New Orleans, LA, USA
N409UA
Airbus A320
According to flight data recorder (FDR) data, the Avionics Smoke warning was active at the time the recording began. Since the caution was inactive at power up, it was most likely caused from contaminants detected before the airplane was powered up. Based on this, when the crew arrived at the airplane, they should have had three primary cues alerting them of an Avionics Smoke event, including: a master caution light illuminated amber; an amber AVIONICS SMOKE warning on the upper Electronic Centralized Aircraft Monitor (ECAM); and Blower and Extract fault lights and Gen 1 Line smoke light illuminated amber on the overhead panel. In addition, when they viewed the status page of the ECAM (as required per the captain's Cockpit Preparation checklist), VENT BLOWER and VENT EXTRACT would have been listed under inoperative systems. It is unlikely that airline personnel would have cleared or canceled this warning without communicating this information with the crew, and the crew stated that they did not cancel the warning. Had the warning been inadvertently cleared or cancelled, the overhead panel lights would have remained illuminated and vent blower and vent extract would have remained inoperative systems. Because the cockpit voice recorder (CVR) did not contain any discussion related to any Avionics Smoke event while on the ground, or after takeoff prior to about 1500 feet (during which time both crewmembers responded with surprise), it is unlikely that the crew had previously seen the warning but purposefully ignored the available cues. It is also possible that the crew did not see the cockpit indications since the captain did not complete at least one step of his cockpit preparation checklist–pushing the recorder ground control switch. Because of this, the CVR and FDR did not begin until the time that the APU started, when it should have started much sooner in the sequence of preflight events. So, the investigation was not able to determine whether the crew completed other checklist items that should have alerted them to the Avionics Smoke warning. Finally, investigators were unable to find any condition in which the caution could be recorded on the FDR but not displayed to the crew. Therefore, although the incident flight crew was not aware of the Avionics Smoke event prior to takeoff, investigators could not determine the reason for this. At 7:10:08.7, the captain began the after takeoff checklist. Item 3 of that checklist is "ECAM memo…checked." When completing this step, the captain detected the Avionics Smoke event on the upper ECAM. Primary cues available were the Avionics Smoke procedure and an amber LAND ASAP message. Although the crew were surprised when they noticed the alert message, there was no corresponding master caution aural warning during this time, which confirms that the Avionics Smoke alert had been active prior to takeoff. The captain then delegated crew duties, assigning the first officer as the pilot flying and indicated that he would complete the ECAM. At 07:10:30.8, the captain began the Avionics Smoke ECAM procedure and stated, "perceptible smoke", referring to the first conditional statement of the procedure "if perceptible smoke". Airbus stated that completion of the Avionics Smoke procedure is dependent on "direct detection by the crew [and] secondary detection by a detector which is considered as a help." Detection by crew can be by sight or smell. According to the procedure, "If perceptible smoke" is a conditional statement and if the crew did not detect smoke, they were not to continue the procedure. After the incident the first officer stated that if Avionics Smoke was detected by the sensor, then there was Avionics Smoke and he was not going to question that. About 38 seconds after the flight crew became aware of the Avionics Smoke warning, the captain stated, "hey you lost your autopilot too." The FDR indicated that the crew received an autothrust message. CVR data suggests that the captain became very apprehensive about the situation. The flight crew concluded that the failure of the autopilot meant that their situation was deteriorating and they needed to land the airplane promptly. Likely adding to the captain's apprehension was the LAND ASAP [i.e. as soon as possible] message displayed on the ECAM. Although an amber LAND ASAP message was presented, discussions with UAL instructors and pilots indicated that, to a pilot, land ASAP means land ASAP, regardless of color. After the incident, the captain stated that during his last proficiency training session, in-flight fires were emphasized. Specifically, pilots were told that delaying landing by a few minutes could be the difference between a successful landing and loss of an aircraft, such as Swissair [flight 111, that