Print from Airliners.net discussion forum
http://www.airliners.net/aviation-forums/general_aviation/read.main/5638163/

Topic: Superjet Pilot Switched Off Taws Before Crash
Username: RubberJungle
Posted 2012-12-18 02:46:39 and read 24594 times.

This doesn't really come as a huge surprise, but the captain of the SSJ100 which crashed in Indonesia turned off the terrain-awareness system after it sounded. Simulations show there was plenty of time to save the aircraft.

http://www.flightglobal.com/news/art...ced-terrain-warning-system-380320/

Why do you think pilots still seem to disregard the "pull up" warning?

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: Semaex
Posted 2012-12-18 03:01:21 and read 24501 times.

When in doubt, trust the aircraft. It's the first rule of IFR flying.

However, there's no rule that suits to pilots who are so full of themselves that they don't even dare think there is something to doubt. It's hopeless and it always will be.   

I'm glad I fly with airlines that choose their pilots based on more important things than who has the biggest cojones.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: nighthawk
Posted 2012-12-18 03:11:20 and read 24424 times.

The article states that the system gave a warning of "terrain ahead, pull up" and "avoid terrain". Are these alerts given in english or Russian on the Superjet and other Russian/Ukrainian built aircraft?

I'm not suggesting this is a factor in the accident, I'm just curious if these aircraft that are predominantly operated in Russia and former Soviet Union countries still use english for alerts.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: tdscanuck
Posted 2012-12-18 05:26:04 and read 23780 times.

Quoting RubberJungle (Thread starter):
Why do you think pilots still seem to disregard the "pull up" warning?

They thought it was a database problem, which means they'd lost situational awareness. Once that happens, confirmation bias becomes a huge issue...the crew believes they know what's going on (even though they don't) so they fit any new information into their theory of what's going on rather than revising the theory.

*If* you believe the terrain database is incorrect, then you *expect* erroneous terrain warnings. In a perverse way, continued warnings in that situation confirm to the crew that they're correct.

Tom.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: rfields5421
Posted 2012-12-18 05:57:02 and read 23547 times.

Quoting RubberJungle (Thread starter):
Why do you think pilots still seem to disregard the "pull up" warning?

The only person on this forum who has flown over that area and has detailed knowledge of the area, the terrain and where the flight was supposed to be is mandala499.

In his detailed posts after the crash, he explained how the pilots had drifted out of the area where the flight was supposed to be taking place.

Also that they were probably fixated on avoiding a higher 'more dangerous' nearby mountain, not realizing they were below the level of the mountain/ ridge they hit.

As Tom said - they lost awareness of where they were.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: dynamo12
Posted 2012-12-18 09:13:21 and read 21214 times.

I'm in a different field, but we have somewhat similar systems.

A question I have for pilots is how often are there actual "database problems" that cause false alarms. What I've noticed is that it only takes a few false alarms for the response to the alarm to degrade very significantly. Particularly if the actual alarm situation is infrequent to non-existent.

Folks designing safety systems, particularly alarms, should spend a very significant amount of time avoiding false alarms. This flows from simple math. If 99.99% of the time there is no problem (1 in 10,000 incident rate), and in even 1% of the no problem time there is a false alarm you end up with 100 false alarms per real alarm. That just doesn't work.

My impression was that TAWS worked to avoid this issue pretty carefully, but I'm curious about the reality.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: JoeCanuck
Posted 2012-12-18 09:32:33 and read 20702 times.

Pilots fly in conditions worse than what the Superjet was flying in every day without problems, even with instrument issues. The chances of any 'new' problem happening is pretty slim.

As Tom mentioned, it's a perception issue...which is a problem when training tells you that something is infallible, like your instruments...as in; If one infallible instrument can fail, what's to say that all of the infallible instruments haven't failed but they just look like they are working? In this case, it seems that nothing failed but the pilot thought something did.

Anybody, including the very experienced and well trained, can be overwhelmed under some circumstances. When that happens, the brain locks up and nothing makes sense.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: redzeppelin
Posted 2012-12-18 09:34:06 and read 20641 times.

I have a lay man's question: How is TAWS connected with a database? I guess that I always assumed that TAWS was based on active radar or some other means of directly detecting an obstruction. But the reference to a database makes it seem that TAWS is comparing the aircraft's position and altitude to a topographic database. Is that the case?

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: travelavnut
Posted 2012-12-18 09:47:10 and read 20297 times.

Quoting redzeppelin (Reply 7):
guess that I always assumed that TAWS was based on active radar or some other means of directly detecting an obstruction.

IIRC a classic GPWS works that way. An EGPWS is also connected to a terrain database to give earlier warnings. Please somebody correct me if I'm wrong  

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: Colombia
Posted 2012-12-18 10:07:56 and read 19806 times.

Quoting tdscanuck (Reply 3):
*If* you believe the terrain database is incorrect, then you *expect* erroneous terrain warnings. In a perverse way, continued warnings in that situation confirm to the crew that they're correct.

I think You are right here, however when flying in areas that one is not familiar with , one should climb to the MEA, MORA, MSA or whatever it is that You have at hand when in doubt, what do You guys think?

Best Regards

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: Dufo
Posted 2012-12-18 10:20:04 and read 19505 times.

This accident is a perfect classroom material, unfortunately with many lives lost.
Flying VFR in IFR in unknown airspace AND not only ignoring the warnings but disabling them completely. Amazing.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: irelayer
Posted 2012-12-18 10:29:12 and read 19287 times.

You can turn off the terrain awareness system? Why would you want to do that? Can you turn off TCAS too? It seems that those critical alert systems should never be overridable.

-IR

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: cbphoto
Posted 2012-12-18 10:33:56 and read 19202 times.

