Crash During Landing Executive Airlines (doing business as American Eagle) Flight 5401 Avions de Transport Regional 72-212, N438AT San Juan, Puerto Rico May 9, 2004 Aircraft Accident Report NTSB/AAR-05/02 PB2005-910402 Notation 7650A National Transportation Safety Board Washington, D.C. THE CORRECTION BELOW IS INCLUDED IN THIS VERSION OF THE PUBLISHED REPORT AIRCRAFT ACCIDENT REPORT NTSB/AAR-05/02 (PB2005-910402) Crash During Landing Executive Airlines (doing business as American Eagle) Flight 5401 Avions de Transport Regional 72-212, N438AT San Juan, Puerto Rico May 9, 2004 • Page number 100.1 between 100 and 101 in Appendix B has been added. (17 November 2005) The page was originally missing from the report. Aircraft Accident Report E SA NTSB/AAR-05/02 PB2005-910402 Notation 7650A Adopted September 7, 2005 R A N S PO FE RI P LU B US UNUM T Y B OA AT I O N NATI ON LT RT A Crash During Landing Executive Airlines (doing business as American Eagle) Flight 5401 Avions de Transport Regional 72-212, N438AT San Juan, Puerto Rico May 9, 2004 R D National Transportation Safety Board 490 L’Enfant Plaza, S.W. Washington, D.C. 20594 National Transportation Safety Board. 2005. Executive Airlines (doing business as American Eagle) Flight 5401, Avions de Transport Regional 72-212, N438AT, San Juan, Puerto Rico, May 9, 2004. Aircraft Accident Report NTSB/AAR-05/02. Washington, DC. Abstract: This report explains the accident involving Executive Airlines (doing business as American Eagle) flight 5401, an Avions de Transport Regional 72-212, which skipped once, bounced hard twice, and then crashed at Luis Muñoz Marin International Airport, San Juan, Puerto Rico. Safety issues discussed in this report focus on flight crew performance, the lack of company bounced landing recovery guidance and training, and malfunctioning flight data recorder potentiometer sensors. Safety recommendations concerning these issues are addressed to the Federal Aviation Administration. The National Transportation Safety Board is an independent Federal agency dedicated to promoting aviation, railroad, highway, marine, pipeline, and hazardous materials safety. Established in 1967, the agency is mandated by Congress through the Independent Safety Board Act of 1974 to investigate transportation accidents, determine the probable causes of the accidents, issue safety recommendations, study transportation safety issues, and evaluate the safety effectiveness of government agencies involved in transportation. The Safety Board makes public its actions and decisions through accident reports, safety studies, special investigation reports, safety recommendations, and statistical reviews. Recent publications are available in their entirety on the Web at . Other information about available publications also may be obtained from the Web site or by contacting: National Transportation Safety Board Public Inquiries Section, RE-51 490 L’Enfant Plaza, S.W. Washington, D.C. 20594 (800) 877-6799 or (202) 314-6551 Safety Board publications may be purchased, by individual copy or by subscription, from the National Technical Information Service. To purchase this publication, order report number PB2005-910402 from: National Technical Information Service 5285 Port Royal Road Springfield, Virginia 22161 (800) 553-6847 or (703) 605-6000 The Independent Safety Board Act, as codified at 49 U.S.C. Section 1154(b), precludes the admission into evidence or use of Board reports related to an incident or accident in a civil action for damages resulting from a matter mentioned in the report. iii Aircraft Accident Report Contents Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii 1. Factual Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 History of Flight . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Injuries to Persons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.3 Damage to Airplane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.4 Other Damage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.5 Personnel Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.5.1 The Captain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.5.2 The First Officer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 1.5.2.1 The First Officer’s Medical History and Prescription Drug Use . . . . . . . . . . . . . 7 1.5.3 The Flight Attendants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1.6 Airplane Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1.6.1 Pitch Control System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 1.6.2 Landing Gear System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 1.6.3 Cockpit Seats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 1.7 Meteorological Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 1.7.1 Airport Weather Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 1.7.2 Additional Wind Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 1.8 Aids to Navigation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 1.9 Communications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 1.10 Airport Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 1.11 Flight Recorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 1.11.1 Cockpit Voice Recorder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 1.11.2 Flight Data Recorder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 1.11.2.1 Validation of Flight Data Recorder Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 1.11.2.2 Aileron Surface Position Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 1.12 Wreckage and Impact Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 1.12.1 General Wreckage Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 1.12.2 Fuselage, Wings, and Engines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 1.12.3 Landing Gear System and Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 1.12.4 Elevator and Rudder Control Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 1.12.5 Cockpit Seats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 1.13 Medical and Pathological Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 1.14 Fire . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 1.15 Survival Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 1.15.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 1.15.2 Evacuation of Passengers and Crewmembers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 1.15.3 Emergency Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 1.16 Tests and Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Contents iv Aircraft Accident Report 1.16.1 Airplane Performance Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 1.16.2 Air Traffic Control Radar Data Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 1.16.3 Cockpit Seat Assembly Metallurgical Examinations . . . . . . . . . . . . . . . . . . . . . . . 23 1.16.3.1 Accident Airplane Cockpit Seat Assemblies . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 1.16.3.2 FAA Vertical Drop Test of ATR-42 Cockpit Seats . . . . . . . . . . . . . . . . . . . . . . 24 1.17 Organizational and Management Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 1.17.1 Flight Crew Training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 1.17.1.1 Simulator Flight Training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 1.17.1.1.1 Observations of Simulator Sessions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 1.17.1.2 Crew Resource Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 1.17.1.3 Bounced Landing Recovery Training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 1.17.2 Operational Guidance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 1.17.2.1 Approach Airspeed Guidance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 1.17.2.2 Before Landing Checklist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 1.17.2.3 Evacuation Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 1.18 Additional Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 1.18.1 Additional Information About Ipeco Cockpit Seats . . . . . . . . . . . . . . . . . . . . . . . . 31 1.18.1.1 General Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 1.18.1.2 Ipeco Cockpit Seat Tensile Strength Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 1.18.1.3 Ipeco’s Postaccident Actions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 1.18.2 Previous Bounced Landing Recovery Guidance and Training-Related Safety Recommendation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 1.18.3 Previous Flight Data Recorder Potentiometer Sensor-Related Safety Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 2. Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 2.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 2.2 Accident Sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 2.3 Bounced Landing Recovery Guidance and Training . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 2.4 Quality of Data Provided by Flight Data Recorder Potentiometer Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 2.5 The First Officer’s Medical Condition and Prescription Drug Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 2.6 Uncoupling of the Pitch Control Uncoupling Mechanism . . . . . . . . . . . . . . . . . . . . . . 43 2.7 Failure of the Left Main Landing Gear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 3. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 3.1 Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 3.2 Probable Cause . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 4. Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 5. Appendixes A: Investigation and Public Hearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 B: Cockpit Voice Recorder Transcript . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 v Aircraft Accident Report Figures 1. Ground track of the accident flight. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1a. 2. Altitude profile of the accident flight. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2. 3. Main landing gear assembly.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 3. 4. Ground scrape marks and the main wreckage. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 4. 5. Interior configuration of the airplane. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 5. 6. Schematic of the cockpit seat assembly. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 vi Aircraft Accident Report Abbreviations AFFF aqueous film forming foam ALAR approach and landing accident reduction AOM Airplane Operating Manual ARFF aircraft rescue and firefighting ASOS Automated Surface Observing System ATC air traffic control ATIS automatic terminal information service ATR Avions de Transport Regional BEA Bureau d’Enquêtes et d’Analyses pour la Sécurité de l’Aviation Civile c.g. center of gravity CAM cockpit area microphone CF center field CFR Code of Federal Regulations CLB climb CRM crew resource management CRZ cruise CVR cockpit voice recorder DFDAU digital flight data acquisition unit DFW Dallas/Fort Worth International Airport FAA Federal Aviation Administration FARs Federal Aviation Regulations FDR flight data recorder fps feet per second GPWS ground proximity warning system HBAW flight standards handbook bulletin for airworthiness IOE initial operating experience KIAS knots indicated airspeed LOFT line-oriented flight training MAZ Eugenio Mariá de Hostos Airport MDCRS Meteorological Data Collection and Reporting System Abbreviations vii Aircraft Accident Report mg milligram MLG main landing gear msl mean sea level NLG nose landing gear PA public address PIC pilot-in-command SB service bulletin SJU Luis Muñoz Marin International Airport SNPRM supplemental notice of proposed rulemaking STC supplemental type certificate STT Cyril E. King International Airport TO takeoff VASI visual approach slope indicator viii Aircraft Accident Report Executive Summary On May 9, 2004, about 1450 Atlantic standard time, Executive Airlines (doing business as American Eagle) flight 5401, an Avions de Transport Regional 72-212, N438AT, skipped once, bounced hard twice, and then crashed at Luis Muñoz Marin International Airport, San Juan, Puerto Rico. The airplane came to a complete stop on a grassy area about 217 feet left of the runway 8 centerline and about 4,317 feet beyond the runway threshold. The captain was seriously injured; the first officer, 2 flight attendants, and 16 of the 22 passengers received minor injuries; and the remaining 6 passengers received no injuries. The airplane was substantially damaged. The airplane was operating under the provisions of 14 Code of Federal Regulations Part 121 as a scheduled passenger flight. Visual meteorological conditions prevailed for the flight, which operated on an instrument flight rules flight plan. The National Transportation Safety Board determines that the probable cause of this accident was the captain’s failure to execute proper techniques to recover from the bounced landings and his subsequent failure to execute a go-around. The safety issues in this report include flight crew performance, the lack of company bounced landing recovery guidance and training, and malfunctioning flight data recorder potentiometer sensors. Safety recommendations concerning bounced landing recovery guidance and training and flight control surface position sensors are addressed to the Federal Aviation Administration. 1 Aircraft Accident Report 1. Factual Information 1.1 History of Flight On May 9, 2004, about 1450 Atlantic standard time,1 Executive Airlines (doing business as American Eagle) flight 5401, an Avions de Transport Regional (ATR) 72-212, N438AT, skipped once, bounced hard twice,2 and then crashed at Luis Muñoz Marin International Airport (SJU), San Juan, Puerto Rico. The airplane came to a complete stop on a grassy area about 217 feet left of the runway 8 centerline and about 4,317 feet beyond the runway threshold. The captain was seriously injured; the first officer, 2 flight attendants, and 16 of the 22 passengers received minor injuries; and the remaining 6 passengers received no injuries. The airplane was substantially damaged. The airplane was operating under the provisions of 14 Code of Federal Regulations (CFR) Part 121 as a scheduled passenger flight. Visual meteorological conditions prevailed for the flight, which operated on an instrument flight rules flight plan. Flight 5401 departed Eugenio Mariá de Hostos Airport (MAZ), Mayagüez, Puerto Rico, for SJU about 1415. The captain was the nonflying pilot for the flight, and the first officer was the flying pilot. The flight crew stated that the takeoff, climb, and en route portions of the flight were uneventful. At 1437:05, as the flight approached the SJU traffic area, the cockpit voice recorder (CVR) recorded the first officer confirming with the captain that automatic terminal information service (ATIS) Juliet, which reported that winds were 060° magnetic at 17 knots and gusting at 23 knots, was current.3 Shortly thereafter, the captain briefed a Vref (the minimum approach airspeed in the landing configuration before the airplane reaches the runway threshold) of 95 knots and told the first officer to “stand by for winds.”4 The first officer asked the captain if he should set his airspeed bug5 to 95 knots, and the captain replied, “yeah.” At 1443:03, a controller from the SJU Terminal Radar Approach Control cautioned the pilots of possible wake turbulence from a preceding Boeing 727.6 The 1 Unless otherwise indicated, all times in this report are Atlantic standard time. 2 For the purposes of this report, the term “skip” refers to a landing airplane that momentarily becomes airborne after contact with the runway. A bounce is similar to a skip; however, the airplane reaches a higher altitude after contact with the runway. A skip or a bounce is typically caused by excessive airspeed or excessive back pressure being applied to the flight controls by the pilot. 3 ATIS information Juliet was based on a 1356 meteorological aerodrome report. For more information about the meteorological conditions that existed before and after the time of the accident, see section 1.7. 4 For information about Executive Airlines’ wind additive to approach airspeed procedures, see section 1.17.2.1. 