TACA aircraft crashed in Honduras
Join Date: Sep 1999
Posts: 541
Likes: 0
Received 0 Likes
on
0 Posts
This is by far the most interesting discussion on this site.
Like RWA,I have been going over and over possible scenarios that could have led to this tragedy.
As I see it we have two choices:
a)all available retardation devices were functioning normallly and the pilot was defeated by a combination of i)longish touchdown ii)wet runway iii)tailwind iv)early application of max REV which precluded the rwy GA option(we know they got spoilers from the "spoiler" call-technically,this is the only call a skipper needs before going with max REV on a short wet runway(involuntary action).However,if the touchdown was delayed(prolonged float or incorrect TCH) ,this is reason enough to throw the landing away despite the spoiler call(judgement action).In this case of course,he went for max REV regardless so its not really relevant.However,I think its interesting to mention because it describes a situation where two independent triggers are momentarily in conflict and as we all know the body doesnt always do what the brain actually wants.Very often the body will complete an involuntary action even though it wasnt really what you wanted to do.
b)A failure of the autobraking system to arm followed by failure of alternate brakes OR failure of autobrake system to arm followed by crew switching off ASKID prematurely(ie they jumped the gun and went straight to the NNC-didnt give alternate brakes a chance to work with ASKID).A looming runway end and knowledge that no overrun exists might have induced panic and led to such an error.The evidence rules out spoilers/reversers as being the cause.That leaves some anomaly in the brakes surely...
Some have declared that this cannot have been the case as the decel call was made and MED autobrake was set.But what if the autobrakes were not working and if the decel call was made on PFD trend alone?They had max REV and spoilers so decel might have "appeared" normal in the early stages.Reversers are efficient as a deceleration device in the high speed regime.They would start to become inefficient just around the time things started to appear wrong..ie 70 knots.On a short wet runway,thats an awfully nasty surprise.The warning for it is inhibited until 80 knots??And they did get a status message which the co-pilot wanted to clear??
Then theres the "quitame" call which we learnt might possibly be translated as "switch off.."
Switch off what?The ASKID/NSWSTRG switch ?Could it be anything else?
The Cardiff incident might help explain one or two things.What helped the skipper in that incident was that he used only IDLE REV so the failure of the autobrakes was readily apparent early on.Runway was longer and dry of course.He attempted to recycle(incorrect procedure) the ASKID/NSWSTRG switch after only 1000 feet runway used.In Toncontin,MAX REV was used so if autobrakes were not working,their failure would not have been apparent to the crew either by status mesage(inhibited),PFD trend(would still be healthy),or seat of the pants feel..Dangerous trap.If he had armed MAX,then the discrepancy in actual and anticipated deceleration would surely have been noticed.But because he only armed MED,the autobrake failure could have been masked by max reverse.
Perhaps an experienced and impartial Airbus pilot can answer the following questions:
a)What does the Decel light actually indicate ?
b)Do you call decel based on PFD trend or on the light?
c)how often do you use manual braking above 70 knots?
d)Have the safety recommendations from the Cardiff/Ibiza incidents been truly addressed by Airbus Industrie?Has the switch been renamed ASKID/NSWTSRG/BSCU?Have Airbus improved automated warnings to flightcrews concerning brake effectiveness following touchdown?
e)If autobrakes are used so trustingly and widely by Airbus pilots,should their failure to arm be cause for more than just a status message which can be inhibited by the computer during the touchdown and rollout phase above 80?
Like RWA,I have been going over and over possible scenarios that could have led to this tragedy.
As I see it we have two choices:
a)all available retardation devices were functioning normallly and the pilot was defeated by a combination of i)longish touchdown ii)wet runway iii)tailwind iv)early application of max REV which precluded the rwy GA option(we know they got spoilers from the "spoiler" call-technically,this is the only call a skipper needs before going with max REV on a short wet runway(involuntary action).However,if the touchdown was delayed(prolonged float or incorrect TCH) ,this is reason enough to throw the landing away despite the spoiler call(judgement action).In this case of course,he went for max REV regardless so its not really relevant.However,I think its interesting to mention because it describes a situation where two independent triggers are momentarily in conflict and as we all know the body doesnt always do what the brain actually wants.Very often the body will complete an involuntary action even though it wasnt really what you wanted to do.
b)A failure of the autobraking system to arm followed by failure of alternate brakes OR failure of autobrake system to arm followed by crew switching off ASKID prematurely(ie they jumped the gun and went straight to the NNC-didnt give alternate brakes a chance to work with ASKID).A looming runway end and knowledge that no overrun exists might have induced panic and led to such an error.The evidence rules out spoilers/reversers as being the cause.That leaves some anomaly in the brakes surely...
Some have declared that this cannot have been the case as the decel call was made and MED autobrake was set.But what if the autobrakes were not working and if the decel call was made on PFD trend alone?They had max REV and spoilers so decel might have "appeared" normal in the early stages.Reversers are efficient as a deceleration device in the high speed regime.They would start to become inefficient just around the time things started to appear wrong..ie 70 knots.On a short wet runway,thats an awfully nasty surprise.The warning for it is inhibited until 80 knots??And they did get a status message which the co-pilot wanted to clear??
Then theres the "quitame" call which we learnt might possibly be translated as "switch off.."
Switch off what?The ASKID/NSWSTRG switch ?Could it be anything else?
The Cardiff incident might help explain one or two things.What helped the skipper in that incident was that he used only IDLE REV so the failure of the autobrakes was readily apparent early on.Runway was longer and dry of course.He attempted to recycle(incorrect procedure) the ASKID/NSWSTRG switch after only 1000 feet runway used.In Toncontin,MAX REV was used so if autobrakes were not working,their failure would not have been apparent to the crew either by status mesage(inhibited),PFD trend(would still be healthy),or seat of the pants feel..Dangerous trap.If he had armed MAX,then the discrepancy in actual and anticipated deceleration would surely have been noticed.But because he only armed MED,the autobrake failure could have been masked by max reverse.