While it doesn't excuse what the pilots did, lets keep one thing in mind, they were on a demo/fun flight. In normal operations this would never have happened, and because they were on this demo flight, they wanted to give the passengers a ride and view of a lifetime and pushed the envelope of safety to do so. Unfortunately, it ended in a tragic way. Many sight seeing flights have done similar things in the past and have had a perfectly safe flight. Regardless of if the database was correct or not, the Captain should have used his best judgment and once he had entered the clouds, should have called off the fun flight and returned to a safe altitude. In the end, we may never know exactly why they continued on the path they did, but sometimes with the greatest rewards, also come the greatest risks!

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: BoeingGuy
Posted 2012-12-18 10:36:12 and read 19106 times.

Quoting irelayer (Reply 11):
You can turn off the terrain awareness system? Why would you want to do that? Can you turn off TCAS too? It seems that those critical alert systems should never be overridable.

On Boeing airplanes you can't turn off all of TAWS. You can inhibit the terrain lookahead, using the terrain database and GPS position, but you cannot turn off the radio-altimeter based GPWS function (the original GPWS without the "E" for "enhanced"). Even the reasons for turning off the terrain look-ahead portion have become mostly obsolete so I'm not really aware of any reason to do it anymore. Current EGPWS systems can handle QFE operations, and most or all airports are in the approved coordinates survey.

Quoting travelavnut (Reply 8):
Quoting redzeppelin (Reply 7):
guess that I always assumed that TAWS was based on active radar or some other means of directly detecting an obstruction.

IIRC a classic GPWS works that way. An EGPWS is also connected to a terrain database to give earlier warnings. Please somebody correct me if I'm wrong

Yes, that's basically correct. GPWS (without the "E") looks at decreasing radio altitude. In other words, the ground is rapidly rising below you. As I said, there's no way to turn this off.

Quoting nighthawk (Reply 2):
The article states that the system gave a warning of "terrain ahead, pull up" and "avoid terrain". Are these alerts given in english or Russian on the Superjet and other Russian/Ukrainian built aircraft?

Ah, they had the ACSS supplier TAWS, not Honeywell EGPWS. Only the former gives the Avoid Terrain alert. All aurals on the flight deck are in English. That's the universal language of aviation.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: RubberJungle
Posted 2012-12-18 11:07:08 and read 18467 times.

Quoting cbphoto (Reply 12):
because they were on this demo flight, they wanted to give the passengers a ride and view of a lifetime and pushed the envelope of safety to do so

There doesn't appear to be much evidence of a sightseeing element. The first demonstration flight turned back before reaching the mountains. The second flight used a different runway, and the investigators say the need to lose altitude was the reason for the early descent and orbit. The crew never intended to fly around the mountain.

http://www.flightglobal.com/news/art...d-as-fighter-investigators-380349/

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: DocLightning
Posted 2012-12-18 11:10:35 and read 18337 times.

Quoting dynamo12 (Reply 5):
Folks designing safety systems, particularly alarms, should spend a very significant amount of time avoiding false alarms. This flows from simple math. If 99.99% of the time there is no problem (1 in 10,000 incident rate), and in even 1% of the no problem time there is a false alarm you end up with 100 false alarms per real alarm. That just doesn't work.

What you have just described is called "positive predictive value." It also explains how 90% of people with a positive HIV test that is "99% accurate" will not actually have HIV.

That said, what would have been the harm in pulling up? He was in IMC flying VFR and got a warning. Frankly, it's another stain on the Russian flying profession and it was all so unnecessary and avoidable.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: rfields5421
Posted 2012-12-18 11:14:49 and read 18236 times.

Quoting Colombia (Reply 9):
one should climb to the MEA, MORA, MSA or whatever it is that You have at hand when in doubt

The flight was supposed to be conducted in an area safe from rising terrain. Well clear of any mountains. The flight was operated at an altitude well above the minimum safe altitude for the area of the demonstration.

The pilots were apparently confused by a different runway for takeoff for this second flight, which allowed them to go a few miles south of the authorized practice area.

Quoting redzeppelin (Reply 7):
How is TAWS connected with a database?

The way I understand the information from the earlier flight, when the plane would turn in certain directions, the TAWS would alert for the highest mountain in the area, even though the pilots were in a turn which would take the plane well clear of that mountain.

Indonesia, and several other island areas, would give alerts for mountains where the pilots had no intention to fly. For example, landing at HNL on Rwy 26L from the east would give TAWS alerts because the TAWS does not know the pilots are planning to turn left two miles out from the runway.

Quoting cbphoto (Reply 12):
because they were on this demo flight, they wanted to give the passengers a ride and view of a lifetime and pushed the envelope of safety to do so.

I've seen nothing to show that occurred in the earlier flight. The pilot did demonstrate the aircraft capabilities, but stayed well clear of the mountains.

We have to remember there were two mountain areas outside the approved demonstration area. The crew appears to have been aware of and careful to keep clear of the 'dangerous' higher mountain.

When they took off for the second flight, IF they followed the previously programmed flight plan and distances, it would have put them five or eight miles too far south.

That crossed out of the approved area into the second lower mountain area. There was no indication I've heard they tried to give the passengers a closeup view of the mountain. They were likely unable to see the second mountain or the ridge they hit until the last few seconds.

The attitude of the impact with the mountain side indicates they likely saw it at the last few seconds and tried to pull up to avoid the collision.

The key information goes back to basics. They allowed the aircraft to drift south of the planned demonstration area, and with limited visibility over the volcanic mountains, they were not aware how close they were to the mountain they crashed into.

Again, referring to a local source who has flown over that area many times, local pilots know to be aware of the area, to exercise extra caution. These two pilots with very little local experience did not.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: Aquila3
Posted 2012-12-18 11:22:43 and read 18070 times.

Quoting irelayer (Reply 11):
You can turn off the terrain awareness system? Why would you want to do that? Can you turn off TCAS too? It seems that those critical alert systems should never be overridable.

