Mid-Air Collision over Southern Germany (merged)
My deep sympathies to the controller in Karlsruhe -- he saw it coming and was helpless to stop it because of the telephones and could no longer reach the Tu154 after it had changed frequency.
TCAS has acquited itself very well and as many have pointed out, by the time an RA has been issued, ATC must be ignored until the RA has been cleared.
There may be a cultural factor here as it seems Russians receive strong conditioning to follow authoritative directives no matter what -- here we see urgent ATC orders trumping a valid TCAS RA.
Perhaps Ruslan could comment.
Question: were the accident a/c on the same ATC frequency?
TCAS has acquited itself very well and as many have pointed out, by the time an RA has been issued, ATC must be ignored until the RA has been cleared.
There may be a cultural factor here as it seems Russians receive strong conditioning to follow authoritative directives no matter what -- here we see urgent ATC orders trumping a valid TCAS RA.
Perhaps Ruslan could comment.
Question: were the accident a/c on the same ATC frequency?
"The INTRODUCER"
About VDL-4. There are epic amounts of information on this subject available: try http://www.nup.nu/ and www.gpc.se and www.avtech.se
Referring to it all as VDL-4 is misleading however. VDL-4 is just the datalink (the ICAO designation for what is otherwise called STDMA). There are other datalinks (Mode S or 1090, and UAT) which would arguably do just as well. That's quite an argument by the way - see the press cuttings on the above sites.
What you should call it is ADS-B (automatic dependent surveillance-broadcast), and the VDL-4-based version - as I think anyone who has been on the SAS flight demos will agree - is truly remarkable. It would make runway incursion-generated accidents a thing of the past (definitely would have saved the Linate and CDG runway collisions) and ditto mid-airs (but then TCAS should have saved this one of course.)
The US has just said that it is not intending to use VDL-4 as its ADS-B datalink - it's common knowledge that there are huge industrial base concerns playing a role in that decision. VDL-4/STDMA is essentially a Swedish concept, although, contrary to self-interested assertions from others, there are no patent issues restricting its use since its inventor, the admirable Hakan Lans, opened it up to everyone. It's also true that there are perfectly respectable technical arguments on both sides too (rather stronger on the VDL-4 side IMHO.)
Unfortunately, the industrial base issues are one of the reasons that ADS-B has been so long coming. But the real problem is to do with devising and implementing the operational procedures needed to make it work. Work it must be made to do, however - there is really no plan B, either in terms of safety or capacity.
Referring to it all as VDL-4 is misleading however. VDL-4 is just the datalink (the ICAO designation for what is otherwise called STDMA). There are other datalinks (Mode S or 1090, and UAT) which would arguably do just as well. That's quite an argument by the way - see the press cuttings on the above sites.
What you should call it is ADS-B (automatic dependent surveillance-broadcast), and the VDL-4-based version - as I think anyone who has been on the SAS flight demos will agree - is truly remarkable. It would make runway incursion-generated accidents a thing of the past (definitely would have saved the Linate and CDG runway collisions) and ditto mid-airs (but then TCAS should have saved this one of course.)
The US has just said that it is not intending to use VDL-4 as its ADS-B datalink - it's common knowledge that there are huge industrial base concerns playing a role in that decision. VDL-4/STDMA is essentially a Swedish concept, although, contrary to self-interested assertions from others, there are no patent issues restricting its use since its inventor, the admirable Hakan Lans, opened it up to everyone. It's also true that there are perfectly respectable technical arguments on both sides too (rather stronger on the VDL-4 side IMHO.)
Unfortunately, the industrial base issues are one of the reasons that ADS-B has been so long coming. But the real problem is to do with devising and implementing the operational procedures needed to make it work. Work it must be made to do, however - there is really no plan B, either in terms of safety or capacity.
Join Date: Sep 2000
Location: Dubai
Posts: 74
Likes: 0
Received 0 Likes
on
0 Posts
RatherbeFlying:
I understand from previous BFU statements that they were on the same frequency.
Can you imagine being in the cruise, and then getting TA, RA and ATC(WITH URGENT COMMAND) in quick succession ? I don't know if simulators can generate the level of confusion/doubt that must have existed for the russian crew.
How would we (you/I) have reacted ? Very Sad indeed.
I understand from previous BFU statements that they were on the same frequency.
Can you imagine being in the cruise, and then getting TA, RA and ATC(WITH URGENT COMMAND) in quick succession ? I don't know if simulators can generate the level of confusion/doubt that must have existed for the russian crew.
How would we (you/I) have reacted ? Very Sad indeed.