Perhaps an experienced and impartial Airbus pilot can answer the following questions:
a)What does the Decel light actually indicate ?
b)Do you call decel based on PFD trend or on the light?
c)how often do you use manual braking above 70 knots?
d)Have the safety recommendations from the Cardiff/Ibiza incidents been truly addressed by Airbus Industrie?Has the switch been renamed ASKID/NSWTSRG/BSCU?Have Airbus improved automated warnings to flightcrews concerning brake effectiveness following touchdown?
e)If autobrakes are used so trustingly and widely by Airbus pilots,should their failure to arm be cause for more than just a status message which can be inhibited by the computer during the touchdown and rollout phase above 80?
Last edited by Rananim; 13th Jul 2008 at 02:52.
Sun worshipper
Join Date: Dec 2001
Location: Paris
Posts: 494
Likes: 0
Received 0 Likes
on
0 Posts
RWA
No. Caution messages are inhibited in this phase.
No. Caution messages are inhibited in this phase.
No. Not the RA read-out (unexploitable because of the terrain), but the baro altitude tape on the PFD
No.They were, at Cardiff, shooting an ILS auto-land. There is no MDA, hence no amber tape on the PFD.
Funny that most training departments disagree with that "prudent" procedure.
A rejected landing is a very demanding maneuver, even more so that it requires a highly developed crew co-ordination, without any hesitation or misunderstanding between the two pilots.
What is recommended is a go-around - down to the flare - if the landing *geometry* doesn't seem right.
So, we stick to the SOPs : Once on the ground -> Full reverse on the T/Ls ; the spoilers will extend, then the reverse lights will come on, followed by engine acceleration...etc...
If I remember that discussion correctly, there was a few older pilots advocating that technique. And I don't recall anyone speaking of "folly". Their piloting error was leaving a T/L in the CL detent, with dire consequences.
The rest of the pilots tended to consider an aborted landing as a very very last resort.
A remark on that kind of procedure : There are SOPs and good SOPs cover all aspects of the airline operations. We cannot afford to have special techniques for special runways...and personal techniques should be eliminated.
We are no longer on a Piper Cub. On the lower ECAM there is an automatic page selection for each flight phase, here the *WHEEL* page,on which a diagram of the spoilers is displayed. Little green arrows above each spoiler confirm their deployment. The inoperative spoilers are shown in amber (that's another one for you to jump on !).
1. the flight crew received a warning,
2. Any warning was about something else
3. The pilot was only referring to the RA readout.
that exactly the same thing (amber warning passing 1,000 feet) happened at Cardiff.
that appears to have been a departure from the most prudent procedure, which I believe is to wait for the ‘spoilers’ call before deploying reverse thrust.
A rejected landing is a very demanding maneuver, even more so that it requires a highly developed crew co-ordination, without any hesitation or misunderstanding between the two pilots.
What is recommended is a go-around - down to the flare - if the landing *geometry* doesn't seem right.
So, we stick to the SOPs : Once on the ground -> Full reverse on the T/Ls ; the spoilers will extend, then the reverse lights will come on, followed by engine acceleration...etc...
That appears to be an entirely reasonable comment, in light of what was said on here a few months ago about the folly of the Congonhas pilot in deploying the reverser straight away?
The rest of the pilots tended to consider an aborted landing as a very very last resort.
A remark on that kind of procedure : There are SOPs and good SOPs cover all aspects of the airline operations. We cannot afford to have special techniques for special runways...and personal techniques should be eliminated.
can anyone please tell me how the PNF checks that spoilers have deployed fully on an A320? Can it be done visually - craning your neck through the side window - or is it just a matter of checking a light on the panel?
Join Date: Sep 2007
Location: Melbourne
Posts: 180
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by Lemurian
No....No....No.....No......
Originally Posted by Lemurian
If I remember that discussion correctly, there was a few older pilots advocating that technique.
Originally Posted by Lemurian
The rest of the pilots tended to consider an aborted landing as a very very last resort.
Originally Posted by Lemurian
We are no longer on a Piper Cub. On the lower ECAM there is an automatic page selection for each flight phase, here the *WHEEL* page,on which a diagram of the spoilers is displayed. Little green arrows above each spoiler confirm their deployment.
So, as I suspected, you can't actually SEE the spoilers from either seat - except maybe the extreme outboard ones? Yet ANOTHER panel display that pilots are taught to rely on, that may have gone wrong?
Please check the photographs on the first page of this thread. As far as I can see, only the extreme outboard spoiler (one of the in-flight spoilers) on the starboard side - that is, on the F/O's side - was deployed when the aeroplane crashed. Care to explain why, in your opinion, the pilots, facing an overrun situation, would actually have RETRACTED the ground spoilers moments before the crash?
We have a basic difference of approach, Lemurian. Most of my professional life was spent getting paid for 'lateral thinking.' You seem to take the view that, A. the 'book' is always right, and - therefore - that, B. anyone who gets into trouble just got what they deserved because they didn't follow the 'book.'
Not good enough. People are dying. Two lots in under a year, in similar circumstances, in the same marque of aeroplane. In that situation, IMO, you have ALSO to consider the possibility that the 'book' may be wrong to start with.
Sun worshipper
Join Date: Dec 2001
Location: Paris
Posts: 494
Likes: 0
Received 0 Likes
on
0 Posts
Blah blah blah blah blah
lateral thinking when you have just seconds or fractions of a second to react ?
When it happens, the book is modified (See Bernd's link, for instance).
Otherwise, just consider that had these aircrews applied what was written in the books, those accidents would have been avoided.
As, according to your well-known agenda, you are just putting the blame on one marque only, see the link of the EASA that Bernd provided and you'll find out that under :
nobody is immune to spoiler -among other systems - problems.