It does not make so much difference. If you have lost awareness, you can simply choose to ignore it, like it seems the AF447 pilots did with the stall indicator.
Very sad stories, both.

BTW, nobody did ever try to implement an active CFIT protection, something like the envelope protection?
Then, of course, you would need something to override it

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: BoeingGuy
Posted 2012-12-18 11:37:01 and read 17765 times.

Quoting rfields5421 (Reply 16):
Quoting redzeppelin (Reply 7):
How is TAWS connected with a database?

The way I understand the information from the earlier flight, when the plane would turn in certain directions, the TAWS would alert for the highest mountain in the area, even though the pilots were in a turn which would take the plane well clear of that mountain.

Not totally true. TAWS' look-ahead vector turns when the airplane is in a bank and turning. It doesn't (yet) know projected LNAV flight path, but it knows if you are, say, turning to the left and the terrain is only instantaneously in front of you but not in front of the way you are turning.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: Semaex
Posted 2012-12-18 11:52:02 and read 17391 times.

Just one question: is (E)GPWS and TWAS the same thing, just different names?

Quoting redzeppelin (Reply 7):
I have a lay man's question: How is TAWS connected with a database? I guess that I always assumed that TAWS was based on active radar or some other means of directly detecting an obstruction. But the reference to a database makes it seem that TAWS is comparing the aircraft's position and altitude to a topographic database. Is that the case?

Concerning GPWS, the system is connected to the database, which references its position through the accelerometers, GPS and rho-rho (DME-DME) or rho-beta (DEM-VOR) information. So the longer the flight is, the less accurate the position becomes and thus the less accurate the position warning will be in case of high terrain. But we're talking about a couple of dozen meters on modern airplanes, so hardly worth mentioning.
However, the radio altimeter is quite accurate at any given time. The problem with the radio altimeter though is that it only starts operating at an altitude of 2500AGL. So if you have high terrain which is rising extremely rapidly, then this margin of 2500ft may not be enough to give a timely warning (as far is I remember the picture of the crashsite in Indonesia, the terrain was rising very fast). The look-ahead function of the GPWS then again relies on the database and position update, unless you use the weather radar for surface scanning, of which I'm not sure the pilots of the SSJ did, since in IMC you ought to use it for weather detection, obviously.
More knowledgable people on the subject feel free to correct me.

Quoting Colombia (Reply 9):
I think You are right here, however when flying in areas that one is not familiar with , one should climb to the MEA, MORA, MSA or whatever it is that You have at hand when in doubt, what do You guys think?

True, but don't forget that this was a demonstration flight. I'm not sure if it was smart in the first place to do this under IMC, but money and fame played a role too I guess.

Quoting irelayer (Reply 11):
You can turn off the terrain awareness system? Why would you want to do that? Can you turn off TCAS too? It seems that those critical alert systems should never be overridable.

You can turn TCAS off by simply selecting the Transponder on a non-S mode. It would be quite annoying to do any aerodrome movement with TCAS on, since it would alarm every single second due to the close proximity of other aircraft in that area.
You can turn GPWS off too, as some of the seven submodes are not necessary at all times (for example the "flaps flaps" warning when you are doing flap-up landings for training purposes).

Quoting Aquila3 (Reply 17):
BTW, nobody did ever try to implement an active CFIT protection, something like the envelope protection?
Then, of course, you would need something to override it

While giving aircraft a CFIT protection would reduce the global amount of fatal accidents by something like 50%, the problem with it is in itself. Think about it: after all, every landing is nothing more than another "controlled flight into terrain", with gear down. Having the aircraft go around by itself everytime it gets near to the ground is rather pointless. It would need such an amount of override for all kinds of situations that probably financially and logically it doesn't make sense to install such a system.
And quite frankly, the GPWS and TCAS do an incredibly good job when handled correctly and responed to actively.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: RickNRoll
Posted 2012-12-18 12:02:04 and read 17165 times.

Quoting tdscanuck (Reply 3):
They thought it was a database problem, which means they'd lost situational awareness. Once that happens, confirmation bias becomes a huge issue...the crew believes they know what's going on (even though they don't) so they fit any new information into their theory of what's going on rather than revising the theory.

*If* you believe the terrain database is incorrect, then you *expect* erroneous terrain warnings. In a perverse way, continued warnings in that situation confirm to the crew that they're correct.

Tom
Quoting Aquila3 (Reply 17):
It does not make so much difference. If you have lost awareness, you can simply choose to ignore it, like it seems the AF447 pilots did with the stall indicator.
Very sad stories, both.

What Tom said instantly reminded me of AF447 as well. Two crashes with experienced crew with planes that performed exactly as designed, that in the same situation most pilots would have avoided the crash.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: sturmovik
Posted 2012-12-18 12:07:49 and read 17075 times.

Quoting nighthawk (Reply 2):
Are these alerts given in english or Russian on the Superjet and other Russian/Ukrainian built aircraft?

The Superjet has a lot of western systems on board, and the warnings were almost certainly in English. A few Russian and Chinese aircraft do have systems that are in the local language, both in terms of what is displayed on screens and what is spoken. However, I doubt this is the case with the SSJ, which has been built with the intention of selling it in markets outside of what is traditional for Russian manufacturers. I may be wrong on this.

Quoting tdscanuck (Reply 3):
confirmation bias

  
I read in one of the news articles that there was a potential buyer in the cockpit who was conversing with the pilots for a good portion of the flight. This may have aided the unfortunate situation, distracting the pilots from reacting correctly to cues they may have otherwise not ignored.

Quoting irelayer (Reply 11):
Can you turn off TCAS too?

