Summary of final two NTSB briefings on Asiana 777 plane crash in San Francisco

On Wednesday, 10 July 2013 and Thursday, 11 July 2013, NTSB conducted their final two press conferences in San Francisco, and covered a variety of issues around the accident, including the operation of the autopilot and autothrottle, damage to the airplane, injuries to the flight attendants, the evacuation of the aircraft, and other initial factual findings from the investigation.

The NTSB emphasized in both of these press conferences that the information was factual in nature, and in many cases had not yet been confirmed or corroborated. For example, statements received from the flight crew still have to be matched up with information from sources such as the cockpit voice recorder (CVR) and the flight data recorder (FDR).

Cockpit automation and its role in the crash
As mentioned in a previous AirSafeNews.com article, use of the autothrottle by the crew to maintain speed was an issue because although the crew was heard on the CVR stating that the target speed was 137 knots, the aircraft was significantly slower than that speed before the crash. In Wednesday's press conference, the NTSB stated that there were five distinct autothrottle modes used in flight, and in the last 2.5 minutes of flight, there were several autothrottle and autopilot modes used.

As explained by the NTSB, the autopilot helps pilots manage pitch, roll, attitude, and heading; while the autothrottle helps to control speed or thrust. The two systems can work together, and the NTSB has to determine, with the help of Boeing, the following:

• Whether autopilot and autothrottle modes were commanded by the pilots or activated inadvertently,
• How the various autopilot and autothrottle modes are designed to work, and
• What are the ways the systems are expected to respond in the various modes.

Comparison to automobile cruise control
NTSB chair Deborah Hersman used an analogy to a much simpler automated system to illustrate the role that an autothrottle plays. Like in an airliner, a car's cruise control can be set to a specific speed, but it is up to the driver to monitor the speed. Also, cruise control may not engage if the car is in a particular mode, for example below a certain speed. While in cruise control, the driver may be allowed to increase or decrease speed within certain limits. Disengaging cruise control can be done by disarming the system or by hitting the car's brake.

Status of the pilots on flight 214
There were a total of four pilots on board, and they consisted of two crews. The first crew consisted of a training captain going through his initial operating experience (IOE) on the 777 and an instructor pilot (IP) who was a training captain. The relief crew consisted of a captain and a first officer (FO) This first crew performed the takeoff from Seoul and flew for several hours before the relief crew took over, and then the first crew flew approximately the last 1.5 hours of the flight.

Shortly before landing, when the aircraft was at around 10,000 feet, the relief FO entered the cockpit and was in the jumpseat for the rest of the flight. The NTSB provided details on the experience of the three pilots in the cockpit:

• The training captain was in the left seat at was the pilot flying (PF),
• The training captain had about 9,700 total hours, including about 5,000 as pilot in command (PIC),
• The PF was hired by Asiana in 1994, and trained in Florida,
• The PF was rated to fly the 737, A320, 747, and 777, and from 2005-2013 flew the A320, serving as an A320 captain before moving to the 777,
• The PF was also a ground school and simulator instructor for the A320 and A321
• The IP was also a 777 captain who served in the South Korean air force for about 10 years before joining Asiana,
• The PF's IOE was to consist of 60 flight hours and 20 flight legs, and had gone through 10 flight legs and about 35 flight hours at the time of the crash,
• The IP had about 13,000 flight hours, including about 3,000 in the 777, and 10,000 as a PIC,
• The IP served as the PIC on flight 214, and was sitting in the right seat,
• This flight was the first time that the PF and the PIC had flown together, and it was the PIC's first trip as an instructor pilot
• The relief FO was a former F-5 and F-16 pilot in the South Korean air force, and had about 4,600 total hours, including 900-1,000 hours in the 777,
• The relief FO had flown to San Francisco five or six times as an observer.