I also wonder what would have happened to one of these 737NGs, had they decided on an aborted landing.( The default hasn't been fixed, yet )
Lateral thinking, you said ?
The problem for you is two-fold :
1/- Your grasp of aero-technology is very limited, your understanding of airplane operations is blinded by prejudice and old memories of some local flying.
2/- You presume, like all of us in the comfort of your favourite arm-chair, to criticize actions about which you have no idea, especially about procedural discipline. Just look up on youtube this video, one of 32 about training and use a bit of imagination to visualize that every day, thousands of aircrew operate through the same vocabulary, the same operating procedures...standardisation is one of the main factors for aircraft safety and having for more years than I could count on insisted on standard ops and phraseology, I certainly can vouch that we have achieved, and still implement that kind of discipline. There is no "Oh ! See ! There is an amber light over there"..., John Wayne style, but this kind of dialogue and co-operation and co-ordination :
YouTube - Airbus A320 Video Training 27 of 32
You are no pilot and you can afford that luxury. I don't in an emergency.
Yes, you summed it up... I would only qualify your second sentence :" Perform any action not approved by the Book at your own peril - and your passengers'". You have to be damned sure of what you're doing and unfortunately, the more time goes by, the less knowledge we have of what is involved behind a switch.Flow diagrams are good for understanding the basics of a system. They have no pretention for accuracy.
And I, very arrogantly, I confess, have no sympathy for what I've called a "very flippant operation", whether at Cardiff or at Habsheim. For the crews at Congonhas and Tai-Pei, on the other hand, I can understand a lapse of concentration, which can happen to anyone of us.
lateral thinking when you have just seconds or fractions of a second to react ?
Two lots in under a year, in similar circumstances, in the same marque of aeroplane. In that situation, IMO, you have ALSO to consider the possibility that the 'book' may be wrong to start with.
Otherwise, just consider that had these aircrews applied what was written in the books, those accidents would have been avoided.
As, according to your well-known agenda, you are just putting the blame on one marque only, see the link of the EASA that Bernd provided and you'll find out that under :
SIN 2007-04 09/03/2007 Boeing 737-800 Spoiler Retraction Failure.
Withdrawn as a consequence of the issuance of FAA Emergency AD 2007-06-51
Withdrawn as a consequence of the issuance of FAA Emergency AD 2007-06-51
I also wonder what would have happened to one of these 737NGs, had they decided on an aborted landing.( The default hasn't been fixed, yet )
Lateral thinking, you said ?
The problem for you is two-fold :
1/- Your grasp of aero-technology is very limited, your understanding of airplane operations is blinded by prejudice and old memories of some local flying.
2/- You presume, like all of us in the comfort of your favourite arm-chair, to criticize actions about which you have no idea, especially about procedural discipline. Just look up on youtube this video, one of 32 about training and use a bit of imagination to visualize that every day, thousands of aircrew operate through the same vocabulary, the same operating procedures...standardisation is one of the main factors for aircraft safety and having for more years than I could count on insisted on standard ops and phraseology, I certainly can vouch that we have achieved, and still implement that kind of discipline. There is no "Oh ! See ! There is an amber light over there"..., John Wayne style, but this kind of dialogue and co-operation and co-ordination :
YouTube - Airbus A320 Video Training 27 of 32
Most of my professional life was spent getting paid for 'lateral thinking.'
You seem to take the view that, A. the 'book' is always right, and - therefore - that, B. anyone who gets into trouble just got what they deserved because they didn't follow the 'book.'
And I, very arrogantly, I confess, have no sympathy for what I've called a "very flippant operation", whether at Cardiff or at Habsheim. For the crews at Congonhas and Tai-Pei, on the other hand, I can understand a lapse of concentration, which can happen to anyone of us.
Perhaps an experienced and impartial Airbus pilot can answer the following questions:
a)What does the Decel light actually indicate ?
b)Do you call decel based on PFD trend or on the light?
c)how often do you use manual braking above 70 knots?
d)Have the safety recommendations from the Cardiff/Ibiza incidents been truly addressed by Airbus Industrie?Has the switch been renamed ASKID/NSWTSRG/BSCU?Have Airbus improved automated warnings to flightcrews concerning brake effectiveness following touchdown?
e)If autobrakes are used so trustingly and widely by Airbus pilots,should their failure to arm be cause for more than just a status message which can be inhibited by the computer during the touchdown and rollout phase above 80?
a)What does the Decel light actually indicate ?
b)Do you call decel based on PFD trend or on the light?
c)how often do you use manual braking above 70 knots?
d)Have the safety recommendations from the Cardiff/Ibiza incidents been truly addressed by Airbus Industrie?Has the switch been renamed ASKID/NSWTSRG/BSCU?Have Airbus improved automated warnings to flightcrews concerning brake effectiveness following touchdown?
e)If autobrakes are used so trustingly and widely by Airbus pilots,should their failure to arm be cause for more than just a status message which can be inhibited by the computer during the touchdown and rollout phase above 80?
b) neither. I call NO DECEL if there is no light.
c) 4 to 12 times per month and certainly every time I fly into LHR
d) I have no idea what safety reccomendations were made based upon Cardiff /Ibiza incidents - may I have some links, please? Switch is still labeled A/SKID & NW STRG. Chances are that Airbus Industrie thinks that every A320 rated pilot knows what the switch does or, in the worst case, knows what B and S in BSCU stand for, therefore obviating the need to add the BSCU label. Also I am unaware of any automated warning to flightcrews concerning brake effectiveness following touchdown, let alone the one that is need of improvement.
e) if they're not armed, there's no blue light in the autobrake selector pushbutton and no autobrake line on ECAM wheel page and that's not something that can be inhibited. Also autobrake fault caution is not inhibited from 1500 ft till 80 kt on rollout.