It can definitely go on standby mode. Switching off the transponder (or a malfunction, perhaps) can cause TCAS to not function, which was one of the contributory causes to the GOL 1907/ Embraer Legacy collision and subsequent crash of the 737. The Legacy jet had its transponder inadvertently switched off, thereby causing a TCAS outage the crew weren't aware of.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: glideslope
Posted 2012-12-18 12:11:47 and read 16982 times.

Situations such as this will never be completely avoided as long as Humans are part of the equation.

  

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: BoeingGuy
Posted 2012-12-18 12:24:49 and read 16697 times.

Quoting Semaex (Reply 19):
Just one question: is (E)GPWS and TWAS the same thing, just different names?

Yes, TAWS is a generic industry term (Terrain Awareness and Warning System). GPWS is an industry generic term, even though Honeywell first developed that system. EGPWS (for "Enhanced") is a Honeywell trademarked term.

Quoting sturmovik (Reply 21):
It can definitely go on standby mode. Switching off the transponder (or a malfunction, perhaps) can cause TCAS to not function, which was one of the contributory causes to the GOL 1907/ Embraer Legacy collision and subsequent crash of the 737. The Legacy jet had its transponder inadvertently switched off, thereby causing a TCAS outage the crew weren't aware of.

That is only half of the story. The Legacy inadvertently turning their Transponder to Standby turned off their own TCAS. However, it caused a bigger issue than that. It also made the Legacy invisible to the 737's TCAS, which was on and in perfect working order. The 737 was doing everything right - TCAS on and in the right place. The 737's TCAS couldn't detect the Legacy because the latter's Transponder was off.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: mandala499
Posted 2012-12-18 12:25:33 and read 16698 times.

Quoting rfields5421 (Reply 4):
The only person on this forum who has flown over that area and has detailed knowledge of the area, the terrain and where the flight was supposed to be is mandala499.

OK... I'm here...   

Quoting rfields5421 (Reply 4):
In his detailed posts after the crash, he explained how the pilots had drifted out of the area where the flight was supposed to be taking place.

I did several simulations to see where the aircraft would end up in zero wind condition and headed out of the right-hand orbit on heading 200... whilst descending from 10,000 to 6,000 at a rate of -1000ft per minute... it ended up a mere 500m from the crash site, same altitude, and about 100ft spare. If they were not distracted in completing their orbit, they wouldn't be facing a wall of terrain but would still be damn close.

Quoting rfields5421 (Reply 4):
Also that they were probably fixated on avoiding a higher 'more dangerous' nearby mountain, not realizing they were below the level of the mountain/ ridge they hit.

Yes, the MORA reflected the bigger mountain (up to 9,900ft high) to the east of the one they hit. This fixation was aided by the perceptions gained from the previous flight, and also the lack of terrain information they had in the area briefing they received... basically, they didn't know there was another mountain at 6000ft nearby... and it was covered by cloud.

Quoting cbphoto (Reply 12):
In normal operations this would never have happened, and because they were on this demo flight, they wanted to give the passengers a ride and view of a lifetime and pushed the envelope of safety to do so.

The CVR nor the FDR revealed any information that indicated intentional "wanting to give the passengers a ride and view of a lifetime and pushed the envelopeof safety to do so". There was no view of a lifetime on that fateful flight. The nice views they had on the first flight, was covered in thick cloud.

Quoting rfields5421 (Reply 16):
The flight was supposed to be conducted in an area safe from rising terrain. Well clear of any mountains. The flight was operated at an altitude well above the minimum safe altitude for the area of the demonstration.

The pilots were apparently confused by a different runway for takeoff for this second flight, which allowed them to go a few miles south of the authorized practice area.

The change of runway, ATC handling them seeing the aircraft type on the radar tag as an Su-30, and lack of terrain information in their area briefing, and purely using approach and enroute charts... sealed their fates.

Quoting rfields5421 (Reply 16):
Again, referring to a local source who has flown over that area many times, local pilots know to be aware of the area, to exercise extra caution. These two pilots with very little local experience did not.

Not quite. Airline pilots don't know... they don't fly to that area. Student pilots out of Halim and Air Force pilots operating out of Halim and nearby Atang Sanjaya generally avoid the area unless in clear VFR. Few, ventured there in marginal VFR or IFR... I did, and even former officials from the CAA of a "safe country" laughed at our preparations and contingencies... at least I'm alive because of that. But many local pilots don't know that area. They just know 'not to venture south of the training area below MSA without knowing where things are. Given that flight path, 9/10 pilots here I asked said they don't even know that mountain was there... they just know the bigger one making that high MORA.

Mandala499

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: JoeCanuck
Posted 2012-12-18 12:57:14 and read 16439 times.

Quoting mandala499 (Reply 24):

Nice to have a first hand perspective....thanks.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: DocLightning
Posted 2012-12-18 13:04:13 and read 16292 times.

Quoting Semaex (Reply 19):
While giving aircraft a CFIT protection would reduce the global amount of fatal accidents by something like 50%, the problem with it is in itself. Think about it: after all, every landing is nothing more than another "controlled flight into terrain", with gear down. Having the aircraft go around by itself everytime it gets near to the ground is rather pointless. It would need such an amount of override for all kinds of situations that probably financially and logically it doesn't make sense to install such a system.

Well, you could arrange the logic architecture to deactivate the protection if the gear is down and the aircraft is otherwise configured for landing.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: BoeingGuy
Posted 2012-12-18 13:21:47 and read 16128 times.

Quoting DocLightning (Reply 26):
Quoting Semaex (Reply 19):
While giving aircraft a CFIT protection would reduce the global amount of fatal accidents by something like 50%, the problem with it is in itself. Think about it: after all, every landing is nothing more than another "controlled flight into terrain", with gear down. Having the aircraft go around by itself everytime it gets near to the ground is rather pointless. It would need such an amount of override for all kinds of situations that probably financially and logically it doesn't make sense to install such a system.