Join Date: Feb 2000
Location: Bit nosey aren't you
Posts: 455
Likes: 0
Received 0 Likes
on
0 Posts
Low Pass,
As Koncordski says, it shouldn't have mattered. From various books:
Primary responsibility to avoid rests with the crews and ATC. Question ATC as early as possible about traffic which is a concern and coordinate.
If you have no SA on the traffic and get a TA and still can't see it then you are not well placed.
By the time TCAS is issueing an RA the time is for actions not words. ATC have had their go at providing collision avoidance and sadly missed the boat. On the ground you do not now have a sufficient update rate to provide sound real time guidance in the gravy strokes.
Now the automatic safety systems come into play which if followed sort the problem out. Quoting from the QRH for a bus:
Respond Promptly and smoothly to an RA
AP off
FD off
Adjust vertical speed to the green sector
Respect stall, GPWS or windshear warnings
Attempt to see the traffic
Notify ATC
What the book then says is:
Once an aircraft in compliance with an RA departs from its assigned ATC clearance the controoler ceases to be responsible for separation until 'Clear of Conflict' is called and the controller reclears the aircraft.
I await the report with interest but all of the data that is appearing seems to show that if the crews had followed the coordinated avoidance maneuvres as called by TCAS we would be talking about an Airmiss.
Ghost
As Koncordski says, it shouldn't have mattered. From various books:
Primary responsibility to avoid rests with the crews and ATC. Question ATC as early as possible about traffic which is a concern and coordinate.
If you have no SA on the traffic and get a TA and still can't see it then you are not well placed.
By the time TCAS is issueing an RA the time is for actions not words. ATC have had their go at providing collision avoidance and sadly missed the boat. On the ground you do not now have a sufficient update rate to provide sound real time guidance in the gravy strokes.
Now the automatic safety systems come into play which if followed sort the problem out. Quoting from the QRH for a bus:
Respond Promptly and smoothly to an RA
AP off
FD off
Adjust vertical speed to the green sector
Respect stall, GPWS or windshear warnings
Attempt to see the traffic
Notify ATC
What the book then says is:
Once an aircraft in compliance with an RA departs from its assigned ATC clearance the controoler ceases to be responsible for separation until 'Clear of Conflict' is called and the controller reclears the aircraft.
I await the report with interest but all of the data that is appearing seems to show that if the crews had followed the coordinated avoidance maneuvres as called by TCAS we would be talking about an Airmiss.
Ghost
Join Date: Nov 1999
Location: London
Posts: 65
Likes: 0
Received 0 Likes
on
0 Posts
ferris,
Re your TCAS incident :
Congratulations for staying cool and getting your point across despite the readiness of others to shoot you down in flames.
I believe there was similar incident in UK airspace last week. Two Aircraft cleared to levels 1000' apart, entirely safe, yet TCAS decides the descending A/C should continue through the others level. The A/C didn't hit so it could be argued that TCAS did it's job, but surely the standard separation provided by ATC was infinitely safer ?
For every airprox TCAS causes there must be hundreds it stops, but we must all be aware of it's limitations and be willing to learn from others experiences as well as our own
Re your TCAS incident :
Congratulations for staying cool and getting your point across despite the readiness of others to shoot you down in flames.
I believe there was similar incident in UK airspace last week. Two Aircraft cleared to levels 1000' apart, entirely safe, yet TCAS decides the descending A/C should continue through the others level. The A/C didn't hit so it could be argued that TCAS did it's job, but surely the standard separation provided by ATC was infinitely safer ?
For every airprox TCAS causes there must be hundreds it stops, but we must all be aware of it's limitations and be willing to learn from others experiences as well as our own
Join Date: Jul 2002
Location: this side of the hill
Posts: 127
Likes: 0
Received 0 Likes
on
0 Posts
Ghost,
Do you still expect to receive an avoiding action call from the ATCO once you will get too close to an other a/c or do you simply expect traffic information? In your logic it seems that this is sufficient since the TCAS of both planes will take over anyway if things turn sour. Let's take the following example. Two a/c on parallel headings with 5 miles in between in IMC. One of the two is turned away from the other but actually turns towards it (really happened, the difference between left and right is not always well known). The only way to achieve separation again is to change levels. Do you expect us to climb or descend one or both, which can be contradicting TCAS or do you want us to tell the a/c that they are closing in rapidly and that they should just sort it out with their TCAS?
Garp
Do you still expect to receive an avoiding action call from the ATCO once you will get too close to an other a/c or do you simply expect traffic information? In your logic it seems that this is sufficient since the TCAS of both planes will take over anyway if things turn sour. Let's take the following example. Two a/c on parallel headings with 5 miles in between in IMC. One of the two is turned away from the other but actually turns towards it (really happened, the difference between left and right is not always well known). The only way to achieve separation again is to change levels. Do you expect us to climb or descend one or both, which can be contradicting TCAS or do you want us to tell the a/c that they are closing in rapidly and that they should just sort it out with their TCAS?