Landing aids in use at the airport
Air traffic control was allowing pilots to operate under visual flight rules (VFR) when flight 214 was approaching the San Francisco airport (SFO), which means that pilots were not required to use the instrument landing systems at the landing runway (28L) or any automated systems on their aircraft. One of the electronic aids that provide aircraft guidance on their glide slope was inoperable, but this had been published for some time and all flight crews using the airport should have been able to see this information. The NTSB has not stated if this crew were aware of this.

A glide slope aid that was in operation at runway 28L were the precision approach path indicator (PAPI) lights, a set of four lights arranged in a horizontal line that provide pilots with a visual indicator of whether the aircraft on the glide slope, above the glide slope, or below the glide slope.

A pilot who is on the glide slope would see two sets of red lights on the left and two sets of white lights on the right. In the example shown here (not from SFO), the three left indicators are red and the right one is white, indicating that the aircraft is slightly below the glide slope. Four red lights would indicate that the aircraft is well beloe the glides slope, and four white lights is an indicator of being well above the glide slope.

Final approach sequence
Over the last two press conferences, the NTSB discussed the following key parts of the final approach:

• The approach path took the aircraft directly over SFO, followed by a wide teardrop left turn to line up with the runway (see below),
  
  (click to enlarge)
• Air traffic control (ATC) called for a maximum airspeed of 180 knots until the aircraft was five miles out,
• The IP recalled that the aircraft was above the intended glide path at 4,000 feet, and that vertical speed mode was set at 1,500 feet per minute,
• ATC gave a landing clearance about 1.5 miles from the runway, about 90 seconds prior to the crash,
• There was a sink rate callout prior to the aircraft reaching 500 feet,
• At about 500 feet, the FP noted a blinding flash of light directly in front of the aircraft but not on the runway,
• The FP stated that he looked away into the cockpit, and was able see the cockpit instruments, including the speed tape,
• There was no mention of the light on the CVR,
• The FP believes it may have been a sun reflection, and the NTSB is determining if this could have been the case,
• There was an automated 500 foot callout about 35 seconds before the crash,
• Shortly after this callout, the landing checklist was completed,
• At about 34 seconds prior to impact, the IP noted that the aircraft was below the glide path at 500 feet, and speed was at about 134 knots, with three red PAPI lights showing, and told the PF to pull back
• Autothrottle was armed and set at 137 knots
• between 500-200 feet, the IP noted that there was a lateral deviation and that the aircraft was low,
• At 200 feet, the IP noted four red PAPI lights, that the speed tape was hatched (a visual indicator of an impending stall), and that the autothrottle had not maintained speed,
• There was an automated 200 foot callout 18 seconds before impact,
• There was an automated 100 foot callout nine seconds before impact,
• Almost immediately after this 100 foot callout, a crew member mentioned airspeed (the NTSB noted that there were no mentions of speed heard on the CVR between 500-100 feet),
• About three seconds before impact, there was a call for a go around,
• The IP established a go around attitude, and went to push the throttles forward manually, but saw that the FP had already done so,
• A second call made for a go around was made by a different crew member about 1.5 seconds before impact.

Crash sequence

• The main landing gear hit the sea wall first, followed by the tail section,
• The main landing gear sheared away from the aircraft as designed, and the wing fuel tanks were not punctured by the gear separation or during the the rest of the crash sequence,
• Cabin flooring and galley components were found on the chevrons in the runway overrun area between the sea wall and the runway threshold,
• The initial impact displaced rocks from the sea wall and some of them were distributed several hundred feet along the debris trail (see photo below),
  
  (click to enlarge)
• All passenger seats stayed inside the cabin, but three flight attendant seats were ejected onto the runway,
• Door 4L detached from the aircraft at some point in the crash sequence
• Six of the 12 flight attendants were interviewed, and they stated that two of the eight escape slides inflated inside the cabin after a secondary impact (from a witness video, it appears that the aircraft rotated almost 360 degrees counter clockwise, with the rear of the cabin rising up at an angle before hitting the ground at the end of the crash sequence),
• The right engine had detached from the wing, had rotated about 90 degrees counter clockwise, and was laying alongside the fuselage (see photo below).