Well, you could arrange the logic architecture to deactivate the protection if the gear is down and the aircraft is otherwise configured for landing.

There's another problem with that description. The TAWS databases know where the airports are. So it inhibits the alerts when you are landing at an actual airport if configured correctly. There are alerts for trying to land and not having the gear or flaps down (there are overrides for this if intentionally doing a flaps up landing for example). However, TAWS knows if you are about to "land" at it's not at an airport, even if in the correct landing configuration.

This is one of the many advantages of EGPWS. GPWS didn't know if you were really at an airport or about to land on a shopping mall. It just knew if you were configured correctly. As long as you were in the correct landing configuration, GPWS didn't give an alert. EGPWS knows if you are actually going to land at an airport.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: awthompson
Posted 2012-12-18 13:26:25 and read 15976 times.

Quoting glideslope (Reply 22):
Situations such as this will never be completely avoided as long as Humans are part of the equation.

We can try but I think air safety is largely as good now as it's likely to get in the west. In Russia and certain other countries, they are not quite as far up the learning curve and have a little way to go yet.

Whilst I love flying in Russia and on the Russian airlines, and have done so in the past month, safety there although improving is not quite up to western standards. I see this an attitude/cultural issue which may slowly improve - unfortunately perhaps as the result of further accidents like the one we are discussing. On recent flights I've taken with the Russian flag carrier, passengers were on their feet before the aircraft came to a standstill, appeals by cabin crew were completely ignored. Also on one flight (see my discussion on Tech/Ops forum), a significant frost layer covering most of the upper wing surface was ignored on departure. Neither of these situations were on their own terribly serious but demonstrated to me a slightly different attitude to safety. In contrast flying last week with a UK airline, upper wing surface was de-iced and flight delayed when first officer saw a tiny amount of frost that was almost invisible.

One positive of the human factor outcome in this accident is that the Superjet itself has come out in the clear.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: DocLightning
Posted 2012-12-18 13:33:18 and read 15876 times.

Quoting awthompson (Reply 28):

One positive of the human factor outcome in this accident is that the Superjet itself has come out in the clear.

Very true.    No aircraft, no matter how safe, is immune to CFIT.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: Semaex
Posted 2012-12-18 14:27:19 and read 14913 times.

Quoting DocLightning (Reply 26):
Well, you could arrange the logic architecture to deactivate the protection if the gear is down and the aircraft is otherwise configured for landing.
Quoting BoeingGuy (Reply 27):
This is one of the many advantages of EGPWS. GPWS didn't know if you were really at an airport or about to land on a shopping mall. It just knew if you were configured correctly. As long as you were in the correct landing configuration, GPWS didn't give an alert. EGPWS knows if you are actually going to land at an airport.

Concerning an "anti-CFIT" program, this situation then however becomes complicated again when the aircraft is out of the normal values for the position of the actual airport. What I mean is: would the system give you a warning if you are an eighth mile away from touchdown, on what you think is on localizer course but way below the NPA glide? The system might think, since you're "close to the airport", that all is well and it will inhibit a warning, which will lull the pilots to think that everything is in limits, even if the GPWS screams for a pull up.

Keep in mind that technology is always just as good as the ones operating it. An increase in the amount of safety features is good, but if it turns out to make pilots think that another kind of safety device must be malfuctioning, then all is pointless.

Take the SSJ accident as an example. The pilots were cleared by a controller (a safety feature) to do something they failed to do eventually, not realizing. Another safety feature (TAWS) told them they have a problem, yet they trusted the instructions by the controller so much that they assumed a system failure of the latter.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: tdscanuck
Posted 2012-12-18 14:43:00 and read 14648 times.

Quoting dynamo12 (Reply 5):
A question I have for pilots is how often are there actual "database problems" that cause false alarms.

It's pretty rare. The more common cause of false alarms is algorithm artifacts, because the system doesn't react that well to very tall very thin obstacles like large antennas.

Quoting dynamo12 (Reply 5):
Folks designing safety systems, particularly alarms, should spend a very significant amount of time avoiding false alarms.

They do.

Quoting Colombia (Reply 9):
I think You are right here, however when flying in areas that one is not familiar with , one should climb to the MEA, MORA, MSA or whatever it is that You have at hand when in doubt, what do You guys think?

Yes, you should. The problem, in this situation, is that the crew thought they were in an area they were adequately familiar with...they had ATC clearance to operate safety and they thought they knew where they were (in which case the clearance was correct and they were safe). The problem was that they weren't where they thought they were. This is almost always true for CFITs.

Quoting irelayer (Reply 11):
You can turn off the terrain awareness system? Why would you want to do that?

You can't actually turn the system off. You can turn off certain functions...for example, if you have an abnormal landing configuration (e.g. Flaps 20 on a Boeing widebody) the checklist will have you turn off some of the warnings to prevent known false alarms during approach.

Quoting irelayer (Reply 11):
Can you turn off TCAS too?

Yes. If you don't, you go nuts at a busy airport. TCAS would have you doing all kinds of bizarre things on climb-out and screaming like a banshee on the ramp.

Quoting irelayer (Reply 11):
It seems that those critical alert systems should never be overridable.

They have to be overridden when operations go outside the normal parameters for those systems (which happens all the time for TCAS, rarely but often enough for EGPWS).

Quoting cbphoto (Reply 12):
While it doesn't excuse what the pilots did, lets keep one thing in mind, they were on a demo/fun flight. In normal operations this would never have happened, and because they were on this demo flight, they wanted to give the passengers a ride and view of a lifetime and pushed the envelope of safety to do so.

It's true that a demo flight (not a "fun flight"...real CAA's don't authorize "fun flights" for customers) operates by different rules, but it is NEVER acceptable to push the safety envelope to do so. It would, in fact, be a direct violation of the FAR's or equivalent of every regulator I'm aware of.