Garp
Join Date: Oct 2001
Location: London
Posts: 183
Likes: 0
Received 0 Likes
on
0 Posts
Vertigo, of the airmiss over Wales last week you said:
''The A/C didn't hit so it could be argued that TCAS did it's job''
Well I understand from colleagues that the E145 passed through the level of the Dash8 (FL180) just 1 and half miles in front travelling in the OPPOSITE DIRECTION so it was entirely luck- not TCAS- that prevented this incident. Without TCAS it wouldnt have happened as the aircraft were cleared to safe levels.
''The A/C didn't hit so it could be argued that TCAS did it's job''
Well I understand from colleagues that the E145 passed through the level of the Dash8 (FL180) just 1 and half miles in front travelling in the OPPOSITE DIRECTION so it was entirely luck- not TCAS- that prevented this incident. Without TCAS it wouldnt have happened as the aircraft were cleared to safe levels.
Join Date: Feb 2000
Location: Bit nosey aren't you
Posts: 455
Likes: 0
Received 0 Likes
on
0 Posts
Vertigo,
Couldn't agree more, the same can be said of the historic picture from a search radar. It has its limitations and when the blips have merged the controller has a tough choice on what information to try and pass.
Interestingly though he will be working on exactly the same info as TCAS namely the Mode 'C' return but with the difference that his will be a sweep out of date whereas TCAS will be working in near real-time. Ferris' problem was caused by a pilot that didn't understand the limitations of TCAS, maneuvring his aircraft in-appropriately.
Ultimately if I were in the situation where I hadn't seen the aircraft approaching on TCAS and then couldn't see it out the window when an RA was activated. I am certain I would follow it. Just as if I was low on SA and a GPWS warning went off I would follow that.
Garp,
I would expect you to notice that the aircraft had turned the wrong way and tell him. If you couldn't get through give avoiding action to both aircraft to avoid TCAS even becoming an issue.
My general point is that TCAS gives a hint at 40 secs and action at about 25 secs. ATC's have the big picture from a long way away but unfortunately do not have the resolution except via Mode 'C' to separate aircraft at close range and that is exactly what TCAS is designed to do.
If you can hand on heart provide the best information to the crews, info that they do not already, then you are contributing SA and must chip in.
I know it is hindsight and when the controller first transmitted in this incident he must have been horrified. That said I would hope ATC SOPs at this late stage would be massively directive to both aircraft, DHL climb, Ruskie descend.
If transmitted just before the coordinated RAs it still might have not helped (life isn't fair) but at 44 secs to go, if just one aircraft had moved in the next 19 seconds it would hopefully have prevented any TCAS calls but at worst meant they didn't conflict with ATC's instructions when they came.
Ghost
Couldn't agree more, the same can be said of the historic picture from a search radar. It has its limitations and when the blips have merged the controller has a tough choice on what information to try and pass.
Interestingly though he will be working on exactly the same info as TCAS namely the Mode 'C' return but with the difference that his will be a sweep out of date whereas TCAS will be working in near real-time. Ferris' problem was caused by a pilot that didn't understand the limitations of TCAS, maneuvring his aircraft in-appropriately.
Ultimately if I were in the situation where I hadn't seen the aircraft approaching on TCAS and then couldn't see it out the window when an RA was activated. I am certain I would follow it. Just as if I was low on SA and a GPWS warning went off I would follow that.
Garp,
I would expect you to notice that the aircraft had turned the wrong way and tell him. If you couldn't get through give avoiding action to both aircraft to avoid TCAS even becoming an issue.
My general point is that TCAS gives a hint at 40 secs and action at about 25 secs. ATC's have the big picture from a long way away but unfortunately do not have the resolution except via Mode 'C' to separate aircraft at close range and that is exactly what TCAS is designed to do.
If you can hand on heart provide the best information to the crews, info that they do not already, then you are contributing SA and must chip in.
I know it is hindsight and when the controller first transmitted in this incident he must have been horrified. That said I would hope ATC SOPs at this late stage would be massively directive to both aircraft, DHL climb, Ruskie descend.
If transmitted just before the coordinated RAs it still might have not helped (life isn't fair) but at 44 secs to go, if just one aircraft had moved in the next 19 seconds it would hopefully have prevented any TCAS calls but at worst meant they didn't conflict with ATC's instructions when they came.