(click to enlarge)

Post-crash actions and fire

• After the aircraft came to a stop, the lead flight attendant (who was near door 1L) went to the cockpit for advice, and was advised not to initiate evacuation (see door layout in photo below),
  
  (click to enlarge)
• Fire extinguisher switches were pulled for both engines and the auxiliary power unit,
• The flight crew was able to communicate with the control tower, and the cabin crew was able to use the public address system to communicate to passengers,
• A flight attendant who was trained as a lead flight attendant was at door 2L, saw fire outside door 2R near row 10 of the aircraft, and sent the other flight attendant at door 2L to the front of the cabin to inform the rest of the crew about the fire and the need to evacuate (in earlier briefings, the NTSB stated that the source of the fire was a ruptured oil tank that leaked fuel onto hot engine parts from the right engine),
• Passenger evacuation began about 90 seconds after the aircraft came to a stop, escape slides were first deployed from door 2L and then from door 1L, and passengers also escaped from door 3R,
• The control tower called for emergency vehicles after the aircraft hit the runway, the first vehicle arrived about two minutes after the crash, and extinguishing agent was first applied about three minutes after the crash,
• Cabin emergency exit lighting came on during the evacuation,
• There were six flight attendants who were injured and hospitalized: three seated in the rear of the plane who were ejected out onto the runway, another flight attendant in the rear who was injured, and two who were injured by the slides that deployed inside the cabin including one at door 1R and a second at door 2R,
• The remaining six flight attendants had evacuated most of the passengers by the time the fire had spread to the cabin,
• Aiport fire crews entered the cabin with a fire hose to help fight the fire,
• Flight attendants helped to fight the fire with fire extinguishers, and also used the extinguishers to help extract the two flight attendants who were trapped by the two escape slides that deployed inside the cabin.

Cabin damage
Prior to the cabin fire, a firefighter entered door 2L and turned right to walk toward the rear of the cabin, and along the way observed that seats in that section were almost pristine, with minimal damage detectable, and that one could just fluff the pillows to get that section ready for the next flight. As he walked toward the rear, he observed more cabin damage, with a sharp contrast between the front and back of the passenger cabin. The photo below shows the pristine area of the cabin that was later damaged by fire.

(click to enlarge)

The NTSB structures team noted that from the cockpit to rear spar of the center wing box, the cabin floor was structurally sound. Aft from the rear spar to doors 3R and 3L, in the passenger seating compartment, support structure were compromised on the right side (flayed out from the aircraft), but still sound on the left side. Between doors 3 and 4, the floor was canted down at an angle, with damage progressively worse towards the back, and there was no cabin floor behind door 4.

Dr. Curtis and Capt. Tom Bunn discuss the crash
The day after the crash, Dr. Curtis of AirSafe.com and Capt. Tom Bunn of the SOAR fear of flying program, who both spent several hours on the day of the crash on cable news programs providing expert commentary, discussed the media's response to the accident and shared their thoughts on the early reports of the crash.

Additional information
AirSafeNews.com 10 July 2013 article on the role of the autothrottle
AirSafeNews.com 8 July 2013 article on early findings of the crash investigation
Other Asiana plane crashes
Other 777 plane crashes
Accident details from Aviation Safety Network
Wikipedia page on this accident

Photos: Wikipedia, NTSB

Asiana 777 accident investigation and the role of the autothrottle

During the Tuesday, 9 July 2013 NTSB press conference, numerous facts about the accident were revealed, including details about the training and experience of the pilots in the cockpit, and the fact that the initial impact not only tore the tail section off the aircraft, but also caused two flight attendants seated in the rear of the aircraft to be ejected out of the cabin. Both survived, but were injured.

Perhaps the most revealing information from the conference was evidence gathered from statements from the pilots, that seemed to indicate that the while the pilots had planned to use the autothrottle to control the aircraft's airspeed during landing, the autothrottle was not engaged at the time the crew was attempting to execute a go around in order to attempt another landing.