Quoting Aquila3 (Reply 17):
BTW, nobody did ever try to implement an active CFIT protection, something like the envelope protection?

The USAF did. It's called Auto-GCAS. http://www.f-16.net/news_article4658.html
Still in testing right now, as far as I know, but very promising.

Quoting Semaex (Reply 19):
Concerning GPWS, the system is connected to the database, which references its position through the accelerometers, GPS and rho-rho (DME-DME) or rho-beta (DEM-VOR) information. So the longer the flight is, the less accurate the position becomes and thus the less accurate the position warning will be in case of high terrain.

Only the inertia (accelerometer) position drifts with time. The GPS, DME-DME, and DME-VOR solutions are time-independent. Most modern INS's will zero-out drift against GPS and come can even re-align (remove drift) in flight. You only get major drift in the accuracy of the position if you're away from VOR's *and* you don't have GPS.

Tom.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: flyinTLow
Posted 2012-12-18 19:32:37 and read 10418 times.

On any aircraft there have to be ways to turn these systems off in case of certain failures which, in case of that failure, could lead to nuisance warnings.

The BASIC GPWS (not the enhanced part of it, as pointed out before), has 7 modes (and from what I was tought in flight school, they are not different with different aircraft types when they have a GPWS System installed), which are among other things, mostly based on radar altitude together with a few other systems:

Mode 1: excessive rate of descent ("sinkrate")
compares barometric VS with radar altitude, gives warning of high vertical speeds below 2500 fpm

Mode 2: excessive terrain closure rate
compares change of barometric altitude with current barometric altitude

Mode 3: altitude loss after takeoff or go-around ("dont't sink")
gives warning if there is a negative VS after T/O or G/A

Mode 4: unsafe terrain clearance and not in landing config ("too low, flaps/gear")
gives warning in case of low radar altitude and Flaps (Mode 4a) or Gear (Mode 4b) are not in landing configuration

Mode 5: deviation below glideslope ("glideslope")
gives warning in case of low radar altitude and more than 1 dot below glide

Mode 6: advisories
Also known as radar altitude call outs

Mode 7: windshear detection
compares air speed with GS

As you can tell, there are multiple systems that work with the system, all which can cause nuisance warnings. This is why, under certain circumstances (for example GS unreliable and performing a LOC Approach, flap failures which would lead to a landing not in normal landing configuration, unreliable air data, etc.) there has to be an option to switch off certain modes of the basic GPWS System in order to not cause nuisance warnings and cause an even higher workload in stressful situations.

As pointed out by the colleages before, the ENHANCED part of the GPWS is a terrain look-ahead system which compares barometric altitude and current aircraft position (based on GPS or IRS or GPIRS, long story there   ) with a worlwide terrain database. This database contains all terrain and most of the time also man made obstacles. This databse has "bowls" around all programed runways where this terrain database is "lowered" in closer proximity of these runways in order to allow for approaches to these runways without EGPWS warnings ("Terrain Ahead").

As you can tell at this point already: if the aircraft position is not absolutly reliable, nuisance warnings can happen. This is why the ENHANCED part of the GPWS must have an inhibit option as well.

The problem with this database is, that it has to be kept up to date. The terrain changes less often, of course. But on the one hand, short-haul aircraft flying in Europe only dont't need the terrain database of southern America, as they are most likely never going to be flying there. Also, new built runways have to be incorporated into the database as well. Now if for some reason you operate an aircraft in areas without terrain database coverage, dispatch will let you know about this, or, as a pilot, you should make yourself aware of this. In that case, the ENHANCED part may be switched off. But this is no excuse to still obey to the other, basic warnings.

On the A320 aircraft, on the overhead panel right above the CM1 seat, you can turn off the entire GPWS System (which then also turns off the ENHANCED part), only the enhanced part (TERR Pushbutton), the glideslope warning (e.g. for LOC Approaches or Circling Approaches, G/S MODE P/B), and the flap warning (for abnormal flap config approaches (FLAP MODE P/B).


I hope that gives a good overview of the 2-part-(E)GPWS System.


P.S.: On the A320, the TCAS can also be switched off or into TA Mode only. The TA mode only for example, is necessary when you have certain performance problems (single engine, flying with gear down other than for approach, etc.) which would not allow you to follow TCAS escape maneuvers. In this case, should a conflict arrise, the TA Mode only TCAS will communicate to the other system that no escape maneuver can be flown, and the other system will then command a more rapid escape)

[edit for TCAS]

[Edited 2012-12-18 19:38:41]

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: BoeingGuy
Posted 2012-12-18 20:01:40 and read 10031 times.

Quoting flyinTLow (Reply 32):
As pointed out by the colleages before, the ENHANCED part of the GPWS is a terrain look-ahead system which compares barometric altitude and current aircraft position (based on GPS or IRS or GPIRS, long story there ) with a worlwide terrain database.

Actually the newer EGPWS's for about the past 10 years do what's called "Geometric Altitude" not just Barometric altitude. It fuses Baro Alt, Radio Alt and GPS alt to get the most accurate altitude. That way, say, if the pilot put in the wrong barometric setting, it would throw out the bad input and still use an accurate altitude. That's also why the current EGPWS can support QFE operations, while there was an AFM limitation to Inhibit the terrain look-ahead feature (the "E" in EGPWS) when doing QFE.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: flyinTLow
Posted 2012-12-18 20:26:56 and read 9676 times.

Quoting BoeingGuy (Reply 33):

absolutly correct. I just used the term barometric to differentiate it from radio.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: sturmovik
Posted 2012-12-18 22:31:14 and read 8392 times.

Quoting BoeingGuy (Reply 23):
That is only half of the story. The Legacy inadvertently turning their Transponder to Standby turned off their own TCAS. However, it caused a bigger issue than that. It also made the Legacy invisible to the 737's TCAS, which was on and in perfect working order. The 737 was doing everything right - TCAS on and in the right place. The 737's TCAS couldn't detect the Legacy because the latter's Transponder was off.