occurred September 2, 1998] and Valujet [flight 592, that occurred May 11, 1996]. The captain said, "he did not want this to be the next Valujet." The captain continued with the Avionics Smoke procedure but did not do so with the necessary thoughtfulness and made several missed steps. For example, the procedure states action item "EMER ELEC PWR…MAN ON" followed by the conditional statement "WHEN EMER GEN AVAIL:" and action item "GEN 2…OFF". In this instance, the captain should have turned on the emergency electrical power (i.e., deployed the RAT13), and then when emergency generator power was available turned off generator 2. Data show that the captain did not manually deploy the RAT prior to turning off generator 2. As a result, when generator 2 was turned off prematurely, there was a brief disruption in the power supply and the airplane entered the emergency electrical configuration. The EMER ELEC procedure and a red LAND ASAP message appeared on the ECAM. This configuration caused the RAT to automatically deploy which restored electrical power to the airplane after about 6 seconds. The airplane remained in the emergency electrical configuration. Therefore, the captain became apprehensive about the Avionics Smoke event and hastily performed the ECAM procedure resulting in the airplane entering the emergency electrical configuration. At 7:12:51.5, the first officer alerted the captain that he had no instruments. Two seconds later the captain took control of the airplane and told the first officer to call the flight attendants. The flight crew did not adequately transfer control of the airplane – the first officer did not brief the captain on the status of the airplane and the captain did not brief the status of the emergency procedures. Over the next two and a half minutes the crew focused primarily on contacting the flight attendants and did not discuss completing the EMER ELEC procedure. About 30 seconds later, the flight crew lowered the landing gear without restoring power to the airplane, per the EMER ELEC procedure, and the airplane began operating on battery power. As a result, the CVR recording ended and no further communications in the cockpit were available with the exception of ATC communications. Completion of the EMER ELEC procedure would have restored power to generators 1 and 2 prior to landing gear extension and maintained electrical power to the airplane. After the incident, the captain said when they lowered the landing gear, operating on battery power was not on his mind. After touchdown, reverser 2 did not deploy, and the airplane veered to the left and exited the runway. The flight crew was not aware that reverser 2 was an inoperative system based on the electrical configuration of the airplane. Had the first officer checked the ECAM status per the Approach Descent Checklist, the inoperative system would have been identified. However, this was not completed likely due to the time constraints. After landing, engine status cues would have alerted the first officer that reverser 2 did not deploy and he should have informed the captain. While it is unknown if the first officer monitored engine status and made the required reverser call out after landing, staff believes it is unlikely because the captain stated in a post incident interview that the airplane departed the runway because of a crosswind. Therefore, the flight crew became distracted by the emergency and focused on landing the airplane without completing necessary checklist items, resulting in the airplane operating on battery power and partial loss of reverse thrust on landing. It is the captain's responsibility as a leader to set the tone in the cockpit for the entire flight, and this is even more critical when a crew is faced with an abnormal situation. CVR data suggests the tone in the cockpit was very casual. For example, prior to performing the before takeoff checklist, the first officer asks the captain "ready to read em and weep?" And just before takeoff, the first officer stated, "let's get…outta here man." The captain then stated "Brakes released. You got it man. Throttles yours. Whatever you want to do." The casual tone in the cockpit during preflight activities and the taxi did not support the creation of a functional team environment conducive to the crew's subsequent attempts to resolve the abnormal situation. This was manifested in the crew's undisciplined management of the situation in that they failed to adequately assess and understand the situation they were presented with. For example, as the captain completed the after takeoff checklist, he noticed the Avionics Smoke warning on the ECAM; however he failed to announce what the warning was. Instead, he delegated the first officer to fly the airplane and stated he would complete the ECAM. There