The NTSB emphasized that this was preliminary factual information that has yet to be corroborated with other data from sources such as the flight data recorder. However, it implies that one of two possible scenarios may have occurred during the latter stages of the flight:

1. The crew intended to use the autothrottle, but did not take all the steps needed to engage the autothrottle, or
2. The flight crew took steps to engage the autothrottle, but the autothrottle either did not engage or it disengaged at some point.

Basic autothrottle operation
In order to understand the possible significance of these preliminary NTSB findings, it helps to have a bit of background knowledge on how autothrottles are used.

In the 777, as in many modern airliners, the autothrottle allows a pilot to control the power setting of an aircraft's engines automatically rather than manually. Flight crews use the autothrottle to maintain, or try to attain a particular value for either speed or thrust without having to manually adjust throttle settings. For example, a pilot may want to maintain a specific airspeed, and would use the autothrottle to maintain that airspeed while the pilot may be manually controlling other aspects of the flight.

The autothrottle can also enhance safety by keeping the aircraft within safe operating limits. For example, if the pilot commands the autothrottle to attain a speed that is at or below a minimum safe speed or above a maximum safe speed, the autothrottl will not allow the aircraft to fly at those unsafe speeds.

Arming and engaging the autothrottle
In the 777, using the autothrottle to control airspeed is a two-step process. First, the autothrottle has to be armed using two switches (one for each engine) on the mode control panel (MCP). Being armed means the autothrottle is available to be used. The second step is to engage the autothrottle, which means it is now being used to control airspeed. The autothrottle is engaged by using an appropriate switch on the mode control panel.

Mode control panel
The cockpit of many modern airliners, including the 777, have a mode control panel (MCP), which contains the controls that the flight crew would need to automatically manage the aircraft's flight, and in the 777, the MCP controls a number of functions, including the autopilot and autothrottle.

Below are two photos depicting the 777. The first is a representation of the MCP from a NASA computer simulator, and the second is from the cockpit of a 777 and shows both the MCP and several cockpit display screens. Note the neither one may represent the exact configuration of the MCP in the Asiana accident aircraft.

(click to enlarge)

(click to enlarge)

The following video describes how a simulated version of the 777 MCP behaves. The first couple of minutes describes how the autothrottle has to be armed before it can be engaged and used to control speed.

---

Is the autothrottle the key to the cause of the accident?
The NTSB emphasized in their press conference that all of the information that they have released so far is factual in nature, and that they have no intention of speculating or deciding upon a cause or causes of the accident at this stage of the investigation. Also, the information gathering stage of the investigation is ongoing, and there may be other facts that the NTSB either has not yet discovered or has not yet released to the public.

Another perspective on this accident
For an excellent perspective on this accident and the revelations from the early part of the investigation, I highly recommend the Slate article of Patrick Smith, a professional airline pilot and recent guest of the AirSafe.com podcast.

Dr. Curtis and Capt. Tom Bunn discuss the crash
The day after the crash, Dr. Curtis of AirSafe.com and Capt. Tom Bunn of the SOAR fear of flying program, who both spent several hours on the day of the crash on cable news programs providing expert commentary, discussed the media's response to the accident and shared their thoughts on the early reports of the crash.

Additional information
AirSafeNews.com 13 July 2013 article
AirSafeNews.com 10 July 2013 article on the role of the autothrottle
AirSafeNews.com 8 July 2013 article on early findings of the crash investigation
Other Asiana plane crashes
Other 777 plane crashes
Accident details from Aviation Safety Network
Wikipedia page on this accident

Photos: 777boeing.com, NASA

Mapping the intersection of mind and computer in the cockpit

The following is based on the article Mapping the intersection of mind and computer from guest writer Christine Negroni

Well my inbox is filling up again with emails, as it did last month when I reported the following story for The New York Times on pilot complacency and cockpit automation.

Prompting the latest flurry of comments is a June 15, 2010 article by Andy Pasztor and Daniel Michaels in the Wall Street Journal about the crash in May 2010 crash of an Afriqiyah Airways A330. Only one of the 104 people on the Airbus A330 from Johannesburg to Tripoli survived the accident.