You are right, of course. I thought I could get away with the phrase 'contributory cause' since this discussion is not about that accident (and that one can open a whole other can of worms).

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: Mir
Posted 2012-12-18 23:01:13 and read 8092 times.

Quoting irelayer (Reply 11):
You can turn off the terrain awareness system? Why would you want to do that? Can you turn off TCAS too? It seems that those critical alert systems should never be overridable.

If they malfunction, then they need to be turned off in order to stop providing false alarms.

Quoting DocLightning (Reply 26):
Well, you could arrange the logic architecture to deactivate the protection if the gear is down and the aircraft is otherwise configured for landing.

You'd also have to provide for the case of a gear-up landing, which would mean a means of disabling the system available to the crew.

-Mir

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: keegd76
Posted 2012-12-19 01:03:41 and read 6950 times.

Quoting mandala499 (Reply 24):
The change of runway, ATC handling them seeing the aircraft type on the radar tag as an Su-30, and lack of terrain information in their area briefing, and purely using approach and enroute charts... sealed their fates.

Didn't know about the radar tag issue. What would cause a civilian jet to appear on radar as a fighter jet? I thought the radar tag was based on information from the aircraft transponder. If so does this mean the transponder was configured incorrectly or was ATC unable to interrogate it properly?

Just curious   

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: Mir
Posted 2012-12-19 01:17:21 and read 6833 times.

Quoting keegd76 (Reply 37):
What would cause a civilian jet to appear on radar as a fighter jet?

An incorrect input into the ATC system, either from an error when filing the flight plan, the system not recognizing the code and assigning what it felt to be the closest known type, or a controller typing in the wrong code (which wouldn't be uncommon when dealing with a new and unfamiliar type).

-Mir

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: RickNRoll
Posted 2012-12-19 01:29:34 and read 6704 times.

Quoting mandala499 (Reply 24):
The change of runway, ATC handling them seeing the aircraft type on the radar tag as an Su-30, and lack of terrain information in their area briefing, and purely using approach and enroute charts... sealed their fates.

That and turning off the TAWS.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: rfields5421
Posted 2012-12-19 05:23:12 and read 5797 times.

Quoting keegd76 (Reply 37):
What would cause a civilian jet to appear on radar as a fighter jet?

Remember this was a new type aircraft never seen in this country before. Su-30 might have been the newest Sukhoi aircraft in the system aircraft type database.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: RubberJungle
Posted 2012-12-19 06:46:09 and read 5558 times.

Quoting keegd76 (Reply 37):
What would cause a civilian jet to appear on radar as a fighter jet?

There was no reference in the flight-planning database for the Sukhoi Superjet, so the flight data officer at Jakarta selected the Su-30 to improvise. When the controller checked the aircraft type on his radar, it simply told him what the data officer had entered as the aircraft type.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: BoeingGuy
Posted 2012-12-19 07:14:29 and read 5482 times.

Quoting Mir (Reply 36):
Quoting irelayer (Reply 11):
You can turn off the terrain awareness system? Why would you want to do that? Can you turn off TCAS too? It seems that those critical alert systems should never be overridable.

If they malfunction, then they need to be turned off in order to stop providing false alarms.

Again, you can't totally turn off TAWS, only the look ahead function. You can't turn off the radio altimeter based alerts. You can inhibit the gear and flap configuration alerts too for non-normal situations that require you to land with flaps-up, for example.

I know of no procedure to totally turn off TCAS in flight. As another poster indicated, there are a few procedures that require you to go to TA ONLY in flight. One example is having an engine shutdown on a twin (because you can't necessarily perform a CLIMB or INCREASE CLIMB Resolution Advisory, should you get one) or something like doing a visual approach to the runway 28s at SFO parallel to another airplane.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: txjim
Posted 2012-12-19 07:32:52 and read 5434 times.

Quoting tdscanuck (Reply 31):
Quoting irelayer (Reply 11):
Can you turn off TCAS too?

Yes. If you don't, you go nuts at a busy airport. TCAS would have you doing all kinds of bizarre things on climb-out and screaming like a banshee on the ramp.

It's also turned off when approaching a tanker

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: BoeingGuy
Posted 2012-12-19 07:50:19 and read 5380 times.

Quoting txjim (Reply 43):
Quoting tdscanuck (Reply 31):
Quoting irelayer (Reply 11):
Can you turn off TCAS too?

Yes. If you don't, you go nuts at a busy airport. TCAS would have you doing all kinds of bizarre things on climb-out and screaming like a banshee on the ramp.

It's also turned off when approaching a tanker

Actually you are probably turning off the Transponder in each case. That does, by default, turn off TCAS (there is also a TCAS OFF mode, which leaves the Transponder on). With your Transponder off, other airplanes like the Tanker won't get TCAS RA's from you either.

TCAS wouldn't be screaming on the ramp. It has some automatic mode inhibits. Even if it's on while on the ground, you only could get TA's and only the visual indications. The aural "TRAFFIC, TRAFFIC" would be automatically inhibited on the ground. Other inhibits are like one that inhibit DESCEND and INCREASE DESCENT commands when below around 1200 feet, for obvious reasons.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: rfields5421
Posted 2012-12-19 08:35:26 and read 5251 times.

One thing to remember about turning off the TAWS is that this aircraft was near the end of the flight and the pilots were descending for their approach. Just ---


Quoting tdscanuck (Reply 31):
The problem, in this situation, is that the crew thought they were in an area they were adequately familiar with...they had ATC clearance to operate safety and they thought they knew where they were (in which case the clearance was correct and they were safe). The problem was that they weren't where they thought they were. This is almost always true for CFITs.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: mandala499
Posted 2012-12-19 10:35:06 and read 5015 times.

