other traffic already in the receiving controller's airspace he would refuse the offered co-ordinartion and request a different level or a delay perhaps rarely with one or two orbits at the same level to ensure separation and to resolve the confliction. This should all take place BEFORE handover and not afterwards.
Suplementary questions would be: was the aircraft at the c0-0rdinated level? or was there a correct co-ordination with a level bust?
Only time will tell but these are questions that must be asked
Join Date: Mar 2002
Location: India
Posts: 346
Likes: 0
Received 0 Likes
on
0 Posts
Findo:
It's ok to say "Hopefully the industry will not allocate blame but talk about cause".
However, given the admission of "breach of duty" by Skyguide, it will be hard to avoid the allocation of responsibility if not blame.
The point is that accidents are rarely caused by "things happening" but by human action.
It's ok to say "Hopefully the industry will not allocate blame but talk about cause".
However, given the admission of "breach of duty" by Skyguide, it will be hard to avoid the allocation of responsibility if not blame.
The point is that accidents are rarely caused by "things happening" but by human action.
Mmmmm PPruuune!
Join Date: Jul 1998
Location: UK
Posts: 597
Likes: 0
Received 0 Likes
on
0 Posts
Firstly my condolences to all concerned.
[/QUOTE] Cause 5. Bad luck. Even though both a/c were descending at the time of the collision, an accident was by no means inevitable as has been suggested on the TV news tonight. The chances of a collision were still small, especially since visibility was good. Instead of the catastrophe that happened, we would probably not find out about the incident for some time and it would be 'just another airprox incident'. [/QUOTE]
In aviation there is no such thing as "Bad Luck".
It is perhaps ironic that the swiss cheese simile is used in our Flight Safety forums but sadly it is no less pertinent when the holes line up as they have here. There is obviously a chain of events which could have been broken at any stage from Flight Planning to the final manoeuvres of both aircraft. Many valid points have come out of this forum already without the speculation of what the respective crews may or may not have reacted to. Those professional amongst us will recognise these points without me re-iterating them again. But for my tuppence worth:
TCAS is not infallible but assuming it was working correctly, it has a perceptable lag time in resolving a DESCEND RA to a CLIMB, add the pilots reaction time and the time to change a 90+ ton aircraft's flight path and it all adds up to getting too close for comfort. Yes, believe and react to TCAS but it should be a last line of defence. I find it inconceivable that there was no lateral separation instruction issued by ATC or requested by either aircaft. Personally, if I get TCAS traffic at the same level on an apparent intercept I would query it well before it got to a TA or RA. I would also be aquiring it visually(assuming VMC).
The case mentioned by Pontious? was BA & Pakistan 747s over the Indian ocean, I think, both at their correct opposite FLs except the PAL 747 TCAS had a bent pin resulting in an incorrect alt input to the TCAS telling it that both a/c were co-alt. TCAS did its thing and instructed PAL to climb (up to BAs level) and BA to descend (down to PALs level) Collision avoided as PALs TCAS switched Alt bands as it climbed and thus fed correct inputs plus to the A/C were visual. The kit may be good but it should be used in conjunction with the "Big Picture".
I'm sorry, but I do not believe these two aircraft should have been placed in that situation on a quiet night in the first place. Having been done so consequent actions are open to debate but it does behove all of us to place ourselves in either cockpit and ask "what would I do?"
God Bless all
[/QUOTE] Cause 5. Bad luck. Even though both a/c were descending at the time of the collision, an accident was by no means inevitable as has been suggested on the TV news tonight. The chances of a collision were still small, especially since visibility was good. Instead of the catastrophe that happened, we would probably not find out about the incident for some time and it would be 'just another airprox incident'. [/QUOTE]
In aviation there is no such thing as "Bad Luck".
It is perhaps ironic that the swiss cheese simile is used in our Flight Safety forums but sadly it is no less pertinent when the holes line up as they have here. There is obviously a chain of events which could have been broken at any stage from Flight Planning to the final manoeuvres of both aircraft. Many valid points have come out of this forum already without the speculation of what the respective crews may or may not have reacted to. Those professional amongst us will recognise these points without me re-iterating them again. But for my tuppence worth:
TCAS is not infallible but assuming it was working correctly, it has a perceptable lag time in resolving a DESCEND RA to a CLIMB, add the pilots reaction time and the time to change a 90+ ton aircraft's flight path and it all adds up to getting too close for comfort. Yes, believe and react to TCAS but it should be a last line of defence. I find it inconceivable that there was no lateral separation instruction issued by ATC or requested by either aircaft. Personally, if I get TCAS traffic at the same level on an apparent intercept I would query it well before it got to a TA or RA. I would also be aquiring it visually(assuming VMC).
The case mentioned by Pontious? was BA & Pakistan 747s over the Indian ocean, I think, both at their correct opposite FLs except the PAL 747 TCAS had a bent pin resulting in an incorrect alt input to the TCAS telling it that both a/c were co-alt. TCAS did its thing and instructed PAL to climb (up to BAs level) and BA to descend (down to PALs level) Collision avoided as PALs TCAS switched Alt bands as it climbed and thus fed correct inputs plus to the A/C were visual. The kit may be good but it should be used in conjunction with the "Big Picture".
I'm sorry, but I do not believe these two aircraft should have been placed in that situation on a quiet night in the first place. Having been done so consequent actions are open to debate but it does behove all of us to place ourselves in either cockpit and ask "what would I do?"
God Bless all
Join Date: Mar 2002
Location: India
Posts: 346
Likes: 0
Received 0 Likes
on
0 Posts
Oldjet Jockey:
Again, I understand more about (Swiss) military ATC than civilian ATC. However, SOP is to have two controllers handle the hand-over, one on screen and the other via telephone.
Your comments thus open up a can of worms, because we now know that one of the two Swiss controllers was AWOL at the time of the collision (and possibly for some time prior to that)
Again, I understand more about (Swiss) military ATC than civilian ATC. However, SOP is to have two controllers handle the hand-over, one on screen and the other via telephone.
Your comments thus open up a can of worms, because we now know that one of the two Swiss controllers was AWOL at the time of the collision (and possibly for some time prior to that)