As I see it we have two choices:
Join Date: Sep 2007
Location: Melbourne
Posts: 180
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by Clandestino
b) neither. I call NO DECEL if there is no light.
Secondly, at what stage in the touchdown/rollout process would you normally expect to make the call?
I ask because touchdown, according to the transcript, was at 15:45:11.7. The F/O called 'manual braking' at 15:45:18.9 and 'DECEL' (actually 'decal' in the transcript) at 15:45:20.4.
Would that sort of timing (8-9 seconds from touchdown) be about right, in your experience?
Originally Posted by Clandestino
As I see it, the best choice is to wait till FDR data are made public and only then go into technicalities.
So it's pretty clear that there's a cover-up going on. Sure, we could just stop talking about the possible causes of the accident for the next year or so.
But, on the available evidence, that would be exactly what the 'investigators' want us (and everyone else) to do.......
As to the Cardiff/Ibiza incidents, this article discusses them pretty fully. It appears that the cause of the glitch has never been found, since the article says:-
"Manufacturer Airbus could not replicate the fault codes recorded by the BSCU BITE (built-in test), the CFDS (Central Fault and Display) nor the flight data recorder (FDR). However, very brief "micro-cut" power interruptions revealed a problem in the separate power supplies for the two BSCU channels. The FDR disclosed that the cycling of the A/SKID & N/W STRNG on finals had caused a swap-over in the active BSCU channels and a consequent silent loss of autobrake arming. In a word, "tricky."
A Very Halting Affair to Remember | Air Safety Week | Find Articles at BNET
Last edited by RWA; 14th Jul 2008 at 04:02. Reason: Edited to add link and explanatory comment
Join Date: Jan 2001
Location: Dallas, TX USA
Posts: 739
Likes: 0
Received 0 Likes
on
0 Posts
I finally found the time to locate the security camera and the aircraft on the runway at 155 kts, as recorded by the security camera. I found the following video on Youtube that provided the necessary information. It's an extended version of the security camera video we've seen earlier.
YouTube - Accidente toncontin
As others may have mentioned, the camera is located very near the north end of the terminal building. In the video, the clues needed to locate the camera are 2 cross taxiways where one is north and one just south of the camera position (the south taxiway is just visible before the camera points down at the end of the video), a linear row of buildings across the runway from the camera, a small building across the runway farther south, and a small road that runs north from the tarmac area located west of the main taxiway and running parallel to it. All are visible in the extended video and a satellite image.
Here's the google map satellite image that helped me to locate the camera:
Google Maps
The satellite image and this airport diagram also helped me to locate the aircraft.
http://www.costaricaaviation.com/cha...G_ARPT_DIA.PDF
The aircraft is recorded in the video (and measured at 155 kts) between the cross taxiway directly in front of the control tower/terminal building, and the next cross taxiway to the north.
The best estimate of the aircraft's nose location on the runway that I could come up with, is from video frame 20.031. Frame 20.031 is the frame I'm most comfortable giving an estimate for, as it has the best visual references. This location is around 2250 feet north of the displaced threshold (horizontal line immediately south of the piano keys in the satellite image), give or take maybe 100 feet.
By frame 20.828 the aircraft has traveled about 263 feet further and is still moving at 155kts ground speed. So by frame 20.828 at around 2500 feet north of the displaced threshold, the aircraft is still moving at 155kts, as captured by the security camera.
If interested, see my post #196 for details about the aircraft's measured ground speed.
YouTube - Accidente toncontin
As others may have mentioned, the camera is located very near the north end of the terminal building. In the video, the clues needed to locate the camera are 2 cross taxiways where one is north and one just south of the camera position (the south taxiway is just visible before the camera points down at the end of the video), a linear row of buildings across the runway from the camera, a small building across the runway farther south, and a small road that runs north from the tarmac area located west of the main taxiway and running parallel to it. All are visible in the extended video and a satellite image.
Here's the google map satellite image that helped me to locate the camera:
Google Maps
The satellite image and this airport diagram also helped me to locate the aircraft.
http://www.costaricaaviation.com/cha...G_ARPT_DIA.PDF
The aircraft is recorded in the video (and measured at 155 kts) between the cross taxiway directly in front of the control tower/terminal building, and the next cross taxiway to the north.
The best estimate of the aircraft's nose location on the runway that I could come up with, is from video frame 20.031. Frame 20.031 is the frame I'm most comfortable giving an estimate for, as it has the best visual references. This location is around 2250 feet north of the displaced threshold (horizontal line immediately south of the piano keys in the satellite image), give or take maybe 100 feet.
By frame 20.828 the aircraft has traveled about 263 feet further and is still moving at 155kts ground speed. So by frame 20.828 at around 2500 feet north of the displaced threshold, the aircraft is still moving at 155kts, as captured by the security camera.
If interested, see my post #196 for details about the aircraft's measured ground speed.
Last edited by Flight Safety; 14th Jul 2008 at 04:30.
Join Date: Jan 2001
Location: Dallas, TX USA
Posts: 739
Likes: 0
Received 0 Likes
on
0 Posts
I just wanted to add that I couldn't locate the camera precisely enough to claim an accuracy any greater than around plus or minus 100 feet. I could locate the camera generally, but for several reasons related to the information I had available, I couldn't locate it more precisely than somewhere on the north end of the terminal building.
Last edited by Flight Safety; 14th Jul 2008 at 11:36.
Join Date: Feb 2008
Location: Edinburgh, UK
Posts: 34
Likes: 0
Received 0 Likes
on
0 Posts
RWA:
Since this seems to be the case on most types, what would you suggest?
How long do investigations into major accidents normally take?
So, as I suspected, you can't actually SEE the spoilers from either seat - except maybe the extreme outboard ones? Yet ANOTHER panel display that pilots are taught to rely on, that may have gone wrong?
So it's pretty clear that there's a cover-up going on.