was no discussion between crewmembers about the situation they were faced with. It is not clear if the first officer was aware of what the warning on the ECAM was. Once the airplane entered the emergency electrical configuration, the captain stopped managing the emergency and the crew's coordination deteriorated further. After the captain stated they were in emergency electrical configuration, the first officer stated "yup confirm. Let's go back." Had the captain been properly managing the abnormal, and now emergency, situation, he should have made the decision to return to the airport rather than the first officer making that decision. In addition, the captain abandoned the EMER ELEC procedure and his pilot monitoring duties. He made radio calls to ATC requesting vectors back to the airport and declaring an emergency. Upon recognizing that the first officer did not have any instruments, the captain assumed control of the airplane. At no point did he delegate the first officer to complete the EMER ELEC procedure but only to inform the flight attendants of the emergency. Completing the EMER ELEC procedure would have resulted in power restoration prior to lowering the landing gear and maintained full use of reverse thrust on landing. After the incident, the first officer stated he did not feel that he had time to be aware of the captain's actions when acting as the pilot flying and said he "took for granted" that the captain completed the ECAM procedure. Finally, during the transfer of flight duties from the first officer to the captain, the first officer stated, "I got the radios", however, subsequent radio communications were made by both crewmembers. Therefore, the captain's failure to set the tone in the cockpit and ineffective management of the emergency resulted in neither crewmember fully understanding the situation they were faced with and subsequent escalation of an abnormal situation to an emergency. The investigation evaluated a number of criteria to determine the extent to which fatigue impacted the flight crew's performance during the incident flight including circadian factors, sleep length, acute or chronic sleep loss, and time since awakening. Based on the pilots' schedules and normal sleeping habits the incident occurred at a time when melatonin is low and body temperature is rising. Therefore, the investigation did not identify any risk of circadian factors in the incident. Investigators also evaluated the flight crew's sleep in the few nights prior to the incident. The captain and the first officer received more than the minimum required rest periods during their trip pairing in the days before the incident, and their flight and duty times in the week and month before the incident would not have precluded them from obtaining adequate sleep. However, both crew members complained of smog, heat and smell during their 29-hour layover in Mexico City on April 2, 2011, which gave them headaches and required use of over the counter pain relief medication. They did not report any difficulties sleeping in Mexico City. The night prior to the incident, the captain obtained about 7.5 hours of sleep, although he said he normally slept about 5 hours per night. This could suggest that he had a sleep debt from previous night's rest that he was trying to overcome, however, staff has no additional information to support that the captain was experiencing a sleep debt. The first officer obtained about 5.5 hours the night before the incident. Although the first officer said he felt rested on the morning of the flight, he received about 1.5 hours less sleep than he normally obtained. Although it is possible that the first officer was experiencing a small acute sleep debt on the morning of the incident, there is no evidence to suggest that this affected his performance during the flight. CVR data indicates he was alert and performed his duties per the captain's delegation and even took an assertive role in deciding to return the flight to the airport. Furthermore, there was no discussion about being tired or yawning heard on the CVR. Neither pilot ate breakfast, but both drank coffee prior to the incident flight. At the time of the incident, the captain had been awake about 2.5 hours and the first officer had been awake about 1.5 hours at the time the Avionics Smoke event was recognized. This was ample time for the body to adjust to being awake and for the crew to maintain alertness and does not believe time since awakening was a factor. Therefore, although it is possible that the first officer was experiencing an acute sleep debt, there is no evidence to suggest that this affected his performance during the flight. His performance was more consistent with poor leadership from the captain and the establishment of a casual tone in the cockpit.