5 An airspeed bug is an orange reference marker on the inside of the airspeed indicator that is set to Vref by the pilot before the approach. Factual Information 2 Aircraft Accident Report captain then told the first officer to “get your speed back. You do not want to take wake turbulence from a seven five.” At 1443:44, the approach controller told the flight crew to reduce the airspeed to 160 knots. After this instruction, the captain again told the first officer to slow down because of the preceding airplane. The first officer replied, “[approach control] said one sixty though I thought.” The captain stated, “yeah, slow it down even more though…just go about one forty.” At 1446:17, the captain stated, “San Juan tower Eagle four zero one ILS [instrument landing system] runway one zero, with eight in sight.” The SJU air traffic control (ATC) tower local controller stated, “one departure prior to your arrival…seven twenty seven, mile final just reported loss of ten knots.” At 1446:33, the local controller cleared the airplane to land on runway 8. The captain acknowledged the clearance. At this time, the first officer turned the airplane left toward runway 8 and transitioned to the visual approach slope indicator (VASI), which is located near the approach end of the runway for glideslope guidance. At 1449:07, during the approach to landing, the captain stated, “you better keep that nose down or get some power up because you’re gonna balloon.”7 The captain then told the first officer to “bring the power back.” About 2 seconds later, the ground proximity warning system (GPWS) alerted “minimums,” and the captain instructed the first officer to get the airplane’s nose up. Four seconds later, the GPWS alerted “glideslope,” and the captain stated, “below the glideslope.” The first officer responded, “correcting.” 6 The 727 landed on runway 10. The accident airplane was initially on approach to runway 10 but was later vectored to runway 8. For information about the Safety Board’s air traffic control (ATC) radar data study, see section 1.16.2. 7 The term “balloon” refers to a landing airplane that rises slightly before touching down. Ballooning is typically caused by excessive airspeed or excessive back pressure being applied to the flight controls by the pilot during the landing flare. Factual Information 3 Aircraft Accident Report At 1449:28, the captain told the first officer to “power in a little bit.” Flight data recorder (FDR) data indicated that the airplane was about 45 feet above ground level8 and traveling at 110 knots indicated airspeed (KIAS) when it crossed the runway 8 threshold 2 seconds later. After the airplane crossed the runway threshold, the captain stated, “power in a little bit, don’t pull the nose up, don’t pull the nose up.” At 1449:39, the captain stated, “you’re ballooning,” and the first officer replied, “all right.” CVR and FDR information indicated that the airplane touched down for the first time about 1449:41 and about 1,600 feet beyond the runway 8 threshold. At this time, the FDR recorded vertical and lateral loads of about 1.3 Gs and -0.10 G, respectively.9 At 1449:41, the captain stated, “get the power,” and, 1 second later, “my aircraft.” The first officer responded, “your airplane.” FDR data indicated that, by 1449:43, the airplane had skipped to an altitude of about 4 feet. CVR and FDR information indicated that the airplane touched down a second time about 1449:45 and about 2,200 feet beyond the runway 8 threshold. FDR data indicated that the airplane then pitched up to an angle of 9° while climbing to an altitude of 37 feet and that the engine torque increased from 10 to 43 percent. About 1449:49, the pitch angle decreased to -3°, and the engine torque started to decrease to 20 percent. The pitch angle continued to decrease to -10°. CVR and FDR information indicated that the airplane touched down a third time about 1449:51 at a bank angle of 7° left wing down and about 3,300 feet beyond the runway 8 threshold. Concurrently, the FDR recorded vertical and lateral loads of about 5 Gs and 0.85 G, respectively. By 1449:54, the pitch angle was 11°, and the airplane had bounced to an altitude of about 24 feet. CVR and FDR information indicated that the airplane touched down a fourth time about 1449:56 (about 15 seconds after the initial touchdown) and about 4,000 feet beyond the runway 8 threshold. FDR data indicated that the airplane pitched down to -7° and that it was banked 29° left wing down. The airplane came to a complete stop about 4,317 feet from the runway threshold. Figure 1 shows the ground track of the accident flight, and figure 1a shows the altitude profile of the accident flight. 8 level. Unless otherwise indicated, altitudes referenced in this report are reported as height above ground 9 One G is equivalent to the acceleration caused by the earth’s gravity (32.174 feet/second2). The Safety Board conducted an airplane performance study, which included airplane trajectory, load, and standard performance calculations. For more information about the Board’s airplane performance study, see section 1.16.1. Factual Information 4 Figure 1. Ground track of the accident flight. Figure 1a. Altitude profile of the accident flight. Aircraft Accident Report Factual Information 5 Aircraft Accident Report 1.2 Injuries to Persons Table 1. Injury chart. Injuries Flight Crew Cabin Crew Fatal 0 0 Serious 1 Minor 1 None Total Passengers Other Total 0 0 0 0 0 0 1 2 16 0 19 0 0 6 0 6 2 2 22 0 26 Note: Title 14 CFR 830.2 defines a serious injury as any injury that (1) requires hospitalization for more than 48 hours, starting within 7 days from the date that the injury was received; (2) results in a fracture of any bone, except simple fractures of fingers, toes, or the nose; (3) causes severe hemorrhages or nerve, muscle, or tendon damage; (4) involves any internal organ; or (5) involves second- or third-degree burns or any burns affecting more than 5 percent of the body surface. A minor injury is any injury that does not qualify as a fatal or serious injury. 1.3 Damage to Airplane The airplane was substantially damaged. 1.4 Other Damage None. 1.5 Personnel Information 1.5.1 The Captain The captain, age 33, was hired by Executive Airlines on January 11, 1999. He held a multiengine airline transport pilot certificate with an ATR-42/-72 type rating.10 (The Federal Aviation Administration [FAA] denotes both airplane models on a pilot’s airman certificate regardless of which simulator or airplane model the pilot used to qualify for the type rating.) The captain’s most recent FAA first-class airman medical certificate was issued on February 10, 2004, and contained the limitation that he “must wear corrective lenses.” According to the captain’s employment application for Executive Airlines, from October 1996 to November 1997, he worked as a pilot-in-command (PIC) at Westwind Aviation, Phoenix, Arizona. From December 1997 to February 1998, the captain held 10 On August 17, 2001, the captain received a notice of disapproval from the Federal Aviation Administration (FAA) after a checkride for his ATR-42/-72 type rating. The notice of disapproval stated that the entire flight check would have to be repeated, except for the portions involving stalls and steep turns. On September 2, 2001, the captain was rechecked successfully, and he received his ATR-42/-72 type rating. Factual Information 6 Aircraft Accident Report nonaviation-related jobs. From March 1998 to January 1999, he worked as a PIC at Sunrise Airlines in Phoenix. Executive Airlines records indicated that the captain had accumulated a total flight time of about 6,071 hours, about 3,814 hours of which were with the company in its ATR-42 and -72 airplanes, including about 1,120 hours as PIC. The captain had flown about 177, 134, 72, and 3.5 hours in the last 90, 60, and 30 days, and 24 hours, respectively. The captain’s last ground training occurred on October 10, 2003; his last PIC proficiency check occurred on October 15, 2003; and his last PIC line check occurred on October 20, 2003. On May 7, 2004, the captain flew the first leg of a 2-day trip sequence, which was a roundtrip between SJU and Flamingo Airport, Antilles Islands, the Netherlands, where he remained overnight. On May 8th, the captain flew the return flight to SJU. He stated that he felt “well rested” for the flights to and from the Antilles Islands. On May 9, the captain reported for standby duty at SJU about 1000. The captain stated that he had slept well the night before. The captain stated that he did not smoke, drink alcohol, or take any medications. 