Join Date: Aug 2005
Location: fl
Posts: 2,525
Likes: 0
Received 0 Likes
on
0 Posts
Looking at the last 20 posts looks like we are in idle mode until we get some FDR info. I still think landing downhill on a wet 5400 ft runway with a cliff dropping 70 ft and landing a tad long in an A320 with a 10 knot tailwind left little brake failure problems to finish the accident. Why did they do that?
Lemaurin Re “#338. The autobrake use does not prevent you from judging by yourself, either visually or through the seat of your pant (ies) whether your deceleration is adequate or not, so your comments on this respect are a bit strange to me.”
As stated, auto brake does not ‘prevent’ judgement of deceleration, we agree; but I suggest that it increases the difficulty / complexity of the judgement.
Finally remember that as an older pilot, presumable with greater experience, the viewpoint of auto systems is not the same as an less experience pilot, who in modern operation probably has less opportunity to gain the high levels of experience to perceive/deduce adequate aircraft/system interaction, e.g. lower frequency of manual brake landings.
The skills issues here are those of perception and system/aircraft performance relationships; the difference between knowledge, as in ‘know-what’ and knowledge, of ‘know-how’.
As stated, auto brake does not ‘prevent’ judgement of deceleration, we agree; but I suggest that it increases the difficulty / complexity of the judgement.
- With pure manual brakes pilots form a relationship (experience) between the tactile input (foot force/position) and the deceleration achieved. Variations in this can be interpreted as a change in runway friction (ignoring brake failures which are normally alerted / indicated independently).
- Using reversers has a similar tactile / deceleration relationship, although with greater variability with speed, but alone does it does not relate to runway friction, i.e. less info.
With brakes and reverse, the result is similar to brakes alone, a complete, meaningful feedback loop. With experience, a pilot might judge the component of deceleration contributed from each system. - When auto brake alone is used, pilots are distanced from the (deceleration) system by the reduced correlation between the input and the sensed deceleration (no tactile feedback from the DECEL light), and auto brake lacks any modulating control – no variable foot force, thus an incomplete loop.
- The combination of auto brake and reverse further distances the pilots from the system as it is more difficult to judge what proportion of the deceleration is being provided by each system. This is adequately explained in the Boeing presentation 'Flt Ops / Flying Technique … Landing on Slippery Runways'. (from #307) where safetypee outlines the problem of perception and the likelihood of misinterpreting poor braking on a low friction runway as a brake failure when reverse is cancelled.
Finally remember that as an older pilot, presumable with greater experience, the viewpoint of auto systems is not the same as an less experience pilot, who in modern operation probably has less opportunity to gain the high levels of experience to perceive/deduce adequate aircraft/system interaction, e.g. lower frequency of manual brake landings.
The skills issues here are those of perception and system/aircraft performance relationships; the difference between knowledge, as in ‘know-what’ and knowledge, of ‘know-how’.
Selection of reverse at touchdown:
In placing REV early in the checklist, presumably Airbus has balanced the risk of failure of rev to cancel in the remote scenario of a GA when on the runway, against the routine need for prompt deceleration. Why then do operators out-think the manufacturer and teach (SOP) a delay, waiting for spoilers?
Use of DECEL light:
The check of this light after touchdown confirms that the required decel level has been achieved and a checks that auto brake has not been inadvertently disabled (Cardiff ?). Thus if the light is ON say nothing (ops normal); if the light is off (with auto brake armed) then call the failure or action – manual braking required.
With auto brake off (no DECEL light), do operators still call for manual brakes as a reminder –audio clutter, error opportunity - call not heard/made with the belief that auto brake is armed – human factors in SOPs !
This supports my previous post illustrating the additional complexity in operations with auto brake; you have to consider another normal and failure case. In addition, the communication uses the weakest method – the human audio channel is the first to degrade under high workload / stress.
Systems failing:
I understand that Airbus has informed operators that there were no malfunctions identified on the FDR; anyone able to confirm this?
When looking for similarities, I note that in accident scenarios humans appear to fail more often than do aircraft systems;- mis-selection, incorrect procedure, mis-assessment of the situation, inappropriate decision, failure to understand implications, etc. Many, if not all of these appear to be associated with this and recent overrun accidents, thus the human and associated factors (met, runway surface, distance, planning) might be chosen as the main theme of the investigation as opposed to the aircraft.
In placing REV early in the checklist, presumably Airbus has balanced the risk of failure of rev to cancel in the remote scenario of a GA when on the runway, against the routine need for prompt deceleration. Why then do operators out-think the manufacturer and teach (SOP) a delay, waiting for spoilers?
Use of DECEL light:
The check of this light after touchdown confirms that the required decel level has been achieved and a checks that auto brake has not been inadvertently disabled (Cardiff ?). Thus if the light is ON say nothing (ops normal); if the light is off (with auto brake armed) then call the failure or action – manual braking required.
With auto brake off (no DECEL light), do operators still call for manual brakes as a reminder –audio clutter, error opportunity - call not heard/made with the belief that auto brake is armed – human factors in SOPs !
This supports my previous post illustrating the additional complexity in operations with auto brake; you have to consider another normal and failure case. In addition, the communication uses the weakest method – the human audio channel is the first to degrade under high workload / stress.
Systems failing:
I understand that Airbus has informed operators that there were no malfunctions identified on the FDR; anyone able to confirm this?
When looking for similarities, I note that in accident scenarios humans appear to fail more often than do aircraft systems;- mis-selection, incorrect procedure, mis-assessment of the situation, inappropriate decision, failure to understand implications, etc. Many, if not all of these appear to be associated with this and recent overrun accidents, thus the human and associated factors (met, runway surface, distance, planning) might be chosen as the main theme of the investigation as opposed to the aircraft.
