Engine Failure on Take-Off : Air Algerie Flight 6289

10 Jun 16 14 Comments

On the 6th of March 2003, a Boeing 737-200 registered as 7T-VEZ, owned and operated by Air Algérie, was scheduled for a domestic passenger flight from Tamanrasset to Algiers via Ghardaïa. All times are given in UTC.

The flight was three hours delayed that day for the scheduled flight DAH 6289 to Ghardaïa and Algiers. There were six crew members (two flight crew and four cabin crew) and 97 passengers on board.

The captain was late for the flight. The first officer did the flight preparation on her own. She identified two errors when it came to the weight of the aircraft: the ramp operations technician had noticed 8,800 instead of 9.800 kilograms and two absent passengers had been included in the count. She corrected these issues and calculated the aircraft’s take-off weight (TOW) as 48,708 kilograms. The maximum authorised weight was 49,500 kilograms. The balance was 22.70% which was within normal range. She correctly approved the weight and balance for the flight: the aircraft was close to the maximum but still within range.

The captain arrived and joined the first officer, inviting the chief flight attendant into the cockpit. The first officer was left to carry out the flight preparation alone while the two men held a conversation. The first officer finished the flight preparations and began the pre-take-off briefing with the V-speeds for the aircraft take-off.

V-speeds are used to define airspeeds important to the operation of aircraft. The actual speeds are specific to the flight but are referred using V-references for clarity. An aircraft’s airspeed indicator will show the key V-speeds as colour-coded arcs. (See Rudy’s comment below, bugs make much more sense!)

V1 is the decision speed, also known as the “engine failure recognition speed”. Once the aircraft exceeds this speed on the take-off run, the take-off should no longer be aborted. V1 is the maximum speed at which the pilot can take the first action to stop the aircraft and still safely come to a halt within the aircraft’s stop distance.

Looking at it from the other side, in the case of an engine failure, V1 is the minimum speed at which the aircraft can continue the take-off and achieve the required height above the runway. Thus if an engine failure occurs after V1, the take-off is continued.

V2 is the take-off safety speed. That’s the minimum speed that needs to be maintained in the event of an engine failure which takes place after V1. If the aircraft is travelling at this speed or greater, the aircraft should be able to climb away safely, even if one engine has failed.

VR is the rotation speed, the speed at which the pilot flying causes the aircraft’s nose wheel to pitch up so that the aircraft leaves the ground.

So before the flight it is standard to brief these these three speeds critical to take-off (or aborting take-off). This ensures that, if there is an engine failure on take-off, it is immediately clear what the next action should be: continue or abort.

One Engine Out Procedure for Tamanrasset

The first officer started the pre-take-off briefing by announcing the speeds for the take-off:
V1:144 kt, VR: 146 kt and V2: 150 kt.

However, the captain interrupted her to speak to the chief flight attendant and her checklist was cut off.

The first officer never resumed the checklist. She asked to perform the flight leg and the captain agreed that she would be Pilot Flying and he would be Pilot Monitoring. They requested start-up clearance and the controller cleared the aircraft to taxi, enter and line up at runway 02.

Four minutes later, the first officer confirmed to the tower that they were ready for departure and were cleared for take-off.

The flight crew powered up the engines and began the take-off run. Fifty-two seconds later, the captain, in his role as Pilot Monitoring, announced rotation (VR) as the aircraft speed reached 150 knots. At the point of rotation, the aircraft airspeed was 160 knots.

The aircraft climbed at a rate of between 1,400 and 1,800 feet per minute with the pitch attitude at 18°.

The first officer asked for the landing gear to be retracted but, as she spoke, there was a sharp thumping noise. The aircraft speed decreased to 158 knots. At this moment, the Boeing 737 was 78 feet (24 metres) above the runway.

The aircraft veered to the left, its magnetic heading changing from 020° (in line with runway 02) to 008° and then correcting back. The first officer cried out in surprise and then said, “What’s going on?”

She couldn’t know it but what had happened was that the left-hand engine had suffered a contained explosion.