HISTORY OF FLIGHT On April 4, 2011, at about 0725 central daylight time (CDT), United Airlines flight 497, an Airbus 320-232, N409UA, departed the left side of runway 19 while conducting an emergency landing due to an avionics smoke warning at the Louis Armstrong New Orleans International Airport (MSY), New Orleans, Louisiana. An emergency evacuation was conducted. There were no injuries to the 104 passengers or five crew members and the airplane sustained only minor damage. The flight was a 14 Code of Federal Regulations part 121 regularly scheduled passenger flight and had originally departed MSY destined for San Francisco International Airport, San Francisco, California. The first officer (FO) was the pilot flying for the flight and the captain was the pilot monitoring. According to flight crew statements and recorded data, the incident flight takeoff began at 0708. At about 0710:10, the cockpit voice recorder (CVR) recorded the captain began the after takeoff checklist and stated, "…wait a minute what do we got here." The captain then states, "okay ECAM I got the uh— uh you got the jet. I got this." The CVR then records the captain beginning the Avionics Smoke ECAM checklist procedure. The FO stated that about that time he became aware of an "avionics smoke" warning electronic centralized aircraft monitoring (ECAM) message and the captain stated that he noticed a "yellow" autothrust ECAM message. The FO reported he pushed the autothrust (ATHR) button on the mode control panel (MCP), but this did not succeed in re-engaging the autothrust. The captain said the autothrust message was followed by a red "LAND ASAP" ECAM message accompanied by the electrical page synoptic display and the "AVIONICS SMOKE" ECAM procedure. The FO leveled the aircraft at 5,000 feet in instrument meteorological conditions (IMC) and retarded the thrust levers to slow the aircraft. The captain stated that the first item on the avionics smoke ECAM procedure was to don oxygen masks, but that he and the FO agreed not to don the masks because there was no smell of smoke. The captain did not recall seeing any conditional statements or a timer in the ECAM procedure. At about 0711:22, the captain stated "line one off" followed by "okay emergency electrical power man on when uh emergency generator available. gen two off." Concurrent with the second generator being switched off, the flight data recorder stopped recording and there was about a six second power interruption of the CVR. When the CVR resumed recording, the captain is discussing the ram air turbine (RAT) and that the airplane is in emergency electrical configuration. The FO then lost his flight instrument displays and began using the captain's primary flight display for altitude, heading and airspeed. A short time later, the captain took control of the airplane. The FO said that once the captain had taken control of the airplane, he stated that he thought they were in the emergency electrical configuration but he didn't know what the ECAM steps were for that configuration. He said he did not follow up on the ECAM actions, and he did not use the Quick Reference Checklist (QRC) while in flight. The first officer then tried to alert the flight attendants (FA), but he said he "did not hear the bell" when he called and the FAs did not respond. He used the pedestal handset to call the FA's, but he got no response. He then opened the cockpit door and told the FAs that they were in an emergency and would be landing immediately. The pilots both stated in interviews that they did not conduct an approach briefing, tune the navigation radios, enter an approach in the FMGC, check the ECAM status page, use the flight manual to determine what systems were affected or lost, conduct an approach descent checklist, conduct an overweight landing checklist, determine the applicable approach speed or landing distance from the FMGC or flight manual, or attempt to repower the electrical system. The captain later said that there was no time to do these things because of the severity of the emergency. The captain said he knew he would be in direct law with no antiskid or nose wheel steering, but he did not recall telling the first officer this. He estimated that the approach speed should be 160 knots, based on the takeoff V2 of 147 knots, rounded up for wind additive. He said that normal landing distance was 3,500 to 4,000 feet and was confident the airplane would stop on the runway. Both pilots stated that they wanted to keep the approach speed above 140 knots in order to avoid stalling the ram air turbine (RAT). The captain told ATC that they would need a vector back to the airport and requested "the longest runway." ATC advised that runway 10, which was 10,104 feet long, was still closed due to the equipment on the runway, but that airport personnel were attempting to clear the runway. At about 07:16:03, as the flightcrew lowered the landing gear, the CVR stopped recording. The captain stated that he could hear the tower on the radio talking to the operations personnel working on runway 10 and he realized that they would not be able to clear that runway in time for the flight to land. The captain then told ATC "we've lost all our instruments, we need a PAR." The captain stated during his interview that he had attitude and compass information but no localizer, and that the screens started to fade during the approach. ATC told the flight that they would provide a no-gyro surveillance