1.5.2 The First Officer The first officer, age 26, was hired by Executive Airlines on March 15, 2004. He held a commercial pilot certificate with single-engine and multiengine land and instrument airplane ratings. The first officer’s most recent FAA airman first-class medical certificate was issued on February 10, 2004, and contained the limitation that he “must wear corrective lenses.” According to the first officer, his only previous aviation-related employment was as a flight instructor in Cessna 172 and Baron airplanes at Windy City Flyers, Wheeling, Illinois. Executive Airlines records indicated that the first officer had accumulated a total flight time of about 2,000 hours, about 20 hours of which were with the company as first officer in its ATR-42 and -72 airplanes. About 18.5 hours of the first officer’s flight time, which included eight landings in the ATR-72, were accumulated during his initial operating experience (IOE).11 Flight 5401 was the first officer’s first scheduled flight since he completed IOE on May 4, 2004. The first officer had flown about 20 hours, all of which were flown in the last 30 days. The first officer’s last recurrent ground training occurred on April 10, 2004, and his last first officer proficiency check occurred on April 26, 2004. On May 8, 2004, the first officer was assigned standby duty at SJU from 0600 to 1400. He was not assigned any flights during this period. On May 9, the first officer reported for standby duty at SJU about 1100. 11 IOE consists of revenue flights flown by pilots after they complete their initial simulator training. These flights are conducted in the presence of a company check airman. Federal regulations require that Part 121 pilots have 20 hours of IOE, which can be reduced by 1 hour (up to 10 hours) for each landing that they have completed. Factual Information 7 Aircraft Accident Report 1.5.2.1 The First Officer’s Medical History and Prescription Drug Use A review of the first officer’s medical records from his personal psychiatrist revealed that, in July 2001, he began seeing the psychiatrist for treatment of various anxiety-related symptoms. The psychiatrist prescribed alprazolam to treat the first officer’s condition.12 Common side effects of alprazolam include drowsiness and light-headedness. The first officer noted on his psychiatric patient information form that he was employed as a part-time flight instructor and that his ambition was to become a commercial airline pilot. In March 2004, the psychiatrist noted in the first officer’s medical records that they discussed the following: heightened anxieties surrounding…his intensive ‘wind-down’ training for full commercial jet pilot licensure….we looked at creative as needed manipulation of alprazolam being mindful of…the need to stay alert. The first officer’s pharmacy refill records indicated that he filled prescriptions for 60 0.25-milligram (mg) alprazolam tablets on July 15, August 19, and November 3, 2001, and on March 28 and May 4, 2004. The first officer stated that he typically took one-half of a 0.25-mg tablet and that he took that dosage about once every 2 or 3 days. He stated that he did not take any alprazolam on the day of the accident and that he thought that the only time he took a whole 0.25-mg tablet in the 72 hours before the accident flight was on May 8 about 2000. A review of the first officer’s three most recent FAA airman medical certificates (dated August 13, 2001; August 7, 2003; and February 10, 2004) revealed that he did not indicate that he was taking alprazolam or being treated by a psychiatrist for anxiety.13 Specifically, the first officer checked the “no” box in response to item No. 18 on the airman medical certificate application, which asks, “Have you ever in your life been diagnosed with, had, or do you presently have any of the following…Mental disorders of any sort, depression, anxiety, etc.?” Further, the first officer did not provide any information about his psychiatric visits in response to item No. 19 on the application, which asks the applicant to list any “visits to health professional within last 3 years.” The Executive Airlines Flight Manual, Chapter 3, “Crew Qualification and Responsibility,” Section 12.5, “Use of Medication (FARs [Federal Aviation Regulations] 91.17),” states, in part, the following: 12 The prescription was for 60 0.25-milligram (mg) tablets with instructions to take one to two tablets every 2 to 3 hours, as needed, and not to exceed 8 mg per day. 13 The National Transportation Safety Board is aware that, after the accident, the FAA revoked the first officer’s airman medical certificate because he allegedly falsified his application. Factual Information 8 Aircraft Accident Report FARs prohibit [a person from] acting or attempting ‘to act as a crewmember of a civil aircraft while using any drug that affects the person’s faculties in any way contrary to safety…’ Crewmembers who are unsure of the side affects of a particular prescription or non-prescription drug are advised to consult their FAA Aeromedical Examiner, or [company] Corporate Medical Director. Although 14 CFR 61.53, “Prohibition on Operations During Medical Deficiency,” does not specifically note anxiety as a disqualifying condition, the FAA “Guide for Aviation Medical Examiners,” dated September 2003, states, “the use of a psychotropic drug is disqualifying for Aeromedical certification purposes. This includes all…anxiolytics [that is, medications used for the treatment of anxiety.]” The first officer stated that he was not taking alprazolam at the time of his last FAA medical examination in February 2004. The investigation determined that the first officer did not consult either an FAA aeromedical examiner or executive airlines’ medical director regarding his use of alprazolam. 1.5.3 The Flight Attendants The flight attendant assigned to the forward jumpseat14 had worked for Executive Airlines as a flight attendant for 6 weeks. This flight attendant had been employed by the company for 6 years in another capacity. On May 8, 2004, she was assigned a 2-day trip sequence, which included flight 5401. On May 9, she began duty at Cyril E. King Airport (STT), St. Thomas, Virgin Islands, about 1130. The flight attendant assigned to the aft jumpseat, who was the lead flight attendant on the accident flight, had worked for Executive Airlines for 4 months and had no previous airline experience. This flight attendant was off duty during the 2 days before the accident. On May 9, 2004, she began duty at STT about 1210. 1.6 Airplane Information The ATR 72-212 is a high-wing, twin turbopropeller, pressurized airplane. The airplane has an overall length of 89 feet 1.5 inches and a wingspan of 88 feet 9 inches. The accident airplane, serial number 438, was delivered new to AMR Leasing Corporation,15 Dallas, Texas, from ATR, Toulouse, France, on March 27, 1995. At the time of the accident, the airplane had 19,276 total flight hours and 18,086 total cycles.16 The airplane was equipped with two Pratt & Whitney 127 PW turbopropeller engines and two Hamilton Standard four-blade propellers. The time since new for the left 14 For a description of the airplane’s interior configuration, see section 1.15.1 and figure 4. 15 AMR Leasing Corporation is owned by AMR Eagle Holding Corporation, which also owns Executive Airlines. 16 An airplane cycle is one complete takeoff and landing sequence. Factual Information 9 Aircraft Accident Report engine was 18,208 hours, and the time since overhaul was 11,709 hours. The time since new for the right engine was 15,637 hours, and the time since overhaul was 8,435 hours. According to the load manifest for flight 5401, the airplane’s takeoff weight was about 36,590 pounds, including 3,960 pounds of passenger weight and 770 pounds of baggage weight,17 and its takeoff center of gravity (c.g.) was -11 inches.18 1.6.1 Pitch Control System Two elevators (left and right) perform pitch control of the airplane. The elevators are movable control surfaces attached to the rear spar of the horizontal stabilizer, which is mounted on top of the vertical stabilizer in a T-tail configuration. Each elevator has a trim tab with an actuator. The elevators are controlled either by manual inputs from the captain or first officer control columns, which are not directly mechanically linked, or the autopilot system. Each control column is connected through a dynamometric rod19 to a cable tension regulator located under the cockpit floor. The cable tension regulator maintains constant cable tension and transmits column movement to two cables that run the length of the fuselage to the aft fuselage elevator cable quadrant, which converts cable movement to pushrod and elevator control bellcrank movement. Rotation of the elevator control bellcrank deflects one elevator, and the pitch uncoupling mechanism (located in the horizontal stabilizer between the left and right elevator bellcranks) moves the other elevator. Pitch uncoupling occurs when opposing rotational forces exist between the left and right elevator control bellcrank shafts. When the pitch control system becomes uncoupled or experiences a malfunction (such as jamming) that restricts one side of the pitch control system, the independent operation of both elevators from either pilot’s control column is allowed. Also, when the pitch control system becomes uncoupled, a microswitch located on the pitch uncoupling mechanism closes, which illuminates the master warning and pitch uncoupling warning lights located on each pilot’s instrument panel and activates a repetitive aural chime.20 17 Average passenger weights were used to calculate the total passenger weight. According to the load manifest for flight 5401, 290 pounds of cargo were stowed in the forward cargo area, and 480 pounds of cargo were stowed in the aft cargo area. The airplane’s maximum certificated gross takeoff weight was 48,501 pounds. 