From the accident report:

Engine power up, airplane acceleration and the standard callouts were all carried out normally until rotation. The problems started suddenly, at the exact moment landing gear retraction was announced. All evidence gathered by the investigation, especially the airplane parts found on the runway near the place where the airplane lifted off, as well as the twelve degree change of heading to the left, show that the crew was then faced with a contained left engine burst.

Examinations showed that the problem originated in the turbine high-pressure stage located just after the combustion chamber. The rupture of a relatively large part caused the immediate destruction of the high-pressure turbine and the subsequent damage to the low-pressure turbine. The burst caused a sharp drop in the rotation speed and in engine thrust, without causing it to shut down.

Apart from the airplane’s tendency to yaw to the left, the damage to the left engine would normally result, in the cockpit, in a sudden reduction in the performance readings for that engine and a visual oil pressure warning, though without a fire alarm.

The airline procedures for dealing with an engine out is to retract the landing gear and continue the climb until the safety altitude. However, the weight of the fully-laden aircraft, the aerodrome altitude and the high temperatures that day were such that even if the flight crew had done everything right, it would have been difficult to maintain a positive rate of climb.

The crew, suddenly confronted with an emergency situation, reacted with confusion. The captain did not repeat back the Pilot Flying’s order to retract the landing gear and he did not retract it. Neither pilot announced the engine failure. Neither pilot mentioned any of the visual warnings, such as the oil pressure warning. No callouts were made regarding the speed, climb rate or trim of the aircraft. From the moment that the engine failed, there was no coordination or cooperation between the crew at all.

The captain asked the first officer to let go of the controls. This is a shift in roles: he was asking her to relinquish control of the aircraft as he was taking over the role of Pilot Flying. It’s unclear as to why he made this split-second decision: we don’t know if the first officer was struggling or if he simply felt that as the captain, he should be flying in an emergency situation. The last thing that the first officer had done was to ask for the landing gear to be retracted. The captain took control instead, leaving the landing gear extended. As Pilot Monitoring, the first officer now couldn’t retract the landing gear until he asked her to. The impact of this is that the aircraft, which already would have struggled to ascend to a safe height, was not configured for best rate of climb.

In any event, the captain clearly took control before he had a chance to assess the situation. The first officer did not request him to and it’s not possible to tell from the Flight Data Recorder whether he took the controls and then spoke or if he asked her to release control before he took action. It’s clear however, that he did not understand the state of the aircraft in that moment; precious time was lost as he tried to decide what to do. This decision could have been better made by leaving control with his first officer or, at the very least, interacting with her.

The captain maintained a high pitch attitude which he couldn’t hope to maintain on only one engine. The aircraft continued to lose speed. He repeated his request for control several times, implying that the first officer had instinctively returned her hands to the controls although she agreed initially that he had control. She asked the captain if she should retract the landing gear: that had been her last request before the emergency and as Pilot Monitoring, it was now her job to retract the landing gear when the captain, as pilot flying, requested it. However, he didn’t reply.

The first officer left the landing gear extended and called the control tower and said, “We have a small problem.” She didn’t appear to know what to do in order to support her captain. She seemed willing to move to a support role: she read back his orders and asked about retracting the gear, as well as speaking to the controller to let them know that there was an emergency. However, he had to repeat his request for control (“let go” and “take your hand off”) until the end of the recording. The investigation report wonders in the end if, perhaps, she was instinctively placing her hands on the control column each time the stall warning sounded.

The aircraft continued to climb and lose speed. About fourteen seconds after the sound and the initial loss of speed, the sound of the stick shaker filled the cockpit. The stick shaker rapidly and noisily vibrates the control yoke to warn the pilots that the aircraft is at risk of a stall. The aircraft had climbed to 398 feet and was travelling at 138 knots. The stick shaker vibrated again. The airspeed was perilously low.

The Ground Proximity Warning System broadcast a warning: Don’t Sink! This is a specific warning which sounds after take-off (when the aircraft is still below 900 feet) in order to warn the flight crew that the aircraft has lost altitude. The stick shaker vibrated continuously as the aircraft began to descend.

