PDA

View Full Version : Pilot was asleep before crash: Black box data


jimmygill
9th Sep 2010, 02:49
Pilot was asleep before crash: Black box data (http://www.deccanherald.com/content/95131/pilot-asleep-crash-black-box.html)

Pilot was asleep before crash: Black box data
Ajith Athrady, New Delhi, Sep 8, DHNS:

Nearly four months after an Air India Express Boeing aircraft crashed during landing at Mangalore airport killing 158 people on board, flight details contained in the black box have revealed that the pilot-in-command was asleep for an hour and 50 minutes when the plane cruised towards its destination.

An analysis of the cockpit voice recorder (CVR)— which contains conversation between two pilots, radio communication between the cockpit crew and others (including conversation with air traffic control personnel) and Digital Flight Data Recorder (DFDR)— was presented before the Court of Inquiry (CoI) headed by Air Marshal (Retd) B N Gokhale here on Wednesday.

The Directorate-General of Civil Aviation (DGCA), which presented the details to the CoI, had sent the CVR and the DFDR to the United States for data retrieval since they were heavily damaged, though the memory chips remained intact. In the CoI, officials from Boeing, the manufacturer of the aircraft, and GE, which was involved in decoding the black box, presented their findings.

Only excerpts of the black box data were released on Wednesday and a detailed report will be submitted to the Centre along with the CoI findings by the end of this month.
The startling details culled from the CVR and the DFDR have now established that pilot error, compounded by others factors like sleeping, caused the crash.

Fingerpointing

The cockpit transcripts and coded DFDR information were released to silence persistent speculation that reasons other than pilot error might have contributed to the doom of the aircraft and the 152 passengers and six crew members.

The analysis of the CVR revealed that there was sound nasal snoring and heavy breathing for nearly two hours, indicating that one of the two pilots—in all likelihood Serbian national Captain Zslatko Glucika, 55—had fallen asleep on his seat on the 200-minute flight of IX-812 that originated in Dubai (local time 1:10 am) on May 22.
What corroborates pilot error as the principal reason behind the crash is the long silence in the cockpit—for as long as 110 minutes—indicating that Glucika, who was believed to have about 2,000 flying hours’ experience, had dozed off ahead of landing. Captain H S Ahluwalia, 40, from Mumbai, was the co-pilot of the ill-fated aircraft.

When it was time for flight IX-812 to land, it was too late. The black box details have established that the aircraft started descent for landing when it was flying at an altitude of 4,400 feet against the normal height of 2,000 feet. Besides, the flight touched down at the 4,638 feet mark on the middle of Mangalore’s table-top runway which has a maximum length of 8,038 feet. According to civil aviation regulations, normal touch-down mark for passenger aircraft is 1,000 feet.

This strongly indicates that Glucika not only reacted late and overshot the runway, but certain standard operating procedures were not followed during landing.

Deccan Herald reported on August 19 the pilots’ last-minute conversation in which Capt Ahluwalia desperately cautioned Glucika to abort landing and “go around”, meaning that the pilot should not attempt a landing and instead try to fly without touching down. The co-pilot is heard on the audio telling Glucika “we don’t have enought runway left”. Capt Ahluwalia’s last word to his commanding pilot was “control” and then the aircraft fell on to a steep slope before exploding into a ball of fire.

According to the decoded information from the black box, the aircraft was not on its glide path while landing. As suspected earlier, the Boeing 737-800 aircraft was flying at an unsually high speed of 139 knots during landing. In the course of the inquiry and inspection of the aircraft’s remains it was found that the plane’s landing gear was found in the takeoff position, suggesting that the pilot tried a “go around.”

“Despite the high speed and landing in the middle of the runway, had the pilot tried to stop the aircraft instead of taking off after making touch down, it would have stopped at least at the end of the run way averting the disaster,” a Boeing official said. In support of the evidence that the pilot tried a “go around,” the DFDR shows that Glucika activated the takeoff gear and that the engine was in powered to high speed. “During normal touch down the engine speed is always low”, the official said.