18 The airplane’s takeoff c.g. limits were from -14.4 inches to -0.2 inch. 19 Each dynamometric rod has two microswitches, which change state when 22.48 pounds are applied to the control column. One microswitch indicates that the control column has been pushed downward, and the other microswitch indicates that the control column has been pushed upward. When the microswitches change state, the autopilot might disengage and trigger the “effort on pitch axis” FDR parameter. 20 According to ATR, recent ATR model airplanes incorporate a device (a switch installed on the first officer’s maintenance panel) that allows the pitch control system to be recoupled. Because recoupling the pitch control system is considered a maintenance action, recoupling can only be performed on the ground. The accident airplane was not equipped with a recoupling device. Factual Information 10 Aircraft Accident Report 1.6.2 Landing Gear System The ATR-72 is equipped with a retractable, fuselage-mounted, tricycle-arranged landing gear system. The landing gear system consists of one forward-retracting, steerable, nose landing gear (NLG) assembly and two inboard-retracting main landing gear (MLG) assemblies, all of which are hydraulically controlled.21 Each MLG assembly consists of a trunnion leg, a trailing arm, a shock absorber, an actuator, and a side brace. Figure 2 shows the MLG assembly. Figure 2. Main landing gear assembly. The ATR-72 landing gear and associated structure were designed to absorb energy equivalent to a maximum airplane descent rate of 10 feet per second (fps) when landing at the airplane’s maximum design landing weight (consistent with the landing design limits 21 The NLG assembly retracts into the nose wheel well, and the MLG assemblies retract into the fuselage. Factual Information 11 Aircraft Accident Report imposed by 14 CFR 25.473 to 25.487). In addition, in accordance with Section 25.723, the ATR-72 MLG is designed to absorb reserve energy equivalent to a maximum airplane descent rate of 12 fps when landing at the airplane’s maximum design landing weight. 1.6.3 Cockpit Seats The accident airplane’s cockpit seats were designed by Ipeco Europe, Ltd. According to Ipeco, the accident cockpit seats were manufactured in 1986, and the sleeve assemblies installed on the seats were manufactured in 2000 and were 2 of 30 assemblies manufactured in a batch run. Ipeco Test Report No. 1057, issued March 25, 1983, indicated that the design of the accident airplane cockpit seats met the static load requirements contained in 14 CFR 25.561 and Technical Standard Order C-39a (9 Gs forward, 1.5 Gs side, 6 Gs down, and 2 Gs up). 1.7 Meteorological Information 1.7.1 Airport Weather Information Weather observations at SJU are made every hour by an Automated Surface Observing System (ASOS),22 which transmits an official meteorological aerodrome report every 56 minutes after the hour. The ASOS is located about 1,900 feet south of the approach end of runway 8, and its wind measuring equipment is installed 33 feet above the ground. About 1356 on the day of the accident, the ASOS reported that visibility was 10 statute miles, clouds were scattered at 3,000 and 4,300 feet and broken at 5,000 feet, and winds were 050° at 17 knots and gusting to 23 knots. At 1456, the ASOS reported that visibility was 10 statute miles; clouds were few at 2,300 feet, scattered at 3,400 feet, and broken at 5,500 feet; and winds were 060° at 15 knots and gusting to 22 knots. The SJU ASOS also provides high-resolution observations that are measured and stored every 5 minutes. About 1445, the ASOS reported that visibility was 10 statute miles, clouds were scattered at 2,300 feet and broken at 3,400 feet, and winds were 050° at 15 knots and gusting to 23 knots. About 1450, the ASOS reported that visibility was 10 statute miles, clouds were scattered at 2,300 feet and broken at 3,400 feet, and winds were 060° at 18 knots and gusting to 22 knots. 1.7.2 Additional Wind Information The 2000 upper air sounding (that is, a vertical profile of atmospheric conditions) from SJU showed northeasterly winds aloft from 15 to 24 knots below 4,000 feet mean sea level (msl).23 22 ASOS is a system that continuously measures weather information, including windspeed and direction, visibility, precipitation, cloud cover, temperature, dew point, and altimeter setting. Factual Information 12 Aircraft Accident Report Center field (CF) wind data,24 which report 2-minute average winds, were obtained from the FAA for the period from 1445:02 to 1452:02. During this 7-minute period, the CF anemometer indicated wind directions of 060° and 070° magnetic, and the CF 2-minute average windspeed ranged from 12 to 16 knots. No wind gusts were reported for this period.25 A Safety Board meteorologist retrieved Meteorological Data Collection and Reporting System (MDCRS)26 reports from the National Oceanic and Atmospheric Administration Forecast System Laboratory archive. Two ascent profiles from airplanes departing STT (about 58 nautical miles from SJU)27 before and after the time of the accident showed 13-knot winds at 1,160 feet msl, 15-knot winds at 1,440 and 1,600 feet msl, and 13-knot winds at 1,670 feet msl. 1.8 Aids to Navigation No problems with any navigational aids were reported. 1.9 Communications No communications problems between the pilots and any of the air traffic controllers who handled the accident flight were reported. 1.10 Airport Information SJU is located 3 miles southeast of San Juan at an elevation of 9 feet msl. The airport has an ATC tower, which provides approach and departure services. SJU has two precision instrument approach runways: runways 8/26 and 10/28. Runway 8/26 is about 10,000 feet long and 200 feet wide. Runway 10/28 is about 8,000 feet long and 150 feet wide. The runway surfaces are constructed of grooved asphalt and are accessible by parallel taxiways. SJU was certificated under 14 CFR Part 139. A standard two-bar VASI is located near the approach end of runway 8. SJU maintains an index D aircraft rescue and firefighting (ARFF) facility, which has six ARFF vehicles.28 23 The National Weather Service typically launches radiosonde balloons about 0700 and 1900. The 2000 upper air sounding was taken shortly after 1900. 24 According to the FAA, the CF anemometer is located midfield, about 1,500 feet south of runway 8, and the wind is measured at 48 feet. 25 According to the FAA, wind gust values are not reported unless the wind gust exceeds the 2-minute average windspeed by 9 or more knots. 26 The MDCRS collects, decodes, and disseminates automated weather reports. 27 No MDCRS data were available for the SJU area. Factual Information 13 Aircraft Accident Report 1.11 Flight Recorders 1.11.1 Cockpit Voice Recorder The accident airplane was equipped with a Fairchild model A-100A CVR, serial number 55031. The exterior of the CVR was not structurally damaged. The tape spool assembly and other components inside the case were not damaged and were generally in good condition. The CVR was sent to the Safety Board’s laboratory in Washington, D.C., for readout and evaluation. The tape was played back normally and without difficulty. The recording started at 1422:02 and continued uninterrupted until 1452:54. The recording consisted of four separate channels of audio information: the cockpit area microphone (CAM), the captain and first officer audio panels, and the public address (PA) system. Hot microphone transmissions were also captured on the flight crew’s respective audio channels. The audio information from all four channels was generally of good quality.29 A transcript of the 31-minute recording was prepared (see appendix B). 