TCAS Descend-RA below Min Safe Alt
Thread Starter
Join Date: Jul 2000
Location: Germany
Posts: 91
Likes: 0
Received 0 Likes
on
0 Posts
TCAS Descend-RA below Min Safe Alt
Hi everyone,
just coming from the SIM, where we got this TCAS-Scenario:
A/C is on downwind, 220 IAS, clean config, under radar vectors below MSA, company rules allow so (min 1000 ft above all relevant obstructions on the flightpath monitored by colour-coded contour-layer chart).
On the TCAS you abserve opposite traffic 1000ft above, which suddenly starts descending. So you get a TA, followed by a 'Descend RA'.
What do you do?
PF responded 'according to the book' by descending as TCAS said with about 1000-1500 fpm. This triggered a GPWS 'Pull Up'. While performing the GPWS Recovery Procedure (on the A320: Side Stick Full Back and Hold, TOGA Power) we ended up climbing at 4000+fpm well ABOVE the intruder, so missing him was plain luck. In the debrief the IP confirmed, that this was the very correct reaction.
I have to disagree and would say it was safer in this very case to disregard the RA, maintain level flight and watch the Intruder on the TCAS-Display.
As 'ultima ratio' I would accept to descent to the max. EGPWS-Altitude + 500 feet, if the relative altitude to the intruder becomes less than 300 ft. (EGPWS Terrain Display is switched on at TOD and displays max Terrain Elevation of the range selected on the ND).
Company Rules say TCAS has to be followed 'unless other emergency situations require appropriate action'.
Any opinions?
Cheers, MAX
just coming from the SIM, where we got this TCAS-Scenario:
A/C is on downwind, 220 IAS, clean config, under radar vectors below MSA, company rules allow so (min 1000 ft above all relevant obstructions on the flightpath monitored by colour-coded contour-layer chart).
On the TCAS you abserve opposite traffic 1000ft above, which suddenly starts descending. So you get a TA, followed by a 'Descend RA'.
What do you do?
PF responded 'according to the book' by descending as TCAS said with about 1000-1500 fpm. This triggered a GPWS 'Pull Up'. While performing the GPWS Recovery Procedure (on the A320: Side Stick Full Back and Hold, TOGA Power) we ended up climbing at 4000+fpm well ABOVE the intruder, so missing him was plain luck. In the debrief the IP confirmed, that this was the very correct reaction.
I have to disagree and would say it was safer in this very case to disregard the RA, maintain level flight and watch the Intruder on the TCAS-Display.
As 'ultima ratio' I would accept to descent to the max. EGPWS-Altitude + 500 feet, if the relative altitude to the intruder becomes less than 300 ft. (EGPWS Terrain Display is switched on at TOD and displays max Terrain Elevation of the range selected on the ND).
Company Rules say TCAS has to be followed 'unless other emergency situations require appropriate action'.
Any opinions?
Cheers, MAX
Join Date: Jul 2002
Location: London
Posts: 171
Likes: 0
Received 0 Likes
on
0 Posts
I'd have to agree with your PF and IP, and follow the RA.
Bear in mind that the RA commands are co-ordinated between the 2 conflicting aircraft, so it can create more problems if you choose to ignore the RA.
Our sop's say to follow any RA's, but to respect stall, gpws and windshear warnings-exactly as your pf did.
Just my opinion of course.
Bear in mind that the RA commands are co-ordinated between the 2 conflicting aircraft, so it can create more problems if you choose to ignore the RA.
Our sop's say to follow any RA's, but to respect stall, gpws and windshear warnings-exactly as your pf did.
Just my opinion of course.
Pegase Driver
Join Date: May 1997
Location: Europe
Age: 74
Posts: 3,682
Likes: 0
Received 0 Likes
on
0 Posts
we ended up climbing at 4000+fpm well ABOVE the intruder, so missing him was plain luck. In the debrief the IP confirmed, that this was the very correct reaction.
All this in theory and with 20/20 hindsight of course...
Join Date: Jul 2006
Location: USA
Posts: 451
Likes: 0
Received 0 Likes
on
0 Posts
And so...
And so...
What's the procedure (SOP) for dealing with an RA? What situations trump the RA maneuver?
1. GPWS or,
2. Windshear or,
3. Stall Warning.
Now then, what will happen in a 'coordinated' TCAS RA? You're told to descend into the ground. But, you don't...as per SOP (and common sense). So, you maybe hold your altitude or climb or go inverted or whatever. And, what you do will 'be told' to the other TCAS...and it will maneuver accordingly.