Ghost
Join Date: Nov 1999
Location: I wish I knew
Posts: 179
Likes: 0
Received 0 Likes
on
0 Posts
Ghostflyer and Konkordski,
I agree that a RA on TCAS should take priority over a controllers instruction. My point is that in this instance, the Tu did what the controller said which happened to be in conflict with TCAS. If the controller was aware of the TCAS warnigns, the accident probably wouldn't have happened unless the controller intentionally issued a cnflicting instruction. You can argue that it's the Tu pilots fault (and I don't) but that's only half the point. How can we avoid the same thing happening in the future? I understand that technology may not yet support this feature, but it will in the future.
I agree that a RA on TCAS should take priority over a controllers instruction. My point is that in this instance, the Tu did what the controller said which happened to be in conflict with TCAS. If the controller was aware of the TCAS warnigns, the accident probably wouldn't have happened unless the controller intentionally issued a cnflicting instruction. You can argue that it's the Tu pilots fault (and I don't) but that's only half the point. How can we avoid the same thing happening in the future? I understand that technology may not yet support this feature, but it will in the future.
Join Date: Apr 2000
Location: 30 West
Age: 65
Posts: 926
Likes: 0
Received 0 Likes
on
0 Posts
I believe that part of the standard reaction to TCAS is that the PNF transmits TCAS Climb/Descent as appropriate and that should inform ATC that they are no longer the controlling body until the threat has passed.
Join Date: Feb 2000
Location: Bit nosey aren't you
Posts: 455
Likes: 0
Received 0 Likes
on
0 Posts
Low-Pass,
Hopefully TCAS-5z will have that functionality. I think we are all agreeing here. It was a sh*tty chain of events. The Swiss collision avoidance system was down otherwise the conflict would have been spotted way earlier. The phones weren't working properly. Maybe ATC stepped on the RA descent call.
As to ATC getting automatic advice of the RA, yep good idea but playing devils advocate, lets say he was transmitting as that happened, the results could still have been the same. Lets face it once the RA has been issued the controller has a 50/50 chance of being right.
In the future when we are certain of the integrity of the system maybe at 30 secs to go ATC are taken out of the loop and TCAS is left to the job. But how would you know you had 30 secs to go? From the ATC collision avoidance system?...aaahh...but that was down. This accident may well be one that proves that life sucks!
Ghost
Hopefully TCAS-5z will have that functionality. I think we are all agreeing here. It was a sh*tty chain of events. The Swiss collision avoidance system was down otherwise the conflict would have been spotted way earlier. The phones weren't working properly. Maybe ATC stepped on the RA descent call.
As to ATC getting automatic advice of the RA, yep good idea but playing devils advocate, lets say he was transmitting as that happened, the results could still have been the same. Lets face it once the RA has been issued the controller has a 50/50 chance of being right.
In the future when we are certain of the integrity of the system maybe at 30 secs to go ATC are taken out of the loop and TCAS is left to the job. But how would you know you had 30 secs to go? From the ATC collision avoidance system?...aaahh...but that was down. This accident may well be one that proves that life sucks!
Ghost
Last edited by Ghostflyer; 8th Jul 2002 at 16:45.
Keeping Danny in Sandwiches
Join Date: May 1999
Location: UK
Age: 76
Posts: 1,294
Likes: 0
Received 0 Likes
on
0 Posts
.Javelin
All very well, however we all know that you can only transmit when the frequency is clear. VHF radio is a pretty basic method of communication.
It is indeed a tragedy if TCAS actually worked and ATC unknowingly countermanded it.
All very well, however we all know that you can only transmit when the frequency is clear. VHF radio is a pretty basic method of communication.
It is indeed a tragedy if TCAS actually worked and ATC unknowingly countermanded it.
on your FM dial
Join Date: May 2000
Location: Bindook
Posts: 114
Likes: 0
Received 0 Likes
on
0 Posts
Garp,
The words “[callsign], TCAS climb [descent]” should be an air traffic controllers cue that a TCAS resolution advisory manoeuvre is taking place. On hearing that cue the controller should stop issuing vertical instructions.
An air traffic controller should continue to issue vertical and/or lateral instructions to converging traffic right up until they hear those words.
But once you hear those words, you have had your chance, and the time for issuing vertical instructions has passed.
The words “[callsign], TCAS climb [descent] complete, returning to FLxxx” should be an air traffic controllers cue that the aircraft has received a “clear of conflict” aural from TCAS and is now returning to the last cleared level. This is your cue that the TCAS vertical manoeuvring is over and you may once again issue vertical instructions.
TCAS “Crossings” and TCAS Training in General
If anything good is to come from this tragedy, perhaps it might be that hundreds of pilots and more than a few air traffic controllers have gone back to the books in an effort to more thoroughly understand TCAS and its related procedures.