1.11.2 Flight Data Recorder The accident airplane was equipped with an L3 Communications Fairchild model F-800 FDR, serial number 3151, which used magnetic tape as the recording medium. The FDR was found to be in good condition. The FDR was sent to the Safety Board’s laboratory for readout and evaluation. The magnetic tape was removed from the FDR, and the data were transcribed directly to a hard disk. About 25 hours of data were recorded on the FDR, including data from the accident flight, and 56 parameters that were pertinent to the circumstances of the accident were verified. 1.11.2.1 Validation of Flight Data Recorder Data The Safety Board recovered data from the accident flight, the last landing before the accident flight,30 the first recorded landing (made about 25 operational hours before the accident flight), and the first recorded flight control ground check and subsequent 28 According to 14 CFR 139.317, an index D ARFF facility is required to have (1) either one firefighting vehicle with 500 pounds of sodium-based dry chemical or Halon 1211 or 450 pounds of potassium-based dry chemical and water with a commensurate quantity of aqueous film forming foam (AFFF) to total 100 gallons and (2) two firefighting vehicles with water and a commensurate quantity of AFFF so that the total quantity of water for foam production carried by all vehicles is at least 4,000 gallons. 29 The Safety Board uses the following categories to classify the levels of CVR recording quality: excellent, good, fair, poor, and unusable. A good quality recording is one in which most of the crew conversations could be accurately and easily understood. At times during the flight 5401 recording, the ambient noise level of the CAM channel made it somewhat difficult to discern sounds or conversations recorded on the other three channels. The transcript that was developed may indicate one to two words that were not intelligible. Any loss in the transcript can be attributed to simultaneous cockpit/radio transmissions that obscure each other. Factual Information 14 Aircraft Accident Report takeoff. However, the Board could not recover three segments of the event data (8 seconds total). The Bureau d’Enquêtes et d’Analyses pour la Sécurité de l’Aviation Civile (BEA) had developed a program to decode the waveforms from F-800 magnetic tape FDRs. With the use of this program, the BEA was able to decode two of the three segments of waveforms; however, because of the very high variation in tape speed, only the last second of the 4-second third segment could be fully recovered.31 1.11.2.2 Aileron Surface Position Sensors In accordance with 14 CFR 121.344, the accident FDR was required to record left and right aileron surface positions no later than August 20, 2001. According to Executive Airlines, the accident airplane was modified on August 7, 2001, in accordance with Supplemental Type Certificate (STC)32 No. ST01310NY, which required adding two new sensors and associated hardware on each wing.33 Although the airplane had been equipped with the required sensors and associated hardware, the left aileron surface position data were invalid; therefore, the airplane did not meet the requirements of 14 CFR 121.344. The aileron sensors installed on the accident airplane were string potentiometers. The string is attached to the aileron linkage so that any aileron movement is registered by the potentiometer, which produces a voltage input to the digital flight data acquisition unit (DFDAU). The DFDAU then converts the voltage to a digital value that is recorded by the FDR and then calibrated by Executive Airlines to determine the aileron surface position. Executive Airlines stated that it added aileron surface position sensors to its 41 ATR-72 airplanes (2 sensors per airplane, for a total of 82 sensors) in accordance with STC No. ST01310NY, and that, in the last 3.5 years, the company has replaced 47 of these sensors, which is a 57 percent failure rate. The company indicated that the sensors are not tracked and, therefore, that the times from installation to failure could not be determined. The company also indicated that the sensors do not incorporate a warning or an indication system. The company further indicated that aileron surface position sensor failures were typically caused by wear or weather-related damage. At the time of the accident, Executive Airlines performed FDR functional checks every 3,000 flight cycles. Executive Airlines indicated that the accident airplane’s last FDR functional check was conducted on January 3, 2003, about 1 year and 5 months after the STC modification and about 1 year and 4 months before the accident. After the accident, Executive Airlines started conducting FDR functional checks every 1,000 flight cycles. 30 The accident flight crew made the airplane’s last landing at MAZ about 1347 on the same day of the accident. 31 Most of the 56 verified parameters recorded from 1449:58.8 to 1450:01.86 (the first 3 seconds of the third segment) were deemed invalid. 32 An STC is issued for major design changes to type-certificated products when the change is not extensive enough to require a new type certificate. 33 Executive Airlines stated that all of its ATR-72s were modified in accordance with STC No. ST01310NY. Factual Information 15 Aircraft Accident Report 1.12 Wreckage and Impact Information 1.12.1 General Wreckage Description The first evidence of ground impact (from the airplane’s third touchdown) was located on runway 8 about 44 feet left of the runway centerline and about 3,361 feet beyond the runway threshold. Various ground impact, gouge, scrape, and tire marks were found near the first ground impact mark. These marks were on a magnetic heading of about 075° and were consistent with the landing gear tires on the accident airplane. The tire marks extended about 40 feet from the first ground impact area in the direction of the main wreckage. No other tire marks that were consistent with the tires on the accident airplane were found beyond the first impact area. An oily spray pattern extended outward to the left of the gouge marks. The oily substance was consistent with the fluid contained in the MLG shock absorbers. Most of the pieces found at the first ground impact area were small sections of the NLG, including the NLG door, pieces of the left MLG, and the fuselage belly fairing. Evidence of another ground impact (from the airplane’s fourth touchdown) was found on the grassy area about 145 feet left of the runway 8 centerline and 4,053 feet beyond the runway threshold (about 692 feet from the initial ground impact area). This evidence included left wing tip scrape marks, left engine propeller strike marks, and a fuselage belly impact impression, which were all on a magnetic heading of about 070°. The marks on the grassy area continued from the second ground impact area to the main wreckage location. Sections of the outboard left wing and the left propeller blades were also found in this area. The main wreckage was located about 217 feet left of the runway 8 centerline and about 4,317 beyond the runway threshold. The main wreckage consisted of most of the airplane structure, except for the lower section of the left MLG tire assembly, which was found about 302 feet left of the runway 8 centerline and 3,944 feet beyond the runway threshold (about 373 feet from the main wreckage). Figure 3 shows ground scrape marks and the main wreckage. Factual Information 16 Aircraft Accident Report Figure 3. Ground scrape marks and the main wreckage. The landing gear handle was found in the down position. The parking brake and gust lock levers were found positioned to off. The flap handle showed 030° of flaps. The captain and first officer internal VmHB30 (minimum high bank at flaps 030°) airspeed bugs were found set at 96 KIAS. The power management selector was found set at CLB (climb). The CVR, left elevator, rudder, and first officer-side fuel pump circuit breakers were found tripped.34 1.12.2 Fuselage, Wings, and Engines The airplane fuselage was found intact and orientated upright. The vertical and horizontal stabilizers remained attached to the tail structure and showed no evidence of impact damage. The left forward hatch exit was found open, and the hatch was found 34 The Safety Board documented the positions of the circuit breakers, switches, and movable controls on the day after the accident. The documented positions of the circuit breakers, switches, and other movable controls may not represent their actual positions after the accident. Emergency medical technicians entered the cockpit during