Accident Details
Probable Cause and Findings
the failure of the flight crew to adequately monitor the system for proper operation and manually activate the system during the flight in icing conditions. Contributing to the accident was the crew's limited training on the manual operation of the anti-ice system and the nonactivation of the automatic ice detection system for reasons that could not be determined.
Aircraft Information
Registered Owner (Historical)
Analysis
HISTORY OF FLIGHTOn February 5, 2014, about 0015 central standard time, an Embraer 145EP (EMB-145), N802HK, operating as Trans States Airlines flight 3395, was landing on runway 36R at Memphis International Airport (MEM), Memphis, Tennessee, when it suddenly rolled to the right, and the right wing struck the runway. The 44 passengers and crew onboard were not injured, and the airplane sustained substantial damage. The flight was operating under the provisions of Title 14 Code of Federal Regulations Part 121 as a scheduled passenger flight from Houston International Airport (IAH), Houston, Texas, to MEM. Instrument meteorological conditions (IMC) prevailed at the time of the accident.
The accident flight was the flight crew's first flight in the accident airplane that day. The first officer was the pilot flying and the captain was the pilot monitoring (PM). According to the flight crew, the preflight inspection was routine, and the anti-ice system was not tested because it was only required before the airplane’s first flight of the day.
The crew indicated that the departure, climb and cruise phases of flight were uneventful. During the descent, the airplane entered a cloud layer about 3,500 ft mean sea level (msl). Air traffic control (ATC) then vectored the flight to a final approach for an instrument landing system (ILS) approach to runway 36L.
According to the crew, before reaching the final approach fix (FAF), they received intermittent localizer course indications on both primary flight displays during the approach. Neither crewmember recalled observing any indications of icing during this approach, and the captain said he noticed no engine indicating and crew alerting system (EICAS) messages for ice.
Inside the FAF, the crew stated the localizer course indications were intermittent again, and they elected to execute a missed approach. According to the aircraft performance study, at 2353, when the go-around was initiated, the airplane was at an altitude of about 2,000 ft. The airplane subsequently climbed to 3,000 ft during the go-around before it returned to and stayed at an altitude of 2,000 ft for about 12 minutes. The study indicated that during the go-around, the airplane spent an additional 19 minutes at an altitude with an increased probability of icing.
The crew notified ATC of the localizer course difficulty they were experiencing on the approach, and the captain requested vectors to the ILS approach to runway 36C. That runway was not available, and ATC cleared the airplane for the ILS approach to 36R.
The captain indicated in a postaccident interview that, while the airplane was level at 2,000 ft, on the base leg to runway 36R, the flight entered clouds. The first officer stated that, near the FAF for the ILS approach to runway 36R, she noticed moisture on the windshield wiper and observed something on the windshield; the captain stated that the wind screen was wet. About 0011, the cockpit voice recorder (CVR) recorded the first officer stating, “are we getting’ ice now,” and the captain replied, “a little bit.” The airplane's ice protection system was in the automatic mode and did not activate nor did the crew manually activate the system. The first officer indicated in a postaccident interview that they did not see the ice light come on and there were no icing messages on the EICAS.
At the FAF, the localizer course signal was uninterrupted, and the airplane's autopilot captured the course. About 0012, the CVR recorded the crew discussing the airplane being configured with landing gear down and 45° of flaps; the captain indicated in a postaccident interview that it was on a stabilized approach at 1,000 and 500 ft. The crew continued the ILS approach, and near the approach minimums (about 400 ft above ground level [agl]), the airplane exited the clouds and the crew observed the landing runway. FDR data indicate that, during the final descent to the runway, the airplane's speed reduced to 130 kts. According to the captain, the first officer announced "landing" and disconnected the autopilot using the control yoke switch when the airplane was about 300 ft above the runway.
After the autopilot was disconnected, the CVR recorded numerous “autopilot” audio messages in the cockpit. The first officer indicated in an interview that she attempted to turn off the audio message but was unable to do so. The CVR recorded the first officer asking, “Why is she not shutting up?” The captain stated that he then held the quick disconnect button for the autopilot to silence it. After the autopilot disengagement, the captain gave a speed warning to the first officer. He stated in an interview that the first officer "got a little slow" (between 5 to 6 kts) during this time and estimated the airplane to be about 100 to 150 ft agl but that the first officer called “correcting” and “got back on speed.” The first officer recalled being slow by about 4 kts.