“The main reason for the accident was that the pilot(s) tried to take off when just 800 feet of the runway was left. It was a wrong judgment while attempting a takeoff,” he said.

The decoding of the black box also shows that the GPS (ground proximity warning system) precision approach and landing system of the aircraft had given several warnings indicating that it was taking a wrong glide path. The CVR records show that both pilots had discussed that the aircraft was following a wrong glide path.

Early on May 22, the weather was fine for landing and visibility 6 km. The runway was dry and clearance had been given for landing under the Instrument Landing System. According to Boeing, the runway length for a 737-800 type aircraft during landing should be 7,500 feet.

Excerpts of pilots’ conversation

As soon as Pilot Captain Z Glucika got clearance of the ATC, he started descending for landing.
Co-Pilot: (Captain H S Ahluwalia): O My God.
Both pilots: Flight is taking wrong path and wrong side.
Co-pilot: Go around
Though flight is still descending and it tried to touch down in middle of the run way, the co-pilot insists on ‘go around’.
Co-pilot: Pull up. (Repeats six times)
Captain: Only 800 metres left. (in runway) (It was suspected that this time may be the aircraft was trying to take off and pilot found that only 800 feet of run way left)
(These conversation took place within eight seconds)

formerijmpilot
9th Sep 2010, 06:11
Conclusions jumped to perhaps too early.

CVR recording showing snoring noises 110 mins at least, before the event.

Any evidence produced to show FTL compliance.

The free exchange between feet and metres is also interesting.

IJM
(felix would be proud)

masalama
9th Sep 2010, 06:22
As usual, our reporters love to give error-filled reports , unless we read the actual transcripts with the DGCA findings, a lot of questions will remain.Some of the errors in my opinion:

1.Glucika, who was believed to have about 2,000 flying hours’ experience
Wow, 2000 hrs and he's hired as a PIC with AIX....maybe they mean 2000 hrs on type.

2.As suspected earlier, the Boeing 737-800 aircraft was flying at an unsually high speed of 139 knots during landing.

OK, any CPL holder will be able to tell you that 139 knots for a 737-800 is not unusual at all.So ineffective and useless are our so-called aviation experts in these newspapers that they'll go any lengths to publish an aviation story riddled with holes ..

3.The decoding of the black box also shows that the GPS (ground proximity warning system) precision approach and landing system of the aircraft

Not only is GPS (Global positioning system) an entirely different piece of equipment but since when did the GPWS become a precision approach and landing system of the aircraft?????

This report is perfect for the non-aviation flying public, I know what they'll be saying , oh look those overpaid pilots sleep in the cockpit and the autopilot does everything and pilots are the reason for every crash. The complex relationship between fatigue, decision making , circardian rhythm , CRM , experience gradient will never get any public attention nor will it be addressed by the regulator.
And as long as we have great reporters like these who are there to publish error prone official reports without any research/investigation of their own , the real issues will not get the attention they deserve.

God help us, masalama.:ok:

jimmygill
9th Sep 2010, 06:42
the GPS, ground proximity warning system, precision approach and landing system of the aircraft had given several warnings

Just replace the '(' and ')' with ','

IndAir967
9th Sep 2010, 07:15
Pilots are human and their body needs a break ..

Yes a lot of pilots take quick naps or power naps on a lot of flights esp
on the ones more than 2hrs cruise time in the middle of the night.
As long as the other person is informed of the nap and he accepts to handle
both aircraft and communication, its fine.

As per the report, I dont see an issue with Sleeping or Approach Speed.

What I dont understand is on an aircraft like the B738 which is equipped with
VNAV and when the crew can constantly check the vertical profile how come
they ended up high ? (if they did as per the above report).
Especially when the guy on the right seat was also pretty experienced on the
aircraft and assuming he commenced decent when the skipper dozed off.

I know AIX has an issue with FDTLs. What about in this particular incident.
If the pilots were above duty then the blame gets vectored onto an different
heading.