The captain maintained the same pitch attitude throughout. The last recorded position was 335 feet above the runway travelling at 126 knots with a magnetic heading of 005°. The GPWS broadcast Don’t Sink again and then both recorders failed.

The rear of the aircraft struck the ground first, with an impact on the right side. The fuel spilling from the right wing immediately caught fire as the aircraft slid along, knocking over an airport perimeter fence and crossing a road.

Fire fighters were on standby and set off as soon as they noticed the aircraft not climbing away as expected.

The tower controller raised the alarm while the aircraft was still in the air and then phoned the Tamanrasset fire service who set off immediately to back up the fire fighters.

A controller told investigators:

…just after the takeoff from runway 02 (1405) a kind of explosion was heard, the alarm was immediately activated, the pilot said we have a small problem. . . the plane began to fall and crashed near the threshold of runway 20; the emergency plan was immediately activated as planned. 1) Aerodrome rescue services at 14:15. 2) Civil services at 14:16. 3) Hospital just afterwards.

On the other side of the road, the rescue teams found the right main landing gear (without wheels) and then the main wreckage, which had already been almost entirely destroyed by fire.

Wreckage

One passenger survived. Sitting near the back without his seatbelt on, he was ejected from the aircraft when it hit the ground and escaped the accident.

No one else survived. The chief flight attendant’s body was found collapsed over the centre console of the cockpit. The rest of the cabin crew and the passengers were all in their places with seat belts still attached.

A ground technician who had worked on the aircraft saw the whole thing.

I was on the parking lot and I saw the plane take off on runway 02. Just after the take-off, the plane swerved slightly to the left, then righted itself on the track and at that moment, I noticed that the plane was losing speed and altitude, still with its landing gear down until the moment of the crash, when there was a total explosion.

The aircraft was destroyed by the impact and the fire. The wreckage showed that the horizontal stabiliser trim was still in normal take-off position. The captain appeared to have intentionally kept an excessive climb angle until impact, so focused on this one parameter that he put the aircraft straight into the stall.

Both engines and their accessories were transported to Brussels to a specialised laboratory. Examination of the left engine showed that the problem stemmed from the high pressure part of the turbine. The part backing of the low-pressure turbine was destroyed as a result, which accentuated the loss of power and led to a sharp drop in the engine rotation speed.

The absence of a part of the distributor and the immediate destruction of the HP turbine blades stopped the expansion of gas normally produced by the HP stage, which, in turn, prevented the cooling triggered by the gas leaving the combustion chamber. Next, excessive temperatures rapidly melted the NGV 2 part of the low pressure spool, which was then directly exposed to the heat of the combustion chamber. The melted metal projections observed in the flame tubes and the other hot sections of the engine showed that the combustion chamber was functioning at the moment of impact.

The rotating parts of the right engine showed deformations which meant that at the moment of impact, the right engine was rotating at low speed and not developing any thrust. The reduction of thrust happened a few seconds after the left-engine lost power, which reduced even further the ability of the Boeing 737-200 to climb away.

Under the circumstances, with one engine out and a constant V2 (150 knots), the maximum vertical speed in the climb should have been around 150 feet (46 meters) per minute with the landing gear down. If they’d retracted the landing gear, the vertical speed would have risen to around 450 feet (137 meters) per minute.

The Commission of Inquiry recreated the accident using the flight simulator, using the conditions on the day of the accident. Qualified crew were put through three scenarios where the left engine failed at rotation speed.

During the first scenario, the crew were asked to follow standard procedures. During the second and third scenarios, the crew were told to leave the landing gear down. For the first scenario, they were told to maintain the V2 speed and for the second scenario, they were to maintain a constant pitch of 18°, emulating the captain’s actions that day.

The simulations showed that the vertical climb speed was maintained between 1,000 and 2,000 feet per minute when the pilots maintained the pitch 18° from the moment of the loss of power, even with the landing gear down. However, when the V2 speed was held constant, as it was on the 6th of March, (see John’s comment below) the vertical speed fluctuated between -300 and +300 feet per minute. The stick shaker consistently activated between 135 and 140 knots.