According to Pasztor and Michaels the landing accident is being seen as one in “which confused pilots got out of sync with the plane's computerized controls and ended up flying an apparently functioning commercial jet into the ground.”

This is no one-off event. A number of studies over the past 15 years indicate pilots fail to adequately monitor what the airplane is doing in one-half to three-quarters of all accidents. So in the wake of the Afriqiyah Airways disaster, what’s the big idea being proposed? More automation. That’s right, Airbus is said to be working to “devise foolproof automated ground-collision avoidance systems” that in cases of emergency transfer control from the pilots to the airplane.

“This is very disturbing”, wrote Hugh Schoelzel, a retired captain who worked as director of safety for TWA. “The more automation we add, the more training and pilot qualification issues arise. I believe in automation, but as an adjunct to basic pilot skills, not as an ‘end-all’.”

While automation may be causing a decrease in piloting skills as Mr. Schoelzel suggests, Professor Missy Cummings of the Massachusetts Institute of Technology says there is another reason to be concerned about cockpit automation; boredom.

Dr. Cummings a former Navy pilot, is director of the humans and automation laboratory at MIT’s department of Aeronautics and Astronautics. Not surprisingly she is a proponent of automation and envisions a future that will include at least some pilotless commercial flights. But first some extremely troublesome problems have to be wrestled to the ground, problems demonstrated by one of Dr. Cummings students, Master’s degree candidate First Lt. Christin S. Hart, who has found that too much automation can prove counter-productive.

“Increased automation can lower an operator’s workload too much, leading to mental underload, which can cause a decrement in vigilance, or sustained alertness, and lead to boredom. It has been shown that boredom produces negative effects on morale, performance, and quality of work,” she wrote in her paper, Assessing the Impact of Low Workload in Supervisory Control of Networked Unmanned Vehicles.

These findings do not surprise Dr. Cummings “The human mind craves stimulation”, she explained to me last week during a visit to her office in Cambridge. Failing to find that stimulation in the task at hand, the mind will wander.

This cuts to the heart of a number of events outlined by industry researchers but takes us at warp speed to the episode last October in which two Northwest Airlines pilots overflew their destination - the Minneapolis airport. The Northwest pilots were doing personal work on their laptops which is not allowed.

“It doesn’t have anything to do with automation,” FAA Administrator Randy Babbitt told me. “Any opportunity for distraction doesn’t have any business in the cockpit. Your focus should be on flying the airplane.”

But if I’m reading Lt. Hart’s study properly, the automation itself is an opportunity for distraction, even as it assists pilots by reducing workload and increasing the precision of calculations and navigation.

This is a conundrum. In today’s cockpit, two highly complex systems – the mind and the computer – come together, even though the contours of that intersection are still being mapped. It is not only unwise to race to a fix that fails appreciate these systems in balance, but it is unlikely to result in success.

Related Content
NTSB opens public docket on Northwest overflight

NTSB to investigate near miss involving a 777 over San Francisco

The NTSB has launched an investigation into a nearly catastrophic collision between a United 777 and a small private aircraft over San Francisco last weekend. The NTSB estimates that the two aircraft missed each other by less than 300 feet.

At about 11:15 a.m. PDT on March 27, United Airlines Flight 889, a B777-222 (N216UA) carrying 251 passengers and 17 crew members on a flight to Beijing, China, had just departed from San Francisco International and was climbing through about 1,100 feet when both air traffic control and the aircraft's collision avoidance system reported air traffic nearby.

The flight crew saw a small aircraft, an Aeronca 11AC (N9270E), in a hard left turn traveling from their 1 o'clock to 3 o'clock position. The first officer interrupted the climb and leveled out the aircraft. Both crew members reported seeing only the underside of the Aeronca as it passed to within an estimated 200-300 feet of the 777.

After the encounter, the flight crew continued to Beijing without further incident.

You can listen to this edited recording of transmissions among air traffic controllers and the two involved aircraft (Source: LiveATC.net).