Join Date: Jan 2006
Location: USA
Posts: 107
Likes: 0
Received 0 Likes
on
0 Posts
Overall, the use of auto brake during landing adds ‘complexity’ to the operation with little benefit
Join Date: Sep 2007
Location: Melbourne
Posts: 180
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by DL-EDI
Since this seems to be the case on most types, what would you suggest?
In light of the fact that only one out of ten spoiler panels was deployed when the aeroplane crashed, and the further information that the First Officer made the spoiler call on the basis of an instrument reading and not from direct observation, we now have to add in another possibility.
Namely that the ground spoilers did not in fact deploy.
Originally Posted by DL-EDI
How long do investigations into major accidents normally take?
The information in the CVR and the FDR, and all other details of what the aeroplane and the crew actually DID, is usually published as soon as it is available.
Announcements as to 'probable causes' vary a lot, since the investigators don't always have enough information on a given accident. But in a straightforward case like this (where all data has been recovered, the aeroplane didn't burn or crash in the sea, the accident was in daylight, and one of the pilots survived) I'd say that six to eight weeks is about typical.
Sure, the formal report (which has to be argued through exhaustively with the lawyers representing all the parties involved) will probably take years to finalise.
But I for one cannot recall a case where absolutely NO factual information has been released six weeks after the event; except for an 'unofficial,' heavily 'doctored' CVR transcript, and an 'unattributed' mention of pilot error.
It is, however, significant (to me anyway) that they've re-opened Toncontin to Cat. C and D traffic. That presumably means that neither the airport nor the runway was to blame. Which only leaves pilot error or some sort of malfunction as 'possible causes.'
There's a cover-up going on, all right. The only question is how long they can keep it up.
Join Date: Sep 2007
Location: Melbourne
Posts: 180
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by PJ2
For RWA and possibly others, there is no SOP requiring a "decel" call nor is there a "spoilers", "spoilers up" or "no spoilers" call as an SOP.
Another possibility. Suppose that the CVR recording was indistinct, and the transcript is therefore inaccurate? And that what the F/O actually called ('decal' in the transcript) was 'NO DECEL'?
As Alf5071h says:-
Thus if the light is ON say nothing (ops normal); if the light is off (with auto brake armed) then call the failure or action – manual braking required.
15:45:18.9 [SIC] manual braking
15:45:20.4 [SIC] no decel
15:45:21.8 [CAM] (sound similar to engine spool up)
15:45:26.4 [CAM] (soun similar to engine spool down)
15:45:27.2 [SIC] SEVENTY Knots
15:45:28.3 [PIC] quíteme quíteme el….!
15:45:28.8 [CAM] (sound similar to throttle movement)
15:45:30.5 [CAM] (sound of click)
15:45:31.5 [CAM] (sounds of single chime)
15:45:33.0 [?] (sounds of inhale/exhale/breathing)
15:45:34.2 [CAM] (sound of thump/clunk)
15:45:34.8 [PIC] Puta.
15:45:35.2 [CAM] (sound of thump/clunk)
15:45:35.7 [CAM] (sound of thump/clunk)
15:45:36.0 [CAM] (very loud sounds begin, and continue to end of recording)
15:45:38.3 [CAM] (end of recording)
15:45:38.5 End of transcript
Last edited by RWA; 15th Jul 2008 at 05:31.
Sun worshipper
Join Date: Dec 2001
Location: Paris
Posts: 494
Likes: 0
Received 0 Likes
on
0 Posts
Let's edit that transcript, remove the sensationalism and concentrate on the landing phase:
"15:45:04.2 [EGPWS] Forty
15:45:04.9 [EGPWS] Thirty
15:45:05.7 [EGPWS] Twenty
15:45:08.2 [EGPWS] FIVE
15:45:11.7 [CAM] (Sound similar touchdown & landing gear compression) T=0
15:45:12.1 [CAM] (sound similar throttle movement full reverse position) T + .4 sec
15:45:13.4 [SIC] Spoiler T + 1.7
15:45:17.0 [SIC] rev(ersers) green T + 5.3
15:45:18.9 [SIC] manual braking T + 7.2
15:45:20.4 [SIC] decal T + 8.7
15:45:21.8 [CAM] sounds of engine spool up T + 10.1
15:45:27.2 [SIC] SEVENTY Knots T + 15.5
15:45:31.5 [CAM] (sounds of single chime) T + 19.8
15:45:34.2 [CAM] (sound of thump/clunk) T + 22.5
There are a few aspects that we can gather :
Fortunately, we have a few professionals, among whom :
5071h
bubbers44
bsieker
I leave that as the final word.
"15:45:04.2 [EGPWS] Forty
15:45:04.9 [EGPWS] Thirty
15:45:05.7 [EGPWS] Twenty
15:45:08.2 [EGPWS] FIVE
15:45:11.7 [CAM] (Sound similar touchdown & landing gear compression) T=0
15:45:12.1 [CAM] (sound similar throttle movement full reverse position) T + .4 sec
15:45:13.4 [SIC] Spoiler T + 1.7
15:45:17.0 [SIC] rev(ersers) green T + 5.3
15:45:18.9 [SIC] manual braking T + 7.2
15:45:20.4 [SIC] decal T + 8.7
15:45:21.8 [CAM] sounds of engine spool up T + 10.1
15:45:27.2 [SIC] SEVENTY Knots T + 15.5
15:45:31.5 [CAM] (sounds of single chime) T + 19.8
15:45:34.2 [CAM] (sound of thump/clunk) T + 22.5
There are a few aspects that we can gather :
- The touch-down phase
The automated Rad Alt calls between 40 ft and 20 ft give us a vertical speed of 800 ft/min, which is about normal for the visual glide slope of 5.4° for Runway 02... Landing seems routine.
The same calculations applied to the times elapsed between 20 ft and 5 ft give a vertical speed of 360 ft/min, revealing the begining of the flare.