the emergency response to help the flight crew evacuate the airplane. Further, an Executive Airlines mechanic entered the cockpit after the emergency response and reportedly turned off all switches and disconnected the battery system. Factual Information 17 Aircraft Accident Report inside the airplane. The aft main entry door was found open, and the handrail was found locked in the up position. A section of about 13 feet of the forward, left side of the lower fuselage was severely crushed. The left forward cargo door and surrounding structure were severely deformed and buckled. A section of about 8 feet of the forward, right side of the lower fuselage was crushed. The damage on both the left and right sides of the lower fuselage extended from the belly to the floor line, and the left side exhibited some evidence of scrape marks just below the floor line. Most of the left belly fairing forward and aft of the left MLG was found crushed and deformed, and a portion was missing. The aft, lower surface of both sides of the fuselage exhibited evidence of scraping, minor crushing damage, and skin wrinkling. The aft fuselage just forward of the tail cone was found wrinkled around its circumference. All eight of the wing attachment fittings and the two shear web supports located on the fuselage were fractured; however, the wing assembly remained near its installed location in the fuselage. The left wing was found rotated counter-clockwise about 15° left (as viewed from aft), and the wing tip and engine propeller hub contacted the ground. The right wing section was found attached to the center wing box and exhibited minimal damage. All control surfaces were found intact and exhibited no damage. All of the fracture surfaces on the wing frames, struts, and shear web were consistent with overload failure. The left and right engine assemblies were found attached to the airframe. All four propeller blades on the left engine were sheared off at the blade root section. A section of about 12 inches was missing from each of the right propeller blades, which remained attached to the engine. The fuselage near the right engine exhibited propeller strike marks. 1.12.3 Landing Gear System and Components The right MLG was found down and locked and remained attached to the fuselage by the trunnion attach points. All of the right MLG components, including the trunnion leg, side brace, shock absorber, and actuator, were found intact and attached to the gear assembly. The right MLG outboard tire was found inflated, and the inboard tire was found deflated. The right MLG inboard tire exhibited some evidence of rubber scraping and minor tears along several circumferential tread lines. No evidence of preexisting damage was found on either tire. Both tires were free to rotate with slight resistance. Both brake assemblies were found intact. Both ground proximity sensors were found attached. The left MLG was fractured circumferentially at its vertical trunnion leg just below the actuator attach point. The upper portion of the vertical trunnion leg, which remained attached to the horizontal trunnion leg, was 4 inches long. The horizontal trunnion leg remained attached to the fuselage by the trunnion attach points, and no evidence of any damage to these components was found. The lower portion of the vertical trunnion leg, which remained attached to the wheel assembly, was 24 inches long. Factual Information 18 Aircraft Accident Report Sections of the vertical trunnion that contained the fractures were removed from the airplane and sent to the Safety Board’s Materials Laboratory for further examination. The examination revealed that the fracture surfaces were consistent with overstress separation. No evidence of fatigue was found. The upper portion of the left MLG side brace remained attached to the fuselage, and the lower portion had separated from the fractured vertical trunnion leg and showed no evidence of scraping or scoring. The secondary side brace remained attached to the left MLG and exhibited minimal damage. The left MLG shock absorber was found separated from the MLG assembly. The shock absorber was found along the wreckage path in the fully compressed condition, and no shock absorber fluid was present. The shock absorber base was fractured about 5 inches from the bottom. The upper portion of the shock absorber was intact. Both left MLG tires were found inflated. The inboard tire exhibited evidence of scoring and rubber tearing over about 22 inches of the circumference of the tire. Both tires rotated freely with slight resistance. Both brake assemblies were found intact. The NLG was found embedded in the belly of the forward fuselage, folded rearward (opposite of its normal position), and rotated counter-clockwise (the right tire was in contact with the ground). The NLG had separated from its fuselage attach points in the wheel well, which was severely crushed. The right attach point had sheared off. The NLG remained attached to the fuselage by the actuating cylinder. Both NLG tires were found deflated and exhibited evidence of rubber scoring and tearing. No evidence of tire rupture or burst was found. Both inner hubs were fractured, and pieces of the hubs were found on runway 8 near the first ground impact area. The wheel well structure was severely crushed. 1.12.4 Elevator and Rudder Control Systems The elevator pitch uncoupling mechanism was found uncoupled. All of the other tail section elevator components were found intact and attached to their respective attachment points. Both elevator control quadrants could be rotated by hand until their respective elevators contacted their up or down stops. No mechanical binding or resistance was felt. The elevator components beneath the cockpit floor, including the pulleys and brackets, were found damaged. All of the elevator cables were found intact and routed through and contained within their respective pulleys. The tail section rudder components were found intact and attached to their respective attachment points. The aft rudder control quadrant showed limited travel in both directions when rotated forward and aft using hand pressure. Factual Information 19 Aircraft Accident Report 1.12.5 Cockpit Seats The captain seat was found attached to the floor at all mounting points, and the floor beneath the mounting brackets was not deformed. The inertia reel was found in the locked position and worked normally. The forward and aft seat adjustment lever would not move. The seat height lever moved freely, but the seat would not lock into any position. The seat was found in the lowest height position and moved freely up and down. The first officer seat was found attached to the floor at all mounting points, and the floor beneath the mounting brackets was not deformed. The first officer seat operated normally. Both seats were removed from the airplane and sent to the Safety Board’s Materials Laboratory for examination.35 1.13 Medical and Pathological Information Required Federal drug and alcohol testing of the captain and first officer were negative for alcohol and drugs of abuse. Company drug and alcohol testing also tested negative for alcohol and a wider range of drugs, including alprazolam. A review of the captain’s postaccident medical records revealed that he sustained a compression fracture of the first lumbar vertebrae. The first officer reported that he sustained a contusion on his forehead. The forward flight attendant reported that she had arm, shoulder, and neck pain and bruises on her arms. The aft flight attendant reported that she had neck and back pain. According to medical records and personal injury reports, 16 of the 22 passengers sustained minor injuries. The remaining six passengers did not report any injuries. 1.14 Fire No evidence of an in-flight or a postcrash fire was found. 1.15 Survival Aspects 1.15.1 General The airplane was configured with 64 passenger seats in a single-aisle configuration. The cockpit contained two flight crew seats and one retractable observer seat. An aft-facing, single-occupancy, retractable flight attendant jumpseat was mounted on the forward bulkhead, and a forward-facing, single-occupancy, retractable flight attendant jumpseat was mounted on the aft bulkhead. Figure 4 shows the interior configuration of the airplane. 