Ain't it a wonderful thing!!!!!
Fly safe,
PantLoad
What's the procedure (SOP) for dealing with an RA? What situations trump the RA maneuver?
1. GPWS or,
2. Windshear or,
3. Stall Warning.
Now then, what will happen in a 'coordinated' TCAS RA? You're told to descend into the ground. But, you don't...as per SOP (and common sense). So, you maybe hold your altitude or climb or go inverted or whatever. And, what you do will 'be told' to the other TCAS...and it will maneuver accordingly.
Ain't it a wonderful thing!!!!!
Fly safe,
PantLoad
Join Date: May 2000
Location: Seattle
Posts: 3,195
Likes: 0
Received 0 Likes
on
0 Posts
When you're being vectored under IFR, MSA and "company rules" don't really come into play. ATC has minimum vectoring altitudes for sectors that are laid out in a manner too complex for pilots to be acutely aware of, but they maintain minimum obstruction clearance. MSA is used for situational awareness and guidance in a lost comm situation, usually after a missed approach.
In your sim session, the PF's reaction was correct. I would do the same in the sim, though my reaction to the GPWS warning MIGHT be tempered IF I was in familiar territory AND I broke out of the clouds and was temporarily VMC during the RA maneuver.
One question I have is how soon after the GPWS Caution did the GPWS Warning sound? While you may not remember in the "heat of the moment," that is yet another indication of how dire your circumstances are with respect to imminent ground contact...
In your sim session, the PF's reaction was correct. I would do the same in the sim, though my reaction to the GPWS warning MIGHT be tempered IF I was in familiar territory AND I broke out of the clouds and was temporarily VMC during the RA maneuver.
One question I have is how soon after the GPWS Caution did the GPWS Warning sound? While you may not remember in the "heat of the moment," that is yet another indication of how dire your circumstances are with respect to imminent ground contact...
Pegase Driver
Join Date: May 1997
Location: Europe
Age: 74
Posts: 3,682
Likes: 0
Received 0 Likes
on
0 Posts
PantLoad :
Well that is theory, as we all know this did not work in Ueberlingen and in a load of other similar events. Sense reversal is currently the subject of a complex discussion process within ARINC.
Perhaps this will be solved with a Version 7.1 , but not today unfortunately.
what you do will 'be told' to the other TCAS...and it will maneuver accordingly.
Ain't it a wonderful thing!!!!!
Ain't it a wonderful thing!!!!!
Perhaps this will be solved with a Version 7.1 , but not today unfortunately.
Join Date: May 2000
Location: Seattle
Posts: 3,195
Likes: 0
Received 0 Likes
on
0 Posts
I don't think it's a software problem, rather a "wetware" problem...
I have had a TCAS RA situation where the other pilot did NOTHING in response to the RA. To add complexity to the situation, the ATC controller (in Japan) first gave us a turn away from the traffic, then changed his mind and turned us back into him. Our TCAS gave the appropriate RA, and we avoided him by 700' vertically, with ZERO lateral separation.
In that case there were AT LEAST 2 human errors (not including the one that got him co-altitude on the same airway in the opposite direction in the first place), but the TCAS still gave the appropriate RA advisories. Virtually any deviation from those advisories would have put us back in a collision situation.
To date I have heard of NO events where both pilots followed TCAS RA advisories and still collided with each other or the ground. If you know of any, please provide references.
I have had a TCAS RA situation where the other pilot did NOTHING in response to the RA. To add complexity to the situation, the ATC controller (in Japan) first gave us a turn away from the traffic, then changed his mind and turned us back into him. Our TCAS gave the appropriate RA, and we avoided him by 700' vertically, with ZERO lateral separation.
In that case there were AT LEAST 2 human errors (not including the one that got him co-altitude on the same airway in the opposite direction in the first place), but the TCAS still gave the appropriate RA advisories. Virtually any deviation from those advisories would have put us back in a collision situation.
To date I have heard of NO events where both pilots followed TCAS RA advisories and still collided with each other or the ground. If you know of any, please provide references.
We've been discussing this very thing (in my company) recently and for what it's worth came to the following conclusion.
The TCAS warns you of a "guaranteed" collision, imminently. The ground poses a threat, but it is subsequent to the anti-collision manoeuvre. The timescale may indeed be short, but we felt that the logical answer was to avoid the imminent threat first (i.e. follow the RA) and avoid the subsequent threat second. RAs clear the conflict quite quickly, which should allow time to pull up, even though MSA will have been "bust".