The average vertical speed between the 5ft call and the touch-down shows an overcooking of the flare...i.e a short floating bit over the runway. Not really a big deal but worth some 260 meters that will, be lacking later. - RWA deceleration theory :
andSuppose that the CVR recording was indistinct, and the transcript is therefore inaccurate? And that what the F/O actually called ('decal' in the transcript) was 'NO DECEL'?
So let's see :the ground spoilers did not in fact deploy.
The aircraft was not decelerating until the supposed "no decel" call and we take the deceleration from the time the sound of spooling-up engines is recorded, right. So from T+10.1 to T+15.5 (the 70 kt call,IAS corrected with 10 kt tailwind), 5.4 seconds have elapsed and the speed was brought from 148 kt (76 m/s) to 80 (41 m/s). That makes us a decel of 6.5 m/s² or .65 G...WhoooooAH ! That's some manual braking on a wet runway !
With this number, they woud have come to a complete stop in a mere 11.7 seconds with a stopping distance of 445 m.
So, really what are we talking about : an airplane with some faults on its major components stopping on the runway. Call the mods ! This is ridiculous ! - The spoilers case
Basing the spoiler-not-deployed theory on the basis of one picture is, to say the least, very bold.In light of the fact that only one out of ten spoiler panels was deployed when the aeroplane crashed, and the further information that the First Officer made the spoiler call on the basis of an instrument reading and not from direct observation, we now have to add in another possibility.
Namely that the ground spoilers did not in fact deploy.
For the spoilers to operate, we need :
a/-Electrical power
b/-Three spoiler/elevator computers, independentfrom one another
c/-Hydraulic power from three different systems powering a servo-jack for each panel
d/-An arming lever, with only an up/down position
e/-A logic.
Knowing also that a hydraulic failure will lock the relevant panel in place, tell me of all the other possibilities that brought that configuration after the crash. Can you ?
Fortunately, we have a few professionals, among whom :
5071h
I understand that Airbus has informed operators that there were no malfunctions identified on the FDR; anyone able to confirm this?
When looking for similarities, I note that in accident scenarios humans appear to fail more often than do aircraft systems;- mis-selection, incorrect procedure, mis-assessment of the situation, inappropriate decision, failure to understand implications, etc. Many, if not all of these appear to be associated with this and recent overrun accidents, thus the human and associated factors (met, runway surface, distance, planning) might be chosen as the main theme of the investigation as opposed to the aircraft.
When looking for similarities, I note that in accident scenarios humans appear to fail more often than do aircraft systems;- mis-selection, incorrect procedure, mis-assessment of the situation, inappropriate decision, failure to understand implications, etc. Many, if not all of these appear to be associated with this and recent overrun accidents, thus the human and associated factors (met, runway surface, distance, planning) might be chosen as the main theme of the investigation as opposed to the aircraft.
I still think landing downhill on a wet 5400 ft runway with a cliff dropping 70 ft and landing a tad long in an A320 with a 10 knot tailwind left little brake failure problems to finish the accident. Why did they do that?
This landing was not an emergency landing.
Join Date: Feb 2008
Location: Edinburgh, UK
Posts: 34
Likes: 0
Received 0 Likes
on
0 Posts
I'm not 'making suggestions
As far as briefings on the progress of investigations, the known facts, and the general lines of enquiry being pursued are concerned, usually no time at all.
The information in the CVR and the FDR, and all other details of what the aeroplane and the crew actually DID, is usually published as soon as it is available.
The information in the CVR and the FDR, and all other details of what the aeroplane and the crew actually DID, is usually published as soon as it is available.
Opinion
NVpilot, this is an excellent forum for sharing and respecting opinions.
In forming an opinion, I attempt to consider as many relevant circumstances as possible.
Perhaps your view and that of Lemaurin previously, places too greater emphasis on normal (routine) operations, particularly in the context of this accident.
I hope that I have included relevant aspects of the situation where the crew may have been stressed by external or internal pressures, i.e. previous ATC exchanges, operational expectation, or a demanding landing on a short, wet runway. It is in these conditions that the weaknesses of systems – be they human, system interface, or procedure, have greater relevance; this is the context of my opinion.
Key factors in avoiding accidents, the human perceptions and behaviors, exist in normal operations. Every day we avoid or detect and correct errors, we perceive operating hazards and choose safe courses of action according to the situation. Somewhere, or somehow this process failed in this landing - ‘the crew thought it was a normal landing’.
You provide a caveat for this landing – ‘I would use manual braking’ – why? What perception (in hindsight), directs this course of action, what’s different about this situation that results in the ‘bending’ of you opinion of the value of auto brake?
Why is this runway any different to a contaminated runway, on which you would use less than the max braking available (an aspect where we have vastly differing opinions)?
May I suggest that just because the runways is short, wet, or requires a steep approach, it does not call for a change in the choice of braking; precision and promptness in flying and retarding actions, yes. However, if there are indications that the landing safety margins might be eroded (tailwind, slippery runway) then manual braking would be a good precaution.
IMHO manual braking provides an earlier indication that the runway friction may not be as expected, by having better sensing than with auto brake; in more extreme circumstances with auto brake and low experience of slippery conditions, low friction might not be detected until reverse is cancelled – too late.
A problem for the investigator is that the thinking activities of perception and choice are not self evident from the FDR or CVR, and even with a crew debrief, they are subject to hindsight bias.
In the context of this accident, the crew appear to have followed normal procedures (calls at touchdown and rollout), including 70 kts. The ‘normal’ habit would be to cancel reverse, but if the runway was judged to be slippery then the habit should be changed - non normal operation. The failure to change implies that the either crew did not perceive that the rollout was not progressing as expected, or that they realize something was wrong and chose an incorrect course of action; it is also possible that the runway friction decreased toward the end of the rollout – too late. The contributions and precursors to these situations are numerous and hidden in the depths of James Reason’s Swiss Cheese model or the ‘risks of organizational accidents’.