Midair collisions that result in passenger deaths on large jet airliners are very rare events. According to AirSafe.com's list of midair collisions, the last midair involving a jet airliner was a 29 September 2006 collision between a Gol Linhas Aereas 737-800 and ERJ135 Legacy 600 executive jet over Brazil. All on board the 737 were killed, and everyone on the ERJ135 survived.

The last airliner midair in the US was a 31 August 1986 collision between an Aeromexico DC9-32 and a Piper Archer over Cerritos, CA. All occupants of both aircraft were killed. More recently, a private plane and a sightseeing helicopter collided over the Hudson River near New York, killing all on board both aircraft.

Photo credits: Drewski2112 (777), Wikipedia (Aeronca)

Letting kids talk on ATC radio not smart, but also not dangerous

Last month, an air traffic controller on two occasions brought a child to work at the control tower at JFK Airport in New York and allowed the child to radio instructions to pilots. By now you have probably heard the exchange between one of the children and a jet on the ground at JFK. Clearly, having a child talking to an aircraft is out of bounds, and the FAA has responded by suspending both the controller and the controllers supervisor pending an investigation of the incident. Like the deliberate crash in Austin and the accidental crash in East Palo Alto discussed previously by AirSafeNews.com and Plane-Crash-Videos.net, audio played an unexpected role in the JFK incident.

Air traffic control transmissions and LiveATC.net
Unlike cockpit voice recordings, which are rarely released to public, transmissions between aircraft and air traffic control can be heard by anyone with the proper equipment. In the old days, one had to be in the local area with a specialized video. Today, thanks to the Internet and a worldwide network of enthusiasts, it is easy to listen to just about any air traffic control transmission. In the last few weeks, one web site, LiveATC.net, has provided the public with recordings of one of the two children at JFK, and the eerily normal conversation the Austin suicide pilot had with ATC shortly before his death. The simple lesson is that any conversation with ATC can be recorded, and if the recording is particularly controversial, it will be widely available to the entire world.

Access differences between cockpits and control towers
Access to airline cockpits and air traffic control towers in the US are subject to many restrictions, with airliner cockpit access being extremely restricted during all phases of flight, and it is very difficult for anyone other than crew member or government official. Control towers on the other hand don't have severe restrictions, and having visitors in the control tower or other ATC facility is a common occurrence.

How dangerous was it to have a child on the radio?
The JFK incident, although good for a few scary headlines, was a clear violation of the rules, but probably not a violation that put aircraft at risk. The child was apparently having a routine conversation with an aircraft that was on the ground, with the air traffic controller directing what the child said. There was no indication that the pilots were told to do anything that was in any way risky.

Is audio surveillance a good thing?
Had the JFK incident happened 15 years ago, it is very likely that anyone other than the the pilots and air traffic controllers tuned to that frequency would have heard it. If the public found out about it, it would not have been through a recording, but through a second or third hand story that would likely escape the attention of the mainstream media.

In contrast with the past, the future will likely have more, and not less private audio surveillance from the likes of LiveATC.net. If you are a pilot or air traffic controller, the safe assumption is that anything that gets broadcast can be recorded. Given the ongoing improvements in computers, storage, and search technology, it would be safe to assume that one day transmissions would not only be recorded, but also accessible and searchable online.

Whether this kind of surveillance is good or bad isn't important. It is a reality that is happening now, and will likely continue to happen in the future. Given these trends, the best advice to any pilot or air traffic controller is to be careful of what you say on the air.

Audio of both incidents on February 16 and 17th
The the following video, courtesy of AirBoyd.tv has audio from both days. Day two starts at around the three minute mark. Clear space is edited out. The radio scanner that recorded the audio clips the audio between both tower channels as it's recorded, so not every word exists on this audio recording. This is the reason for missing audio in some sentences.