Unsurprisingly, in every test, the pilot monitoring found it difficult to take the controls during the initial climb with one engine out.

3.2 Probable Causes
The accident was caused by the loss of an engine during a critical phase of flight,
the non-retraction of the landing gear after the engine failure, and the Captain, the
PNF, taking over control of the airplane before having clearly identified the
problem.

The following factors probably contributed to the accident:

  • the perfunctory flight preparation, which meant that the crew were not equipped
    to face the situation that occurred at a critical moment of the flight;
  • the coincidence between the moment the failure occurred and the request to
    retract the landing gear;
  • the speed of the event that left the crew little time to recover the situation;
  • maintaining an inappropriate rate of climb, taking into account the failure of one
    engine;
  • the absence of any teamwork after the engine failure, which led to a failure to
    detect and correct parameters related to the conduct of the flight (speed, rate of
    climb, configuration, etc.);
  • the takeoff weight being close to the maximum with a high aerodrome altitude
    and high temperature;
  • the rocky environment around the aerodrome, unsuitable for an emergency
    landing.

Among the recommendation is that Air Algérie and other operators review their Cockpit Resource Management training programs to ensure that crew are aware of the strict respect required for handover procedures and task-sharing. Investigators noted that, although Air Algérie published a flight safety bulletin for personnel a few times a year, with studies of accidents and incidents, the occurrences were almost all based on other airlines and since August 2000 had only ever concerned foreign operators. No event reported by an Air Algérie crew was analysed in the flight safety bulletins.

In addition, the report includes a recommendation that the Algerian Ministry of Transport set up a permanent organisation for the investigation of civil aviation accidents and incidents. Unfortunately, when I tried to access the Transport Ministry to find out if this had happened, the website wasn’t working.

You can read the full accident report on the French BEA site.

14 Comments

  • Impossible not to wonder if gender played a role in the interpersonal dynamics: two men in conversation talking over a woman, even though what she had to say was important; a less-informed male captain taking over from a female Pilot Flying.

    • It does feel like a component, doesn’t it. But then the bad cockpit resource management goes beyond that: ignoring/not completing the briefing, non-essential conversation, and above all the senior cabin crew member was clearly *still* standing in the cockpit as they took off.

      The flight crew certainly were never a team. But there’s also the fact the captain did not appear to have heard his first officer at all when she brought up the landing gear.

      It’s an impossible situation for her, too, because although it is a first officer’s duty to take control if the captain is not managing the situation, what you really can’t do is *take back* control having just had it taken off you. So the first officer really was powerless to do anything at all, in my opinion.

      There’s so many layers that it really ends up becoming an accident that was simply waiting to happen.

      • I am not at all a pilot, and don’t know the correct protocols.

        Would it have been “wrong” but life-saving for the first officer to retract the landing gear without specific approval from the captain? We might be getting into semantics, but she had asked for the landing gear to be raised. After his take-over, she would have been PNF, which gives her the responsibility to raise the landing gear. Justification would be that the pilot flying (herself, just seconds before) had asked for them to be raised.

        And I agree with Abi that gender seems to be a big part of the problem. She seemed to have a better sense of what was happening, but was repeatedly ignored before and after takeoff.

        • As Pilot Monitoring, she absolutely not change the configuration of the aircraft without a direct request from the Pilot Flying. Colgan Air flight 3407 is an example of this, where the Pilot Monitoring retracted the flaps which contributed to the stall/crash of the aircraft.

          I think it is an interesting argument that her request made as Pilot Flying could then be carried out by her as Pilot Monitoring when he took control. However, the point is that as Pilot Flying he needs to know the state of the aircraft and she’s changed it after he’s taken control, so it just sounds really dangerous to me. She did the right thing to bring it up but he was too focused on the pitch (apparently to the exclusion of all else) to take in the information.