approach. The captain did not advise the FO as he continued the descent through the cleared altitude of 2000 feet. The FO stated that he did not recall if they were cleared to descend out of 2000 feet. According to the ATC recording, the crew reported they "we're at 1000 feet now and we've got water contact, where are we from the airport?" The controller replied that he was at 330 degrees from the airport and said on their present heading they would be "set up for the shoreline 19." After some additional communication with ATC, the captain said "I've got it" and the controller replied "wind 180 at 16 gusts to 20, cleared to land." The captain stated that he landed with full flaps and used the PAPI for vertical guidance. The captain stated that he landed on the centerline, approximately 1,500 feet down the runway. He also state that he "got on the brakes," used full reverse, and used right rudder to keep the airplane in the center of the runway. The FO stated that on touchdown the cockpit door swung open and that he turned and shouted "remain seated, remain seated." He said that when the engines came out of reverse the captain went to the tiller but the aircraft pulled to the left. The captain had said "I can't control it," and "we're going to evacuate." The airplane continued to veer to the left and the captain indicated he stepped harder on the right brake pedal. The airplane departed the left side of runway 19 approximately 5000 feet from the threshold at a low speed, and the nose gear sank into the soft groundoff the side of the runway. Once the aircraft came to a stop, the flightcrew used the QRC to conduct the evacuation. The captain set off the evacuation signal and the first officer then silenced it. The first officer said the 1R slide did not inflate and he yelled "go the other way." He went down the 1L slide and began helping the passengers get away from the airplane. The captain came out a few minutes later with a megaphone, which he used to direct the passengers away from the aircraft. INJURIES TO PERSONS: There were no injuries to the 104 passengers, which included 3 children and 1 infant lap child, two flight crew, or three flight attendants. DAMAGE TO AIRCRAFT: Post incident inspection of the airplane revealed no evidence of fire or smoke in the avionics compartment, ducts, or anywhere on the airplane. The nose landing gear bracket and weight-on-wheels switch assembly was broken and an electrical conduit in the area of the nose landing gear received minor damage. The right main landing gear tires were found deflated. The tread of the outboard tire had two flat spots, one of which was an open hole. The inboard tire (#3) was found deflated, the beads of the tire were not mounted on the wheel halves, and heavy abrasion was found to one tread shoulder. PERSONNEL INFORMATION: The captain, age 50, was hired by United Airlines on July 17, 1995. He held an Airline Transport Pilot certificate, multi-engine land, with type ratings in A320, B737, B747-4, B767, B777, BAE125, CE500, and HS125. He held an FAA first class medical certificate with limitations: must have glasses for near and far vision. Company records indicate that he had 15,000 hours total time with 1,487 hours on the A320. He had no previous accidents, incidents, or violations. The incident flight was the first flight of the day for the Captain. The first officer, age 51, was hired by United Airlines on April 13, 1998. He held an Airline Transport Pilot certificate, multi-engine land, with type ratings in the A320, B757, B767, and BA3100. He held an FAA first class medical certificate with the following limitations: must wear lenses for distant vision, must have glasses for near vision. He reported a total of 11,500 flight hours, with 1,154 hours in the A320. He had no previous accidents, incidents, or violations. The incident flight was the first flight of the day for the FO. The three flight attendants were all current and qualified on the Airbus A320. AIRCRAFT INFORMATION: The incident airplane was an Airbus Industries A320-232, registration N409UA, manufacturer serial number 462, and was equipped with two IAE Aerospace V2500 turbofan engines and an Auxiliary Power International Corporation (Hamilton Sundstrand Company) Auxiliary Power Unit (APU). The airplane was purchased new by United Air Lines, Inc. from Airbus Industries on March 21, 1994. The airplane had accumulated 58,253:02 flight hours and 21,414 cycles at the time of the incident. The electrical system generates and distributes AC and DC power to airplane systems. Three generators provide AC power, two batteries provide power for APU start and for emergency power, and an emergency generator provides AC power in the event all three main generators fail. System operation is normally automatic. An electrical control panel, located in the center part of the overhead panel, includes generator on/off push button switches for generator one, generator two, the APU generator, and for each of the two batteries. An emergency electrical power control panel, located on the left side of the overhead panel, includes GEN LINE 1, RAT & EMER GEN, and MAN ON push button switches. Smoke detectors are located in the lavatories, avionics compartment, and cargo compartments. The avionics compartment detector is a Cerberus Model CG7GO ionic smoke detector, and is installed on the air extraction duct of the avionics ventilation system. The ionic type of smoke detector was installed