The TCAS warns you of a "guaranteed" collision, imminently. The ground poses a threat, but it is subsequent to the anti-collision manoeuvre. The timescale may indeed be short, but we felt that the logical answer was to avoid the imminent threat first (i.e. follow the RA) and avoid the subsequent threat second. RAs clear the conflict quite quickly, which should allow time to pull up, even though MSA will have been "bust".
Join Date: Dec 2001
Location: London
Posts: 186
Likes: 0
Received 0 Likes
on
0 Posts
ATC Watcher. As I understand it the sense reversal at Ueberlingen did work as intended. (TCAS can order an RA in the reverse direction to the original RA if the "master" TCAS detects failure of the other aircraft to comply iwth its RA).
But TCAS sense reversal is inhibited whenever the two aircraft are vertically separated by less than 100 ft. Removing that inhibition is one of the matters I believe is being discussed.
But TCAS sense reversal is inhibited whenever the two aircraft are vertically separated by less than 100 ft. Removing that inhibition is one of the matters I believe is being discussed.
Pegase Driver
Join Date: May 1997
Location: Europe
Age: 74
Posts: 3,682
Likes: 0
Received 0 Likes
on
0 Posts
Frangible : it depends on the definition of " as intended ". I took it in the global sense , intended to prevent the 2 a/c factually hitting each other.
From what I was told and understood :
In he current system (7.0) The initial RA sense will not be reversed until it has been displayed for at least nine seconds (unless the low Mode S address aircraft has a vertical rate higher than 2500 feet per minute and acts contrary to the RA). But : if both aircraft are going at the same rate towards each other ( one following its RA and the other one acting against it ) No RA reversal occurs. I believe that is the problem , But maybe we are talking about the same thing.
I understand Correction Proposal 112 E should rectify this and allow sense reversal at all times . ( decision on this next December I believe ? )
From what I was told and understood :
In he current system (7.0) The initial RA sense will not be reversed until it has been displayed for at least nine seconds (unless the low Mode S address aircraft has a vertical rate higher than 2500 feet per minute and acts contrary to the RA). But : if both aircraft are going at the same rate towards each other ( one following its RA and the other one acting against it ) No RA reversal occurs. I believe that is the problem , But maybe we are talking about the same thing.
I understand Correction Proposal 112 E should rectify this and allow sense reversal at all times . ( decision on this next December I believe ? )
Join Date: May 2000
Location: Seattle
Posts: 3,195
Likes: 0
Received 0 Likes
on
0 Posts
Still, the upcoming software release is only designed to correct a "wetware" problem: Pilots following contrary ATC instructions in spite of TCAS RA advisories.
If Pilots follow TCAS RAs instead of contravening ATC instructions, as is required by FAA (and other countries', AFAIK) regulations, there would be no requirement for the software change. It extends the "reach" of TCAS to situations for which I doubt it was originally designed.
While the change might add an extra margin in those cases where Pilots follow contravening ATC instructions, will we again blame TCAS when, in the next-generation conflict, a Pilot finally realizes he should follow his TCAS, and reverses at the same time the other Pilot receives his reversal RA?
If Pilots follow TCAS RAs instead of contravening ATC instructions, as is required by FAA (and other countries', AFAIK) regulations, there would be no requirement for the software change. It extends the "reach" of TCAS to situations for which I doubt it was originally designed.
While the change might add an extra margin in those cases where Pilots follow contravening ATC instructions, will we again blame TCAS when, in the next-generation conflict, a Pilot finally realizes he should follow his TCAS, and reverses at the same time the other Pilot receives his reversal RA?
Join Date: Nov 2004
Location: Orstralia
Posts: 171
Likes: 0
Received 0 Likes
on
0 Posts
Most interesting thread, and something I had wondered about. Is interesting to see a divergence of opinion - as experienced at Uberlingen...
Isn't this a false sense of security? I didn't think the TCAS traffic display was reliable in depicting brg/dist good enuf to base separation on.
...safer in this very case to disregard the RA, maintain level flight and watch the Intruder on the TCAS-Display.
Join Date: May 2000
Location: Seattle
Posts: 3,195
Likes: 0
Received 0 Likes
on
0 Posts
Yes, indeed! See the references to the Eurocontrol ACAS bulletins I posted here:
http://www.pprune.org/forums/showthr...=291507&page=8
http://www.pprune.org/forums/showthr...=291507&page=8
Pegase Driver
Join Date: May 1997
Location: Europe
Age: 74
Posts: 3,682
Likes: 0
Received 0 Likes
on
0 Posts
Intruder :
I am not familiar with the term " wetware" , but I guess you mean : "humanware " correct ?