May be the Boeing guys can enlighten me much more..

jimmygill
9th Sep 2010, 08:16
Yes a lot of pilots take quick naps or power naps on a lot of flights esp
on the ones more than 2hrs cruise time in the middle of the night.
As long as the other person is informed of the nap and he accepts to handle
both aircraft and communication, its fine.


If they need a nap there should be a three pilot crew at least.
Human performance immediately after waking is probably at its bottom.
A two crew aircraft doesn't mean one sleeping crew and one awake.

How would a sleeping f/o know that captain has become incapacitated?


Disclaimer: My observation is without regards to this accident.

nikaviator
9th Sep 2010, 10:00
jimmy...unlikely scenario (3 pilots) ....with the trend of low cost operators' scheming for one pilot operations...:ugh:

Airline CEO: Nix co-pilot, save money - CNN.com (http://edition.cnn.com/2010/TRAVEL/09/06/ryanair.ceo.comments/index.html?iref=obnetwork#fbid=GYN5gNWkvXL&wom=false)

slayerdude
9th Sep 2010, 10:28
were there 2 captains on the flt deck... both pilot and co-pilot were ranked by the report as captain... was the guy on the right seat a training captain acting as co-pilot or were they both captains???? maybe a cockpit gradient thing

jimmygill
9th Sep 2010, 10:29
jimmy...unlikely scenario (3 pilots)

I know its unlikely, but that should not be an excuse for sleeping, no power nap or meditation. Its not a regular office.

iflytb20
9th Sep 2010, 10:50
were there 2 captains on the flt deck... both pilot and co-pilot were ranked by the report as captain... was the guy on the right seat a training captain acting as co-pilot or were they both captains???? maybe a cockpit gradient thing

In India ALL pilots are called Captains. The media usually distinguished between the Capt and the FO by calling them Pilot and Co-Pilot.

What I dont understand is on an aircraft like the B738 which is equipped with VNAV and when the crew can constantly check the vertical profile how come they ended up high ? (if they did as per the above report).

VNAV will only show you what you program it as. If you put the wrong height / speed values, VNAV can very easily put you in trouble. The best way would be to do a DME / Height cross-check throughout the descent.

IndAir967
9th Sep 2010, 11:15
I know its unlikely, but that should not be an excuse for sleeping, no power nap or meditation. Its not a regular office.

For once I disagree with u Jimmy. I dont know the kind of operations you are currently involved in. But when you do an midnight turnaround from an subcontinental destination back to the sandpit,
fatigue catches up with you and an 20-30 minute nap on an 4.5 hour sector wouldnt affect the safety or operation of the aircraft. As long as the other crew member gives his consent to the rest. Its a common happening especially in the A6 Registered Medium Haul Routes.
Courtesy: Our Ridiculous Roster :\

But what I do agree is when the crew involved is the Pilot Flying then he
must not be sleeping until the last minute as the body and mind will take some
time to return to active duty. The rest usually applies to PNF or PM.

If you put the wrong height / speed values, VNAV can very easily put you in trouble. The best way would be to do a DME / Height cross-check throughout the descent.

Agreed ! But if you end up programming the FMGS with wrong input, then
the discussion proceeds in an all new direction. But as of now it seems there
isint any FMGS programming issues on that ill fated aircraft.

goldenseth
9th Sep 2010, 11:27
I didn't aware that CVR on a B738 can record for more than 2 hours. The article says 'the CVR revealed that there was sound nasal snoring and heavy breathing for nearly two hours' + some time whthout snoring (I assume he didn't snore while awake) before landing.
It's true that we (pilots) learn new things everyday.:eek:
Don't do :zzz: inflight!

jimmygill
9th Sep 2010, 11:56
Mine are armchair operations, I take naps very often. But I don't have 150 plus passengers who rely on me for their safety.

20-30 minute nap on an 4.5 hour sector wouldnt affect the safety or operation of the aircraft

It does effect the safety of operations else the nap would have been in the SOP for long medium haul operations.