According to the crew, when the airplane was about 20-40 ft agl, as the first officer was applying control inputs to adjust for a crosswind, a rapid roll to the right occurred. According to the performance study, the airplane began to roll quickly to the right just before 0015, followed by stick shaker activation. One of the airplane's two angle of attack (AOA) sensors reached 15°, and the indicated airspeed was 113 kts; the airplane's maximum roll attitude, which coincided with the stick shaker activation, was 28° right wing down. The airplane's wing struck the runway, and the airplane landed hard on the right side of the runway. The CVR recording and postaccident crew statements indicated that the pilots believed that the sudden roll to the right was caused by a rudder hardover. However, FDR data did not show any evidence of a large rudder surface deflection.
About 40 minutes after the airplane landed, taxied to, and arrived at the gate, the crew observed ice on the wings, horizontal stabilizer, and both engine inlets. The first officer indicated that the airplane was covered in ice, and the captain indicated observing “a lot of ice” on the leading edge of the wings.
PERSONNEL INFORMATIONAll crewmembers were current and qualified in accordance with Federal Aviation Administration (FAA) regulations and Trans States Airlines requirements.
The Captain
The captain was hired by Trans States Airlines in August 2005. A review of the captain's FAA records did not reveal any prior accidents, incidents, or enforcement actions, and a review of his driving records showed no revocations or suspensions. He held an Airline Transport Pilot certificate and had accumulated 6,400 hours of total flight experience, of which about 5,600 was in the Embraer 145.
The captain was based in St. Louis, Missouri, and was on the first day of a 5-day reserve period. The accident flight occurred on his fourth leg of the day. The captain did not recall what time he awoke on February 4 but stated that he had normal sleep and felt rested. He said he departed his home in St Louis about 1115 and checked in for duty about 1300 for a scheduled departure of 1345. On February 3, he was at his residence and did not recall what time he awoke but spent the day at home. He did not recall what time he went to bed but stated that he slept "well." On February 2, he had returned from vacation about 1100 and said he had normal sleep that night.
The First Officer
The first officer was hired by Trans States Airlines in September 2012. A review of FAA records did not reveal any prior accidents, incidents, or enforcement actions, and a review of her driving records showed no revocations or suspensions. She held an Airline Transport Pilot certificate and had accumulated 930 flight hours as second in command in the Embraer 145.
The first officer was also based in St. Louis. The accident flight was her third leg of the day. She indicated that, on February 4, she awoke about 0900 and had slept "well" and that, on February 3, she awoke about 0530 and slept "well." She flew from MEM to Chicago, Chicago to Moline, and Moline to Chicago. She checked into the hotel for a 27-hour layover and went to bed about 2200. On February 2, she commuted in the night prior. She indicated that she awoke about 0400 and had slept well. She flew two flight legs. Her duty day ended at 1100, and she was in bed about 2000.
AIRCRAFT INFORMATIONIce Detection System
The ice detection system is used to detect and alert the crew about the formation of ice. It is the primary source to automatically activate the airplane’s anti-icing systems for the wings and engines. The ice detection system is comprised of two identical ice detection circuits that operate independently during all phases of flight. Each ice detection system circuit has the following components: ice detector, ice detector relay, ice protection overhead panel, circuit breakers, and data acquisition unit.
The ice detector is a one-piece unit including the sensor and processing electronics that detect the presence of ice. When either of the ice detectors detect ice, an advisory message "ICE CONDITION" is shown on the EICAS display, a signal would be sent to the anti-ice system valves to activate them to open, and a signal would be sent to the full authority digital engine control to activate the automated engine icing thrust setting logic to limit the thrust to a minimum acceptable level. The icing signal stays active for 60 seconds. At the same time the icing signal is activated, the ice detector heaters are turned on to deice the detector strut and probe. When the sensing probe is deiced, it is ready to sense ice again. If the icing condition continues and the ice thickness switching level is reached before 60 seconds has passed, the icing signal is continuous. All anti-ice functions operate when one or both detectors detect ice.
The ice detectors are self-monitored through built-in test circuits. A detected failure in either detector would cause a change to the status output ...
Data Source
Data provided by the National Transportation Safety Board (NTSB). For more information on this event, visit the NTSB Records Search website. NTSB# DCA14FA058