Mixing highly complex automated system with humans reactions has always been a good recipe for disasters, not only in aviation.
You said :
Absolutely agreed . But the fact is pilots do not always follow the RAs in that case as many large scale data collection studies show, and post 2002 incidents investigations still show.
So many solutions : I give you the main 4 : 1 : change/retrain the current pilots and/or replace them by kids that will not challenge computer generated advices, 2 : link ACAS to the auto pilot, 3 : modify the software, 4. Accept a certain non compliance percentage and the accidents/loss of lives that will result because of that non-compliance.
Incidentally , for the anecdote , Airbus, just after the 5 A320 landing accidents following its introduction , proposed a mixture of solutions 1 and 3 .
I am not familiar with the term " wetware" , but I guess you mean : "humanware " correct ?
Mixing highly complex automated system with humans reactions has always been a good recipe for disasters, not only in aviation.
You said :
If Pilots follow TCAS RAs instead of contravening ATC instructions,[...] there would be no requirement for the software change.
So many solutions : I give you the main 4 : 1 : change/retrain the current pilots and/or replace them by kids that will not challenge computer generated advices, 2 : link ACAS to the auto pilot, 3 : modify the software, 4. Accept a certain non compliance percentage and the accidents/loss of lives that will result because of that non-compliance.
Incidentally , for the anecdote , Airbus, just after the 5 A320 landing accidents following its introduction , proposed a mixture of solutions 1 and 3 .
Join Date: May 2000
Location: Seattle
Posts: 3,195
Likes: 0
Received 0 Likes
on
0 Posts
[quote'I am not familiar with the term " wetware" , but I guess you mean : "humanware " correct ?[/quote]
Yes. You need to read a bit more science fiction.
I agree here. The problem is that we are going to make the system even more complex, such that it will be even less understandable by the Pilots.
I would prefer #2 for the current generations of airplanes (I fly the 747 Classic, and I doubt it would be included...). I see 2 problems with #3 that will significantly reduce its theoretical effectiveness:
a. Programming in reversals will not be immediately accepted by all Pilots. TCAS has been around for how long now, and we still have incidents caused allegedly because Pilots don't understand the advisories. How can we expect those same Pilots to understand and comply with even more complex commands and maneuvers, especially when there is even less margin for error?
b. There will still be Pilots who ignore TCAS RAs, and controllers who give contravening instructions. With all the increased information going to ATC via Mode S, it is even more likely that a controller will reverse his contravening call in the middle of an incident, at the same time TCAS gives a reversal RA to the other Pilot. Hence, there will likely be as many incidents where an RA reversal will be harmful rather than helpful.
I suspect there were good reasons for programming in the reversal delays/inhibits that are currently in the TCAS RA logic. I can't accept as raw fact the need to change the logic because of a few incidents, including the Ueberlingen mishap. In fact, I submit that it would be better for the education of the Pilots to be monitored by the appropriate regulatory officials in the airlines where non-compliance with RAs is a factor.
Yes. You need to read a bit more science fiction.
Mixing highly complex automated system with humans reactions has always been a good recipe for disasters, not only in aviation.
So many solutions : I give you the main 4 : 1 : change/retrain the current pilots and/or replace them by kids that will not challenge computer generated advices, 2 : link ACAS to the auto pilot, 3 : modify the software, 4. Accept a certain non compliance percentage and the accidents/loss of lives that will result because of that non-compliance.
a. Programming in reversals will not be immediately accepted by all Pilots. TCAS has been around for how long now, and we still have incidents caused allegedly because Pilots don't understand the advisories. How can we expect those same Pilots to understand and comply with even more complex commands and maneuvers, especially when there is even less margin for error?
b. There will still be Pilots who ignore TCAS RAs, and controllers who give contravening instructions. With all the increased information going to ATC via Mode S, it is even more likely that a controller will reverse his contravening call in the middle of an incident, at the same time TCAS gives a reversal RA to the other Pilot. Hence, there will likely be as many incidents where an RA reversal will be harmful rather than helpful.
I suspect there were good reasons for programming in the reversal delays/inhibits that are currently in the TCAS RA logic. I can't accept as raw fact the need to change the logic because of a few incidents, including the Ueberlingen mishap. In fact, I submit that it would be better for the education of the Pilots to be monitored by the appropriate regulatory officials in the airlines where non-compliance with RAs is a factor.