As I have told you, If I am taking a nap I have no clue what my f/o is doing, when he fainted on the controls. There will be pilot flying, but no pilot monitoring.

A sleeping pilot is an absent pilot.

If you try to answer this you will hopefully agree with me:

If one pilot is sleeping will you prefer the other to wear the oxygen mask or not?



And here is a news sotry, to prove that certain actions may be safe in certain conditions and fatal in others..
Joanne Kitchen stabbed then throttled for calling out wrong name during sex - mirror.co.uk (http://www.mirror.co.uk/news/top-stories/2010/09/08/joanne-kitchen-stabbed-then-throttled-for-calling-out-wrong-name-during-sex-115875-22545434/)

iflytb20
9th Sep 2010, 12:08
Agreed ! But if you end up programming the FMGS with wrong input, then
the discussion proceeds in an all new direction. But as of now it seems there
isint any FMGS programming issues on that ill fated aircraft. Let me rephrase that. A lot of waypoints have the "or Above" constraints. As per the FCTM, if we make the last "or Above" waypoint a hard altitude, it usually permits a normal deceleration for flaps and gear extension. If you leave it as an "or Above" waypoint, if the winds are not exactly as planned or if the ATC give late descent clearance, one can easily end up a bit high. I'm not saying they programmed it wrong. All i'm saying is thay if you are not monitoring the descent properly, you could end up a bit high.

masalama
9th Sep 2010, 12:57
Jimmy said:If they need a nap there should be a three pilot crew at least.
Human performance immediately after waking is probably at its bottom.
A two crew aircraft doesn't mean one sleeping crew and one awake.

How would a sleeping f/o know that captain has become incapacitated?

I know its unlikely, but that should not be an excuse for sleeping, no power nap or meditation. Its not a regular office.



You would be surprised to find my friend jimmy that many airlines around the world have already authorized and it's part of their SOP's short naps in flight and I'm not talking bunk beds but two man crew ops.
It's called strategic napping and has been found very effective through a great number of studies , most notably NASA/FAA on alertness managemet ...here's an excerpt from the findings:

"One effective alertness strategy needs further discussion given the context of this article. An extensive scientific literature exists of laboratory studies demonstrating that naps can improve alertness and performance. A NASA study extended these findings to operational settings by examining planned naps in pilots during actual flight operations. Results from the NASA study showed that a short, 40-minute planned nap resulted in a 34% improvement in performance and a 100% improvement in alertness. In real-world 24/7 operational settings, planned naps can be a critical strategy to address the spontaneous and uncontrolled sleep episodes that are known to occur."

This article and others like it showing how airlines/regulators around the world are addressing sleep/fatigue /alertness management issues head on are available on the net , if interested .

I'm not saying that fatigue was the sole reason for this accident but it could be a contributing factor .Decision-making and problem solving skills go for a toss when the body/mind are tired , the earlier that airlines/regulators wake up to this the safer our skies will be.Do we know the sleep/rest history of the two pilots involved and what was the sequence of flights before the accident one?

How can pilots help? Don't fly if you know you haven't had proper rest/sleep , try to be disciplined about getting that proper rest at home, layovers etc....I don't think any airline in India would fire somebody if he reports in sick genuinely due improper rest/sleep as long as it's not a chronic recurring problem... I've been lucky so far as it's mainly domestic ops at our airline with no flying on the back of the clock...keeping my fingers crossed.

masalama:ok:

jimmygill
9th Sep 2010, 13:57
You are right, I am surprised..

In hindsight DGCA's latest circular on requiring cabin crew to chat with pilots every half an hour, might just disturb that short nap.

I guess there is also a study how these benefits deteriorate when nap is extended.