I wish to express my gratitude to vanman who posted a very useful link to the Honeywell TCAS II Users Manual in a TCAS Requirements thread on the Questions forum, and to Capt Pit Bull who has made a number of most informative and well written posts in regards to TCAS.
To both of you, a sincere thank you.
In particular, Capt Pit Bull has very eloquently described the TCAS “crossing RA” in the Nasty Incident thread on the ATC Issues forum.
I must admit to having only had a very vague recollection of the term “crossing” in relation to the operation of TCAS. Prior to reading Capt Pit Bull’s post, I would not have been able to explain the significance of the word “crossing” in a TCAS aural. Obviously, there is no time to attempt to dig the coffee-stained manuals out from under the crew-meals when you are in receipt of a TCAS RA.
I was concerned that I might not have the knowledge I ought to, and I was further concerned that it was only me.
During the last week I conducted a very unscientific survey of five jet captains. Each one was separately asked “What does the ‘crossing’ part of a TCAS aural ‘climb crossing climb’ mean?”
They all said that they had never heard of the TCAS word ‘crossing’, and had no idea what it meant. That disturbed me even more.
There seems to be a real lack of understanding of TCAS by both pilots and air traffic controllers. This is evidenced, in part, by the reckless and/or misguided way that pilots continue to race up (or down) to cleared levels just 1,000 feet apart using unnecessarily high rates of climb (or descent).
From Capt Pit Bull’s posts, it seems that the only way you can generate a “crossing RA” is to use high rates of climb and/or descent close to your cleared level. Yet the apparently well-trained and experienced crews of a major flag carrier’s regional airline have managed to do just that! How can this be?
It seems that whilst regulatory authorities have mandated the installation of TCAS hardware, they have not ensured that crews receive adequate initial and recurrent TCAS training. The requirement to fit TCAS hardware only does half the job, leaving many pilots and air traffic controllers with an apparent distrust of TCAS due to misunderstandings brought about by poor training.
The words “[callsign], TCAS climb [descent]” should be an air traffic controllers cue that a TCAS resolution advisory manoeuvre is taking place. On hearing that cue the controller should stop issuing vertical instructions.
An air traffic controller should continue to issue vertical and/or lateral instructions to converging traffic right up until they hear those words.
But once you hear those words, you have had your chance, and the time for issuing vertical instructions has passed.
The words “[callsign], TCAS climb [descent] complete, returning to FLxxx” should be an air traffic controllers cue that the aircraft has received a “clear of conflict” aural from TCAS and is now returning to the last cleared level. This is your cue that the TCAS vertical manoeuvring is over and you may once again issue vertical instructions.
TCAS “Crossings” and TCAS Training in General
If anything good is to come from this tragedy, perhaps it might be that hundreds of pilots and more than a few air traffic controllers have gone back to the books in an effort to more thoroughly understand TCAS and its related procedures.
I wish to express my gratitude to vanman who posted a very useful link to the Honeywell TCAS II Users Manual in a TCAS Requirements thread on the Questions forum, and to Capt Pit Bull who has made a number of most informative and well written posts in regards to TCAS.
To both of you, a sincere thank you.
In particular, Capt Pit Bull has very eloquently described the TCAS “crossing RA” in the Nasty Incident thread on the ATC Issues forum.
I must admit to having only had a very vague recollection of the term “crossing” in relation to the operation of TCAS. Prior to reading Capt Pit Bull’s post, I would not have been able to explain the significance of the word “crossing” in a TCAS aural. Obviously, there is no time to attempt to dig the coffee-stained manuals out from under the crew-meals when you are in receipt of a TCAS RA.
I was concerned that I might not have the knowledge I ought to, and I was further concerned that it was only me.
During the last week I conducted a very unscientific survey of five jet captains. Each one was separately asked “What does the ‘crossing’ part of a TCAS aural ‘climb crossing climb’ mean?”
They all said that they had never heard of the TCAS word ‘crossing’, and had no idea what it meant. That disturbed me even more.
There seems to be a real lack of understanding of TCAS by both pilots and air traffic controllers. This is evidenced, in part, by the reckless and/or misguided way that pilots continue to race up (or down) to cleared levels just 1,000 feet apart using unnecessarily high rates of climb (or descent).
From Capt Pit Bull’s posts, it seems that the only way you can generate a “crossing RA” is to use high rates of climb and/or descent close to your cleared level. Yet the apparently well-trained and experienced crews of a major flag carrier’s regional airline have managed to do just that! How can this be?