Pegase Driver
Join Date: May 1997
Location: Europe
Age: 74
Posts: 3,682
Likes: 0
Received 0 Likes
on
0 Posts
Intruder :
No, I do not read science fiction anymore, but maybe I should go back to it
I accept and follow your reasoning above, and would be willing to agree with your points , especially about the more complex human environment more sense reversals would induce , both in the cockpit and , in case of RA downlink, with the controllers. We might indeed open a can full of (new) worms .
But there are 2 points I disagree :
You said :
The incidents are not few and the consequences of a " mishap" as you put it are the major issue . For a refresher, in Ueberlingen we were " lucky" that one aircraft was half empty , the other one was a cargo and all the pieces fell in the countryside.
Now, with Murphy around, the next one could well be between a full 747 on an Asian route (i.e full ) with an Mediterranean charter 767 ( i.e also full ) which could kill 5 or 600 people. If the pieces fell above a city of populated areas the damage would also be quite significant. The political fallout of such an accident , especially after the lessons of Ueberlingen would no be sustainable for any Aviation official and politician.
That does not work today for many reasons. One of them starts with current priority TCAS training is being given. I cannot talk for the US, but I know of at least one large European carrier , whose TCAS training consist of a 6 pages photocopy document being tossed at you, followed by the inclusion of one or 2 RAs during the Sims checks . And we all know that no realistic ATC interface R/T ever takes place on most Sim sessions.
So, in Europe at least, we reap what has been planted .
No, I do not read science fiction anymore, but maybe I should go back to it
I accept and follow your reasoning above, and would be willing to agree with your points , especially about the more complex human environment more sense reversals would induce , both in the cockpit and , in case of RA downlink, with the controllers. We might indeed open a can full of (new) worms .
But there are 2 points I disagree :
You said :
I can't accept as raw fact the need to change the logic because of a few incidents, including the Ueberlingen mishap. .
Now, with Murphy around, the next one could well be between a full 747 on an Asian route (i.e full ) with an Mediterranean charter 767 ( i.e also full ) which could kill 5 or 600 people. If the pieces fell above a city of populated areas the damage would also be quite significant. The political fallout of such an accident , especially after the lessons of Ueberlingen would no be sustainable for any Aviation official and politician.
In fact, I submit that it would be better for the education of the Pilots to be monitored by the appropriate regulatory officials in the airlines where non-compliance with RAs is a factor
So, in Europe at least, we reap what has been planted .
Join Date: May 2000
Location: Seattle
Posts: 3,195
Likes: 0
Received 0 Likes
on
0 Posts
Yep... "Safety first" isn't worth the ink it's printed with in many places.
However, I am advocating that the FAA/CAA/JAA/whoever MAKE TCAS a priority in airlines' training! We get a lot more than you describe, and the scenarios we are presented with in the sim are very educating (e.g., do one with an engine out, when shifted to TA Only, and watch the pretty pictures as an airplane comes co-altitude...).
If safety really IS important, and TCAS really IS an important piece of it, the regulators have to get off their fat @$$3$ and DO the oversight they are paid to do!
However, I am advocating that the FAA/CAA/JAA/whoever MAKE TCAS a priority in airlines' training! We get a lot more than you describe, and the scenarios we are presented with in the sim are very educating (e.g., do one with an engine out, when shifted to TA Only, and watch the pretty pictures as an airplane comes co-altitude...).
If safety really IS important, and TCAS really IS an important piece of it, the regulators have to get off their fat @$$3$ and DO the oversight they are paid to do!
Pegase Driver
Join Date: May 1997
Location: Europe
Age: 74
Posts: 3,682
Likes: 0
Received 0 Likes
on
0 Posts
With that, I again fully agree with you ....
I am glad to hear hat your airline is taking the training seriously. The US generally has less "philosophical problems " with following computers advices that the " old Europe " .
However where it comes down to the real test in real life , olds habits die hard. :
incidentally in a carbon-copy of the Ueberlingen scenario that occurred in 2005 over France, the pilot that decided to follow the ATC clearance instead of the RA was not Russian, but a US trained 767 Captain of one of the major US carriers.
I am glad to hear hat your airline is taking the training seriously. The US generally has less "philosophical problems " with following computers advices that the " old Europe " .
However where it comes down to the real test in real life , olds habits die hard. :
incidentally in a carbon-copy of the Ueberlingen scenario that occurred in 2005 over France, the pilot that decided to follow the ATC clearance instead of the RA was not Russian, but a US trained 767 Captain of one of the major US carriers.