The view above assumes a slippery runway which some people have dismissed; I remain open minded. The runway charts are annotated ‘during rainy season, touchdown zones extremely slippery’; it was or had been raining. Reports of recent runway improvements may have improved the surface, but from other incident examples, new asphalt (and rain) can produce a very slippery greasy surface.
Although the runway has been assessed as safe after the accident, the details are not available; I also note that … “there is not, at present, a common friction index for all ground friction measuring devices. Hence it is not practicable at the present time to determine aeroplane performance on the basis of an internationally accepted friction index measured by ground friction devices”. CS 25 - AMC 25.1591
It may be very difficult to determine the significant contributing factors in this accident, but if as it appears that “in the Captain’s opinion” the attempt to land was safe, then the investigators might determine (in their opinion) some factors which show that the decision was imprudent or more likely the situation was not as expected.
But that’s just an opinion.
In forming an opinion, I attempt to consider as many relevant circumstances as possible.
Perhaps your view and that of Lemaurin previously, places too greater emphasis on normal (routine) operations, particularly in the context of this accident.
I hope that I have included relevant aspects of the situation where the crew may have been stressed by external or internal pressures, i.e. previous ATC exchanges, operational expectation, or a demanding landing on a short, wet runway. It is in these conditions that the weaknesses of systems – be they human, system interface, or procedure, have greater relevance; this is the context of my opinion.
Key factors in avoiding accidents, the human perceptions and behaviors, exist in normal operations. Every day we avoid or detect and correct errors, we perceive operating hazards and choose safe courses of action according to the situation. Somewhere, or somehow this process failed in this landing - ‘the crew thought it was a normal landing’.
You provide a caveat for this landing – ‘I would use manual braking’ – why? What perception (in hindsight), directs this course of action, what’s different about this situation that results in the ‘bending’ of you opinion of the value of auto brake?
Why is this runway any different to a contaminated runway, on which you would use less than the max braking available (an aspect where we have vastly differing opinions)?
May I suggest that just because the runways is short, wet, or requires a steep approach, it does not call for a change in the choice of braking; precision and promptness in flying and retarding actions, yes. However, if there are indications that the landing safety margins might be eroded (tailwind, slippery runway) then manual braking would be a good precaution.
IMHO manual braking provides an earlier indication that the runway friction may not be as expected, by having better sensing than with auto brake; in more extreme circumstances with auto brake and low experience of slippery conditions, low friction might not be detected until reverse is cancelled – too late.
A problem for the investigator is that the thinking activities of perception and choice are not self evident from the FDR or CVR, and even with a crew debrief, they are subject to hindsight bias.
In the context of this accident, the crew appear to have followed normal procedures (calls at touchdown and rollout), including 70 kts. The ‘normal’ habit would be to cancel reverse, but if the runway was judged to be slippery then the habit should be changed - non normal operation. The failure to change implies that the either crew did not perceive that the rollout was not progressing as expected, or that they realize something was wrong and chose an incorrect course of action; it is also possible that the runway friction decreased toward the end of the rollout – too late. The contributions and precursors to these situations are numerous and hidden in the depths of James Reason’s Swiss Cheese model or the ‘risks of organizational accidents’.
The view above assumes a slippery runway which some people have dismissed; I remain open minded. The runway charts are annotated ‘during rainy season, touchdown zones extremely slippery’; it was or had been raining. Reports of recent runway improvements may have improved the surface, but from other incident examples, new asphalt (and rain) can produce a very slippery greasy surface.
Although the runway has been assessed as safe after the accident, the details are not available; I also note that … “there is not, at present, a common friction index for all ground friction measuring devices. Hence it is not practicable at the present time to determine aeroplane performance on the basis of an internationally accepted friction index measured by ground friction devices”. CS 25 - AMC 25.1591
It may be very difficult to determine the significant contributing factors in this accident, but if as it appears that “in the Captain’s opinion” the attempt to land was safe, then the investigators might determine (in their opinion) some factors which show that the decision was imprudent or more likely the situation was not as expected.
But that’s just an opinion.
Join Date: Aug 2005
Location: fl
Posts: 2,525
Likes: 0
Received 0 Likes
on
0 Posts
I found my sink rate around 1200 fpm on a 135 knot approach with no wind on the 757 as normal. With a 10 knot tailwind it must have been about 1400 fpm to clear the hills and touch down in the 800 ft zone we used past the displaced threshold. This is the last 50 feet after clearing that hill by 40 feet. My friend who operates our airline at TGU said they excavated another 300 feet out of that hill to extend the runway to the south. That might help a bit on take off until they move the operation to the military field about 30 miles west of TGU. I am not sure how this affects landing when they complete it since terrain may not allow the landing threshold to change much.
Join Date: Aug 2005
Location: fl
Posts: 2,525
Likes: 0
Received 0 Likes
on
0 Posts
Lemurian, Thanks for the support. My last post showed my 1200 fpm as normal and I noticed you showed the 40 to 20 ft calls at 800fpm. The reason for that is we always got a 40 call crossing a hill turning final and that hill was well above field elevation. You could not land at TGU if you didn't do that portion before flair at about 1200 fpm. Even though we don't use the light airplane vasi, we use about the same angle, over 5 degrees to clear those hills turning final. Sometimes with more than a 10 knot headwind we could do a silent approach because we didn't exceed 1200fpm but most of the time you heard sink rate, sink rate before landing.
Thanks for your interest in TGU. I love that airport and the people that work there. I spent my last seven years flying there because of the wonderful people and the fun and safe approach. It even made flying fun again flying for a mega airline that doesn't want people to have fun.
Thanks for your interest in TGU. I love that airport and the people that work there. I spent my last seven years flying there because of the wonderful people and the fun and safe approach. It even made flying fun again flying for a mega airline that doesn't want people to have fun.