Thanks, I will explore this further.

flyjet787
10th Sep 2010, 22:51
MUMBAI: Capt Z Glusica (http://timesofindia.indiatimes.com/topic/search?q=Capt%20Z%20Glusica), the commander of the Air India Express (http://timesofindia.indiatimes.com/topic/search?q=Air%20India%20Express) flight that crashed in Mangalore, was asleep for over an hour after the ill-fated flight took off from Dubai. While that may seem unthinkable and counter intuitive to flyers, it did not raise brows in the court of inquiry hearings, held this week in Delhi (http://timesofindia.indiatimes.com/topic/Delhi), as scientific studies have long proved that the practice of pilots taking naps in turns during a flight leads to higher levels of physiological alertness. Several foreign airlines encourage it and last year, Air India had written to the Directorate General (http://timesofindia.indiatimes.com/topic/search?q=Directorate%20General) of Civil Aviation (DGCA) urging that controlled rest in cockpit be made legal in India.

Taking a nap in the cockpit is strictly prohibited by the Indian regulator. But airlines in India have long tuned into international studies that have established that a short, controlled nap only leads to better flight safety. One of the first studies on this subject was done in 1990 when NASA (http://timesofindia.indiatimes.com/topic/NASA) compared performance of pilots who were allowed to take naps to those who weren't. The researchers found that pre-planned cockpit rest resulted in better behavioural and physiological alertness. In the last four-five years, several foreign airlines like British Airways (http://timesofindia.indiatimes.com/topic/British-Airways) and Qantas (http://timesofindia.indiatimes.com/topic/search?q=Qantas) took a fresh perspective after scientific studies proved the positive effects of cockpit napping (during low-activity periods) on air safety. These airlines have formulated strict guidelines on controlled rest during flights.

This practice, pilots say, is common in India and abroad, especially on long-haul flights in the night. " Air Canada (http://timesofindia.indiatimes.com/topic/Air-Canada) and some carriers in Australia (http://timesofindia.indiatimes.com/topic/search?q=Australia) are allowed to take controlled naps, lasting 35-40 minutes. We had written to the DGCA (http://timesofindia.indiatimes.com/topic/search?q=DGCA) last year asking that it should be allowed in India,'' said Capt V Kulkarni (http://timesofindia.indiatimes.com/topic/search?q=Capt%20V%20Kulkarni), training in-charge, Air India Express in his deposition before the court of inquiry. Some co-pilots who deposed before the court too spoke about flights where they took "short periods of rest with eyes shut'' in turns. "Then we have coffee,'' said a co-pilot.

Capt A Ranganathan, an air safety expert said that cockpit napping should be allowed. "The pilot rest rules followed in India were formulated in 1992 when there were no ultra-long haul flights and few late-night multi-sector flights to operate. But now, controlled rest in cockpits should be allowed as napping is an excellent fatigue-mitigation tool,'' he said.

It was probably for these reasons that no one raised brows when in the court of inquiry into Mangalore crash it was revealed that the cockpit voice recorder recovered from the wreckage played sounds of snoring and heavy breathing in the first 100 minutes after the flight took off from Dubai (http://timesofindia.indiatimes.com/topic/Dubai) late in the night. During this period, only the first officer's voice was heard. If anything, the commander's nap only raised questions about the out-dated pilot rest rules followed by airlines in India.

Though International Civil Aviation Organisation (http://timesofindia.indiatimes.com/topic/search?q=International%20Civil%20Aviation%20Organisation) (ICAO) had set a deadline for its member countries to implement scientifically back pilot rest rules, India has not yet complied to the order. The only aspect that the Indian regulator has looked into is the possibility of both pilots falling asleep at controls. Following an incident in India some years ago and a couple of them in US, the DGCA introduced a norm wherein flight attendants have to check on the pilots every half-an-hour during low-activity periods in the cockpit.

Read more: 'Snoozing on flights helps tired pilots' - The Times of India (http://timesofindia.indiatimes.com/city/mumbai/Snoozing-on-flights-helps-tired-pilots/articleshow/6533040.cms#ixzz0zAabgts0) 'Snoozing on flights helps tired pilots' - The Times of India (http://timesofindia.indiatimes.com/city/mumbai/Snoozing-on-flights-helps-tired-pilots/articleshow/6533040.cms#ixzz0zAabgts0)