It seems that whilst regulatory authorities have mandated the installation of TCAS hardware, they have not ensured that crews receive adequate initial and recurrent TCAS training. The requirement to fit TCAS hardware only does half the job, leaving many pilots and air traffic controllers with an apparent distrust of TCAS due to misunderstandings brought about by poor training.
Join Date: Aug 1999
Location: England
Posts: 1,050
Likes: 0
Received 0 Likes
on
0 Posts
To answer Cockpit Visits question
TCAS continuously assesses the situation during an RA to see if a modification to the RA is required.
If TCAS sees, that inspite of the RA being flown, the intruder is still predicted to be within the vertical thresholds (i.e. the intruder must have manouevered towards you), then it will issue a revised RA.
This RA will take the form of either a strengthening of the original RA, or a reversal.
Strengthening:
"Climb" (1500 - 2000 fpm) becomes "Increase climb" (2500-3000 fpm).
"Descend" becomes "Increase Descent".
Reversal:
"Climb" becomes "Descend Now".
"Descend" becomes "Climb Now".
These manoeuvres should make it possible for the situation to be ressolved safely.
If you recieve one of these RAs, one of two things has happened:
a) the other aircraft has not got TCAS, and has for some reason unfortunately choosen a very bad time to manoeuvre in the same sense as you.
b) the other aircraft has got TCAS, and the crew have manouevered in the opposire sense to their RA (which, to reiterate, is strictly against training guidelines, and highly dangerous).
Based on what has been reported on the last page or so, it appears that the t154 did get a climb RA, and did manouevre opposite to it.
So, how come TCAS didn't save the day?
There are basically four possible scenarios.
Scenario 1.
We know the 757 left its level fractionally before the t154. We know it would have had a commanded vertical speed of -1500 to -2000 fpm. This would have put both its present altitude, and its predicted flight path, UNDER the t154.
One of TCASs design objectives is to avoid issueing RAs that will make the aircraft cross the intruders altitude. Rationale: Regardless of what happens horizontally, if the aircraft are never at the same level, they cannot hit one another.
Therefore, as TCAS saw the t154 start to descend, it probably would have selected a "strengthening" rather than a "Reversal", and given the b757 an "Increase descent". Simultaneously, the t154 would have been given an "Increase Climb".
Problem is, the t154 appear to have made the decision to follow ATCs avoiding action. It would seem reasonable that they would have initiated a descent well in excess of 1500 fpm and would therefore been matching the b757 as it steepened its descent.
What we are seeing here is the t154 manoevre opposite the sense of its RA not just once (descending instead of climbing), but twice (increasing its descent instead of commencing a climb).
In other words, TCAS would have been defeated by two manoeuvres opposite to given RAs.
Scenario 2.
As before for the initial RA, and opposit manoeuvre by the t154.
On receipt of the strengthened RA, and realising that they had heard the other aircraft say "TCAS Descent", the t154 crew followed their RA and climbed
Unfortunately, the b757 crew, having heard the t154 crew acknowledge their instruction to descend from ATC, decided to climb again. i.e. manoeuvred against the sense of their RA. The resulting levelling off / climbing, by both aircraft, lead to collision.
In other words, TCAS would have been defeated by two manoeuvres opposite to given RAs.
Scenario 3.
As before for initial RA, but this time TCAS choose the reversal scenario. (This might happen if the t154s manoeuvre was particularly sudden, such that it managed to 'out pitch' the b757 before TCAS reassessed i.a.w. scenario 1).
In this case, the b757 would have got a "Climb Now" RA, whilst the t154 would have got a "Descend Now" RA.
The b757, with their thus far proven track record of complying with RAs, followed the new instruction.
The t154 crew, having heard the b757s initial call of 'TCAS descent', and bearing in mind the "climb" RA they had been receiving, decided that they had better climb. Having (finally) made this decision, they were unable to change their minds yet again, and therefore initiated a climb inspite of the fact that during this manoeuvre TCAS told them to descend. i.e. a manouevre in the opposite sense to an RA.
Again, the resulting levelling off / climbing, by both aircraft, lead to collision.
In other words, TCAS would have been defeated by two manoeuvres opposite to given RAs.
Scenario 4
As per scenario 3, but this time the t154 crew respond correctly and comply with their RA.
The b757 crew, for some reason (perhaps the visual illusion I described in Tech Log), decided to increase their descent rate instead of climbing. Again, a manoeuvre opposite the sense of an RA.
Both aircraft increased their descent rates and collided.
In other words, TCAS would have been defeated by two manoeuvres opposite to given RAs.
(All Scenarios:
There may perhaps even have been a third round of RAs, but by this stage range was too close for anything to save the day.)