in the A320 up to MSN number 1540. A320 airplanes following those line numbers and on all A318 airplanes, the ionic type of detector was no longer installed, and optical smoke detectors were installed. The A320 series airplanes cool the avionics compartment with outside air when the airplane is on the ground. When avionics smoke is first detected, the detection is indicated by an aural single chime (SC), the illumination of the SMOKE light on the EMER ELEC PWR panel, the illumination of BLOWER and EXTRACT FAULT lights on the VENTILATION panel and MASTER CAUTION lights, and by an ECAM caution on the Engine and Warning Display (E/WD). Five minutes after smoke is detected in the avionics compartment, the caution "latches" if the detection is still active, and the caution remains in effect until the warning system is reset. If avionics smoke is detected, ECAM directs the airplane be placed in the smoke configuration. This requires placing the GEN 1 LINE switch on the EMER ELEC PWR panel to OFF. When this occurs, the GEN 1 line contactor opens, GEN 2 powers AC busses 1 and 2 (main galley bus is shed) through the tie bus, and GEN 1 powers a fuel pump in each wing tank. This permits removing AC power from all busses during the AVIONICS SMOKE ECAM procedure without removing power from the fuel pumps. An amber SMOKE light is located in the upper half of the GEN LINE 1 switch, and illuminates when there is an avionics smoke warning. When pushed, the GEN LINE 1 switch stops the generator from supplying power to its normal buses, but continues to provide power to a fuel pump in each wing. A red FAULT light is located in the upper half of the RAT AND EMER GEN switch which illuminates red when AC busses 1 and 2 are lost and the emergency generator is not supplying power and the nose gear is retracted (some A320's). The MAN ON switch is a red guarded switch. It is normally left in the guarded AUTO position, which will cause the RAT to extend automatically with the loss of AC busses 1 and 2 and speed above 100 knots. When the guard is opened and the ON position is selected, the RAT extends and couples the emergency generator to the electrical system The ram air turbine automatically extends if both AC busses 1 and 2 lose electrical power above 100 knots; however, a minimum airspeed of 140 knots is required to provide sufficient blue hydraulic pressure to operate the emergency generator. The RAT pressurizes the blue hydraulic system, which powers the emergency generator via a hydraulic motor. A generator control unit (GCU) controls the generator speed, voltage, generator line contactor, and start-up. The emergency generator is automatically coupled to the electrical system after RAT extension is complete. On the incident aircraft, the emergency generator operates only with the landing gear retracted; if the landing gear is extended and AC busses 1 and 2 lose power, the emergency generator does not operate until the nose gear is retracted and the EMER ELEC PWR MAN ON switch is selected ON. If the nose gear is extended after RAT deployment, the emergency generator drops off line and cannot be manually recoupled until the nose gear is retracted. After the emergency generator drops off line, power is transferred to the batteries. The RAT can be stowed only on the ground. According to the "Emergency Generator/Battery Powered Equipment" contained in the UAL Flight Manual, the upper ECAM is displayed when the aircraft is on emergency generator power or on battery power, except after speed falls below 50 knots. The captain's and FO's audio control panels (ACP), the cabin intercommunication data system (CIDS), the interphone, the left cockpit loudspeaker, the captain's primary flight display (PFD), the ECAM control panel, the brake pressure indicator, and the cabin and cockpit emergency lights operate normally on either the emergency generator or battery power. The captain's Flight Management Guidance Computer (FMGC), the captain's multipurpose control display unit (MCDU), the captain's navigation display (ND) and the #1 flight augmentation computer (FAC) are powered by the emergency generator but not by battery power. Two batteries, BAT 1 and BAT 2, are installed. They are not connected in parallel, and each is connected directly to its respective HOT BAT bus. In addition to powering their respective HOT BAT busses, the batteries can also supply power for APU starting, operation of the static inverter (used in the emergency electrical configuration), and powering the DC ESS bus. In flight, battery endurance is approximately 22 minutes. In-flight, the batteries are the only source of electrical power when the generators, including the RAT-driven emergency generator, are not available. This includes flight with the nose landing gear extended and in during the approximately 8 seconds required from initiating extension of the RAT until the emergency generator is on-line. The ECAM system presents airplane engine and system data on two identical CRT displays located on the center instrument panel below the glareshield. The upper screen is the engine/warning display (E/WD) and the lower screen is the system display (SD). The E/WD has priority over the SD. If the upper ECAM scr
the captain's failure to properly recognize and manage the abnormal condition, resulting in it escalating to an in-flight emergency.
Source: NTSB Aviation Accident Database
Aviation Accidents App
In-Depth Access to Aviation Accident Reports