NTSB to Open Up Public Docket on Northwest Airlines A320 Overflight in October 2009

Detailed information from the NTSB investigation into the October 21, 2009 incident involving the distracted Northwest Airlines pilots have been released to the public. The incident, which resulted in a large amount of media attention, occurred when the pilots of Northwest Airlines flight 188 was out of contact with air traffic control for about for 77 minutes while cruising from San Diego to Minneapolis.

The Airbus A320 (N03274) overflew Minneapolis by more than 100 miles before re-establishing radio contact with air traffic controllers and landing at the destination airport. There were no injuries to the 149 aircraft occupants, or damage to the aircraft.

Overview of Final Stages of the Flight

Reportedly, one of the reasons why the pilots did not respond to radio calls was because they were using their laptop computers in the cockpit and were distracted. While using laptops in the cockpit is allowed by the FAA, use of personal computers in the cockpit was not allowed by Delta (which owns Northwest). Both pilots were dismissed by the airline, and the FAA revoked their licenses.

Included in the docket will be factual reports from several of the NTSB-led groups involved in the investigation, including the Operations Group, Air Traffic Control Group, Survival Factors Group, Cockpit Voice Recorder Group and Flight Data Recorder Group. The docket will be available on Wednesday morning, December 16th. To view the contents of the docket, visit the the public docket page for this incident.

Related Information
Northwest Flight 188 public docket page
Previous AirSafeNews.com Incident Update
Original AirSafeNews.com article on this incident
Delta Airlines Statement 26 October 2009
FAA air traffic control transcripts and audio recordings
NTSB Update from 26 October 2009

Updates on Two Recent Qantas Accidents Involving an A330 and a 747

Second Interim Report on Qantas Accident of 7 October 2008
The Australian Transport Safety Bureau (ATSB) released its second interim report about a 7 October 2008 event when a Qantas A330-300 (VH-QPA, Flight 72) experienced an unexpected and rapid change in altitude while the aircraft was in cruise at 37,000 feet. The accident resulted in injuries to 110 passengers and nine crew members.

The first interim report identified two significant safety factors. First, one of the air data computers provided incorrect data that was not detected by the aircraft systems. Second, the aircraft's flight control computers did not process some of the aircraft's attitude data in a specific situation.

The second interim report did not identify the cause of the October 2008 event, but detailed the safety actions that have been taken that would prevent a recurrence of the event, including modifications to the A330's primary flight control computer and changes in flight crew operational procedures.


The ATSB addressed the possible relationship between the October 2008 Qantas A330 event and the June 2009 crash of an Air France A330-200 (F-GZCP, Flight 447). Although both investigations are ongoing, the ATSB pointed out several key differences between the two events:

• The air data computers on the two aircraft were different models, and constructed by different manufacturers.

• The cockpit messages and maintenance fault messages from both flights showed a significantly different sequence and pattern of events, with the maintenance messages that were transmitted by the Air France aircraft prior to the accident showing inconsistencies between the measured airspeeds and the associated consequences on other aircraft systems. No such messages were recorded by by the Qantas aircraft.

• The airspeed sensors (pitot probes), which were a issue of great concern in the Air France Accident, were not an issue in the Qantas accident because they were different airspeed sensor models made by different manufacturers.

Update on Qantas 747 Decompression Event
On 25 July 2008, a Qantas 747-400 (VH-OJK, Flight 30) experienced a rapid decompression while cruising at 29,000 feet after an oxygen cylinder that was part of the emergency oxygen system exploded and blew a hole in the fuselage. In their second interim report on the decompression event, the ATSB reported that there is no evidence of a safety problem with the oxygen bottles of the type involved in the accident.

ATSB reported that among the actions taken by Qantas were a fleet-wide safety inspections of oxygen system installations and a revision of flight crew emergency procedures, including the introduction of a new depressurisation checklist. Also, various tests have not been able to replicate the cylinder failure that initiated the accident.

Additional Information
You can find more details about these three accident investigations, including links to the interim reports from the investigating authorities, at the following pages:
Qantas 747 Depressurization Accident 25 July 2008
Qantas A330 In Flight Upset Accident 7 October 2008
Air France A330 Crash 1 June 2009