          • Thank you for the insight and explanation. I work with computers, where if there are problems, we can generally recreate the initial conditions and go forward a different way. Different built-in instincts.

  • “The first officer never resumed the checklist …” is the “And the call was coming from _inside_ the house …” of the aviation industry.

    Dun dun dun dun dun …

  • I think that Abi has a point, gender probably did play a role. But this whole sordid story does come down to an unbelievably incompetent and probably arrogant captain. The first officer did try and gave up when she was totally ignored.
    Many investigations seem to confirm that a subordinate crew member, when consistently put down, will eventually give up.
    How that captain managed to get through his command training is a question the airline in question – Air Algerie – will have to answer.
    As a btw: It may be a feature on this aircraft but I have never come across an ASI with coloured arcs marking the “V-speeds”.
    They are movable “bugs” that are set after the speeds have been calculated as part of the pre-flight preparations. They are not fixed but vary with the outside air temperature, aircraft weight etc.
    The crew will obtain them, after listening to the ATIS, by consulting the TL (or RTOW) tables for the given type of aircraft and the expected runway for take-off.
    On aircraft with a “glass cockpit” they are set differently, dependent on the sophistication of the instrumentation, but they still are representing the calculated take-off speeds (V1, Vr, V2, Vfr)

    • I’ve revised that because using bugs makes much more sense. I wondered if it was a digital display that changed the colour codes once the speeds were set, which sounds awesome but I have never actually seen.

  • Hi Sylvia,
    Nice write up!

    One small thing niggles me though. In the following lines there seems to be something a little bit off.

    “The simulations showed that the vertical climb speed was maintained between 1,000 and 2,000 feet per minute when the pilots maintained the pitch 18° from the moment of the loss of power, even with the landing gear down. However, when the V2 speed was held constant, as it was on the 6th of March, the vertical speed fluctuated between -300 and +300 feet per minute. The stick shaker consistently activated between 135 and 140 knots.”

    To my knowledge maintaining V2 is correct and maintaining (18 degrees) pitch is not. It would appear that the test flights prove the opposite which seems strange to me..
    Also, in the very last sentence, it states that the crew tried to maintain V2, which they clearly did not. The crew maintained pitch and lost control due to dropping airspeed.

    Indeed a very sad story of lack of CRM and “rabbit caught in the headlight”-effect.

    Cheers,
    Keep up the good work!

    John

    • You are right! Having gone back to the original source, that’s what they seem to have done for the test flights but the refernce that the crew attempted to maintain V2 is clearly wrong and my misreading / losing track of where I was.

      I’ve done a quick fix above. Here’s the exact results:

      • a change in the magnetic heading by about fifteen degrees towards the left after correction by the crew;
      • vertical climb speed between 1,000 and 2,000 ft/min when the pitch is maintained at 18° from the moment of loss of power in the left engine and with the landing gear down;
      • vertical speed fluctuating between – 300 and + 300 ft/min, when the V2 speed is held constant;
      • stick shaker activated between 135 and 140 kt;
      • difficulty for the pilot to take the controls during the initial climb with one engine out, especially after a simulator pause comparable to a transfer of control in a critical situation.
  • … yet another airline to add to our no-fly list : this “” lack of coordination “” (panic) upfront reminds me of the scandalous Asiana/SFO approach of a few years ago – but there extreme relaxation (carelessness) lead to panic and then fatalities …

    • Yes, that’s a good point. I should mention that this was 2003 so quite some time ago. I should really look into their more recent incident.

  • Not gender but I would suggest that the captain did not contribute in the required pre flight activities, checks and briefings. perhaps they were beneath him and he did think any of the protocols could endanger the aircraft so he didn’t need to take any of it seriously. In addition his role surely was to support, guide, advise the pilot in control, his mind seemed to be elsewhere. The captain wasn’t at any time ready or prepared to pilot the aircraft let alone deal with an emergency. Complacency then mechanical engine failure. Rather than there being a gender issue I would suggest a seniority issue.

Post a comment:

Your email address will not be published. Required fields are marked *

*
*
*