I trust my non too subtle post is reinforcing the point:
Don't manoeuvre opposite to an RA
Your Ops manual probably gives you the freedom not to follow an RA. Can I reiterate that it is one thing not to follow an RA, it is another thing altogether (and incredibly more dangerous) to manoeuvre Opposite.
However, even if somebody does, TCAS can still save the day as long as somebosy doesn't manoeuvre opposite the next round of RAs.
So, if something happens during an encounter that means you have to manoeuvre opposite an RA, or for whatever reason you suddenly realise that you have accidentally done so, whatever you do, follow the next one.
Historically, there are incidents where TCAS has saved the day at the second RA due to non compliance with the first.
There have also been incidents where 2 RAs have been manoeuvred opposite, and the only thing that saved the day was fortuitous lateral seperation.
CPB
TCAS continuously assesses the situation during an RA to see if a modification to the RA is required.
If TCAS sees, that inspite of the RA being flown, the intruder is still predicted to be within the vertical thresholds (i.e. the intruder must have manouevered towards you), then it will issue a revised RA.
This RA will take the form of either a strengthening of the original RA, or a reversal.
Strengthening:
"Climb" (1500 - 2000 fpm) becomes "Increase climb" (2500-3000 fpm).
"Descend" becomes "Increase Descent".
Reversal:
"Climb" becomes "Descend Now".
"Descend" becomes "Climb Now".
These manoeuvres should make it possible for the situation to be ressolved safely.
If you recieve one of these RAs, one of two things has happened:
a) the other aircraft has not got TCAS, and has for some reason unfortunately choosen a very bad time to manoeuvre in the same sense as you.
b) the other aircraft has got TCAS, and the crew have manouevered in the opposire sense to their RA (which, to reiterate, is strictly against training guidelines, and highly dangerous).
Based on what has been reported on the last page or so, it appears that the t154 did get a climb RA, and did manouevre opposite to it.
So, how come TCAS didn't save the day?
There are basically four possible scenarios.
Scenario 1.
We know the 757 left its level fractionally before the t154. We know it would have had a commanded vertical speed of -1500 to -2000 fpm. This would have put both its present altitude, and its predicted flight path, UNDER the t154.
One of TCASs design objectives is to avoid issueing RAs that will make the aircraft cross the intruders altitude. Rationale: Regardless of what happens horizontally, if the aircraft are never at the same level, they cannot hit one another.
Therefore, as TCAS saw the t154 start to descend, it probably would have selected a "strengthening" rather than a "Reversal", and given the b757 an "Increase descent". Simultaneously, the t154 would have been given an "Increase Climb".
Problem is, the t154 appear to have made the decision to follow ATCs avoiding action. It would seem reasonable that they would have initiated a descent well in excess of 1500 fpm and would therefore been matching the b757 as it steepened its descent.
What we are seeing here is the t154 manoevre opposite the sense of its RA not just once (descending instead of climbing), but twice (increasing its descent instead of commencing a climb).
In other words, TCAS would have been defeated by two manoeuvres opposite to given RAs.
Scenario 2.
As before for the initial RA, and opposit manoeuvre by the t154.
On receipt of the strengthened RA, and realising that they had heard the other aircraft say "TCAS Descent", the t154 crew followed their RA and climbed
Unfortunately, the b757 crew, having heard the t154 crew acknowledge their instruction to descend from ATC, decided to climb again. i.e. manoeuvred against the sense of their RA. The resulting levelling off / climbing, by both aircraft, lead to collision.
In other words, TCAS would have been defeated by two manoeuvres opposite to given RAs.
Scenario 3.
As before for initial RA, but this time TCAS choose the reversal scenario. (This might happen if the t154s manoeuvre was particularly sudden, such that it managed to 'out pitch' the b757 before TCAS reassessed i.a.w. scenario 1).
In this case, the b757 would have got a "Climb Now" RA, whilst the t154 would have got a "Descend Now" RA.
The b757, with their thus far proven track record of complying with RAs, followed the new instruction.
The t154 crew, having heard the b757s initial call of 'TCAS descent', and bearing in mind the "climb" RA they had been receiving, decided that they had better climb. Having (finally) made this decision, they were unable to change their minds yet again, and therefore initiated a climb inspite of the fact that during this manoeuvre TCAS told them to descend. i.e. a manouevre in the opposite sense to an RA.
Again, the resulting levelling off / climbing, by both aircraft, lead to collision.
In other words, TCAS would have been defeated by two manoeuvres opposite to given RAs.
Scenario 4
As per scenario 3, but this time the t154 crew respond correctly and comply with their RA.