Quoting RickNRoll (Reply 39):
That and turning off the TAWS.

If they didn't get distracted, no TAWS (with basic GPWS still on) wouldn't result in the accident.
They switched the TAWS off after getting the warning because they think it was erroneous.
The question that must be asked, is factors contributed to their decision to switch it off after it sounded the warning.
It's the same as in AF447.. why did the PF ignore over 50 seconds of stall warning? That case, it wasn't switched off, but ignored.
So, if the TAWS wasn't switched off after the initial warning, would they have believed it?
Inadequate terrain information in the pre-flight briefing was a factor mentioned in the report.

Quoting rfields5421 (Reply 40):
Remember this was a new type aircraft never seen in this country before. Su-30 might have been the newest Sukhoi aircraft in the system aircraft type database.

I'll add further to this...
Until this accident happened, an overwhelming majority in the industry here, didn't know Sukhoi made a jetliner... and a higher proportion ATC people not knowing that at that time won't be a surprise to me.

And, almost all of the Sukhois flying around Jakarta are Su-30s (and fewer Su-27s), and since 2nd half of 2011, the Air Force's Sukhois visit Jakarta Halim Airportevery few weeks/months. So, this adds to the assumption factor. This flight being organized by the local Sukhoi distributor (who sold the Sukhoi military jets), and added military presence surrounding the flight, again, adding to the factors leading to that assumption.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: Aquila3
Posted 2012-12-19 10:36:25 and read 4993 times.

Quoting tdscanuck (Reply 31):
Quoting Aquila3 (Reply 17):
BTW, nobody did ever try to implement an active CFIT protection, something like the envelope protection?

The USAF did. It's called Auto-GCAS. http://www.f-16.net/news_article4658.html
Still in testing right now, as far as I know, but very promising.

Thanks, Tom. That sounds like the right way to me. Glad that USAF invests conspicuous budget in this research, involving the top notch of US technology suppliers. Of course it is much easier to make a terrain avoidance maneuver with a F16 (see THRUST) but in a few years we might have it implemented in commercial airliners, in analogy with what happened with FBW. I am quite sure that many lives could be saved this way.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: Gatorman96
Posted 2012-12-19 11:02:41 and read 4918 times.

There is a good chance I missed this fact somewhere along the line and I am in NO way placing any blame as this and all accidents are very complex and have many layers. Wouldn't ATC have known that the jet was off course and if so, they would've certainly notified the pilots. No?

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: rfields5421
Posted 2012-12-19 11:47:56 and read 4821 times.

Quoting Gatorman96 (Reply 48):
Wouldn't ATC have known that the jet was off course and if so, they would've certainly notified the pilots. No?

Mandala499 provided some background on this back right after the crash.

I'm writing this from memory.

There were a couple factors which limited the ATC knowing about the aircraft being off course.

One is the time of day of the flight was a very busy time of day for the controllers.

Another is the aircraft was to fly to a certain cleared area and maneuver at pilot's discretion. There wasn't a set 'course' for the demo flight to follow - just to operate in a general area.

So the controllers were not monitoring every turn, etc.

Also -

Quoting mandala499 (Reply 24):
Student pilots out of Halim and Air Force pilots operating out of Halim and nearby Atang Sanjaya generally avoid the area unless in clear VFR. Few, ventured there in marginal VFR or IFR... I did, and even former officials from the CAA of a "safe country" laughed at our preparations and contingencies... at least I'm alive because of that. But many local pilots don't know that area. They just know 'not to venture south of the training area below MSA without knowing where things are. Given that flight path, 9/10 pilots here I asked said they don't even know that mountain was there... they just know the bigger one making that high MORA.

There is no reason to believe ATC would have had more idea/ warning about this area than the pilots who fly in the area.

A key factor in my opinion is that the pilots fixated on avoiding the 'dangerous' mountain and crashed into a ridge on a much lower mountain.

As Tom said above - they thought there were in one location when they really were a few miles south.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: mandala499
Posted 2012-12-20 09:54:11 and read 4259 times.

I'll add to Rfields5421 as usual..

Quoting Gatorman96 (Reply 48):
Wouldn't ATC have known that the jet was off course and if so, they would've certainly notified the pilots. No?

The final report stated the controller on duty was over burdened. He had to control his sector with 14-20 aircraft at a time passing through, without an assistant, and also doing supervisor duty. He didn't realize the aircraft was missing for 20 mins...
And add to the fact that the radar tag was displaying the aircraft as an Su-30, he wasn't overly concerned... until 20mins.

Quoting rfields5421 (Reply 49):
Another is the aircraft was to fly to a certain cleared area and maneuver at pilot's discretion. There wasn't a set 'course' for the demo flight to follow - just to operate in a general area.

This was one change from the previous 'assumptions'... the flight was filed to fly along HLM R-200, but no distance along that radial was set... and they arbitrarily chose DME20 in the first flight prior to turning back on vectors... DME20, was above and right in the middle of the Atang/Bogor training area... add that with Su-30 radar tag... yeah... we now know the outcome.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: 747WanSui
Posted 2012-12-20 19:33:54 and read 3839 times.

I hope Mayday can do an episode on this crash at some point - I certainly think the details of this crash would make for a good episode for that series.

Topic: RE: Superjet Pilot Switched Off Taws Before Crash
Username: JoeCanuck
Posted 2012-12-20 21:08:45 and read 3706 times.

Quoting 747WanSui (Reply 51):

Absolutely...I think there's always something to learn from the tragic, 'human error(s)' accidents...even if it's; 'there, but for the grace of god, go I'.


The messages in this discussion express the views of the author of the message, not necessarily the views of Airliners.net or any entity associated with Airliners.net.

Copyright © Lundgren Aerospace. All rights reserved.
http://www.airliners.net/