The b757 crew, for some reason (perhaps the visual illusion I described in Tech Log), decided to increase their descent rate instead of climbing. Again, a manoeuvre opposite the sense of an RA.
Both aircraft increased their descent rates and collided.
In other words, TCAS would have been defeated by two manoeuvres opposite to given RAs.
(All Scenarios:
There may perhaps even have been a third round of RAs, but by this stage range was too close for anything to save the day.)
I trust my non too subtle post is reinforcing the point:
Don't manoeuvre opposite to an RA
Your Ops manual probably gives you the freedom not to follow an RA. Can I reiterate that it is one thing not to follow an RA, it is another thing altogether (and incredibly more dangerous) to manoeuvre Opposite.
However, even if somebody does, TCAS can still save the day as long as somebosy doesn't manoeuvre opposite the next round of RAs.
So, if something happens during an encounter that means you have to manoeuvre opposite an RA, or for whatever reason you suddenly realise that you have accidentally done so, whatever you do, follow the next one.
Historically, there are incidents where TCAS has saved the day at the second RA due to non compliance with the first.
There have also been incidents where 2 RAs have been manoeuvred opposite, and the only thing that saved the day was fortuitous lateral seperation.
CPB
spekesoftly - the SOP is to avoid nuisance TCAS alerts in RVSM airspace only. Higher climb/descent rates may well be required elsewhere, particularly near (but outside) regulated airspace.
So Beags, in reply to an earlier post that your ageing 4 jet has no TCAS fitted, does that mean that the Tri jet still does not have it fitted, the Comet like thing does not have it and the ageing prop fleet and maybe the newer prop fleet does not have it either means if i understand TCAS correctly means that you will not get the opposite command to me therefore the separation is going to be less in fact half, makes me feel an awful lot safer (not) when I hear the Ascot callsign, maybe its time that the Military were forced to update their fleets, I am here writting this because TCAS worked for me in Greek airsapce, the computor model that the CAA made from our tapes gave very little distance on a head to head. If it had been one of your Tri jets without TCAS I probably would not have been here writing this.
Food for thought at your next Flight Safety meeting eh!!!
Fly safely out there.
Food for thought at your next Flight Safety meeting eh!!!
Fly safely out there.
Join Date: Sep 2000
Location: Dubai
Posts: 74
Likes: 0
Received 0 Likes
on
0 Posts
At the risk of losing the (patent) (copyright) benefits of something that is not out there yet.......
Does anyone know of a simple, PC based, program out there there that can simulate various TCAS situations.
I am not talking about MS Flight Sim or anything else that requires flying and navigating all over the place to get a single RA.
I am talking about training orientated TCAS situations, invloving various models of TCAS and multiple aircraft. Grading and post miss (or hit) commentary or evaluation could be in there too....
Anyone ???
Does anyone know of a simple, PC based, program out there there that can simulate various TCAS situations.
I am not talking about MS Flight Sim or anything else that requires flying and navigating all over the place to get a single RA.
I am talking about training orientated TCAS situations, invloving various models of TCAS and multiple aircraft. Grading and post miss (or hit) commentary or evaluation could be in there too....
Anyone ???
Join Date: Aug 1999
Location: England
Posts: 1,050
Likes: 0
Received 0 Likes
on
0 Posts
BIK_116.80
Thank you for your kind words. However, just to clarify, high vertical rates near your cleared level are not the only way, just a way. Also a common way. Other scenarios possible, but involving ATC errors.
Beagle.
SOP versus Airmanship. It is not, I presume, SOP to intnetionally use high rates outside RVSM.
Fergineer.
Don't worry too much. TCAS manoeuvres you till its safety criteria are matched. If the other guy gets out of your way (i.e. coordinated TCAS) then you don't go so far from your original flight path, but get a certain separation (altitude dependant).
If the other aircraft doesn't manoeuvre, your corrective RA will last for longer before it slackens, but the size of the RA versus the time you have to manoeuvre means that you should still get the same ammount of seperation.
CPB
Thank you for your kind words. However, just to clarify, high vertical rates near your cleared level are not the only way, just a way. Also a common way. Other scenarios possible, but involving ATC errors.
Beagle.
SOP versus Airmanship. It is not, I presume, SOP to intnetionally use high rates outside RVSM.
Fergineer.
Don't worry too much. TCAS manoeuvres you till its safety criteria are matched. If the other guy gets out of your way (i.e. coordinated TCAS) then you don't go so far from your original flight path, but get a certain separation (altitude dependant).
If the other aircraft doesn't manoeuvre, your corrective RA will last for longer before it slackens, but the size of the RA versus the time you have to manoeuvre means that you should still get the same ammount of seperation.
CPB