35 For information about the metallurgical examinations of the cockpit seat assemblies, see section 1.16.3. Factual Information 20 Aircraft Accident Report Figure 4. Interior configuration of the airplane. As shown in figure 4, the airplane had a main entry door on the left side of the aft fuselage, emergency hatch exits on each side of the first row of passenger seats, and a service door on the right side of the aft fuselage. The airplane was equipped with all required cabin emergency equipment, as specified in Executive Airlines’ In-Flight Manual. 1.15.2 Evacuation of Passengers and Crewmembers The cabin crew initiated the evacuation of the passengers in accordance with standard company procedures. The forward flight attendant stated that she looked out the Factual Information 21 Aircraft Accident Report forward left emergency exit window and saw a lot of “crash debris” on the ground and that she then looked out the right emergency exit window and thought that the exit looked as if it were too far above the ground to use; therefore, she decided not to open either forward exit. This flight attendant stated that most of the passengers were seated near the rear of the airplane.36 The aft flight attendant stated that, after she assessed conditions, she asked a nonrevenue pilot who was sitting in the back of the airplane to open the service door. This flight attendant stated that, while he opened the service door, she opened the main entry door. She added that emergency response personnel were waiting outside the airplane and that they assisted the passengers as they exited the airplane through the main entry door. 1.15.3 Emergency Response An ARFF specialist in a fire truck was positioned between the Executive Airlines operations area and the taxiway adjacent to runway 8 (about 1,500 feet from the approach end of the runway). The ARFF specialist stated that, about 1455, he watched the accident airplane make the approach to landing. The ARFF specialist stated that, after the second touchdown, the airplane pitched up “sharply” and that he called the ARFF station because he thought that something might be wrong. The ARFF specialist stated that he then turned on the vehicle beacon and siren and visually tracked the airplane until it came to a complete stop. The ARFF specialist stated that he drove to the location where the airplane had stopped and then approached it from the left aft side. He stated that, because he saw “black and white” smoke coming from near the left engine, he “hosed [it] down.” By 1500, four additional ARFF vehicles and five additional ARFF personnel and G.E.S. Ambulance Service ambulances had arrived. About 1515, the State Emergency Medical, San Juan Municipal, Carolina Municipal, and Guaynabo City Emergency Medical Service units arrived. Additionally, five fire trucks and crews from off-airport mutual aid services responded to the accident. According to the airport operations coordinator, he coordinated with the operations supervisor and ARFF and security personnel to escort these units to the main terminal building. The captain and the injured passengers were transported to area hospitals. 1.16 Tests and Research 1.16.1 Airplane Performance Study The Safety Board conducted an airplane performance study, which used CVR transcript information and FDR and radar data37 that were correlated to a common time reference. The study integrated FDR lateral, vertical, and longitudinal loads with pitch, 36 According to passenger information provided by Executive Airlines, the passengers were assigned to seats 4A, 5B and 5C, 6B and 6C, 7A and 7D, 8A through 8D, 9A through 9D, 10A through 10D, 11A, 14D, and 16A. Factual Information 22 Aircraft Accident Report roll, and yaw values and wreckage survey data and determined the accident airplane’s ground track, the corresponding time history of the airplane’s motions, and the estimated load factors on the left MLG and cockpit area. The airplane performance study also derived windspeed and direction to determine whether the winds had affected the airplane’s performance. Before and during the time that the CVR recorded the sound of the first thump (1449:41), the airplane encountered an 8-knot decrease followed by an 11-knot increase in windspeed. Immediately thereafter, the CVR recorded the captain’s statement, “my aircraft.” At this time, FDR data showed the engine torque decrease from about 30 to 10 percent, which is slightly above the flight idle position, and the elevators deflect from 5° to -2°. At 1449:45, the CVR recorded the sound of a second thump. FDR data showed that, immediately thereafter, the airplane pitch angle decreased to -4°, and the derived windspeed increased by 6 knots. The pitch angle then increased to about 9° while climbing to an altitude of 37 feet, and the engine torque started to increase from 10 to 43 percent. About 3 seconds later, the engine torque started to reduce to 20 percent, and the pitch angle decreased to -3°. While the pitch angle continued to decrease to -10°, the derived windspeed decreased by 8 knots, and the elevator deflection began to increase to 4°. The airplane performance study showed that, about 1449:51, when the airplane was about 3,300 feet beyond the runway 8 threshold, the CVR recorded a very loud bang, and the FDR recorded vertical and lateral loads of about 5 Gs and 0.85 G, respectively. The average vertical load for the left side of the cockpit area was calculated to be about 12 Gs. It is possible that the vertical loads experienced in the cockpit during the third touchdown were more than 12 Gs; however, this value could not be calculated because of the low FDR sampling rate. The airplane’s descent rate was determined to be about 19 to 32 fps. During the last touchdown, when the most substantial damage to the airplane most likely occurred (especially to the left side of the cockpit), the left bank angle recorded by the FDR was 29° left wing down. The average vertical loads experienced in the cockpit during the last touchdown could not be determined because the FDR data became unreliable at this point and because of the airplane’s orientation. 1.16.2 Air Traffic Control Radar Data Study The Safety Board conducted an ATC radar data study to evaluate the separation of radar tracks associated with the accident airplane and the preceding 727 that landed on runway 10 about 3 minutes before the accident airplane landed on runway 8. The study began at 1444:26, when the 727 was about 5.7 miles west of SJU on final approach to runway 10 and the accident flight was about 4.3 miles behind the 727. The study ended at 1446:42, when the last radar return associated with the 727 was received. At their closest 37 Radar data from the U.S. Air Force 84th Radar Evaluation Squadron, which were recorded by the SJU Airport Surveillance Radar-8 sensor, were used in the calculations. Factual Information 23 Aircraft Accident Report point, the airplanes were separated about 4.3 miles laterally and 400 feet vertically, which is greater than the minimum lateral separation specified in Federal requirements.38 1.16.3 Cockpit Seat Assembly Metallurgical Examinations 1.16.3.1 Accident Airplane Cockpit Seat Assemblies The accident airplane’s cockpit seat assemblies were sent to the Safety Board’s Materials Laboratory for metallurgical examination. Figure 5 shows a schematic of the cockpit seat assembly. Figure 5. Schematic of the cockpit seat assembly. 38 FAA Order 7110.65, “Air Traffic Control,” paragraphs 4-5-1, 5-5-4, and 5-5-5, require that airplanes be separated at least 3 miles laterally and/or 1,000 feet vertically. Factual Information 24 Aircraft Accident Report Visual examinations of the captain seat revealed that the weld that attached the end cap to the sleeve was fractured at both welded joints, releasing the end cap from the sleeve. Visual examinations of the fractured welded joints revealed the presence of dark, oxidized, preexisting regions that were consistent with weld discontinuities at the inner diameter side of both joints. On the forward sidewall fracture, intermittent weld discontinuities extended along most of the 4.2-inch circumferential length of the joint to a depth of about 0.03 to 0.06 inch.39 On the aft sidewall fracture, the weld discontinuity was continuous and extended along the entire circumferential length of the joint to a depth of about 0.03 to 0.06 inch. A Safety Board materials engineer estimated that the weld discontinuities extended through about 25 percent of the cross-sectional area on the forward side and about 40 percent of the cross-sectional area on the aft side. The discontinuities contained repeating columnar patterns with curved features at an angle to the surface, which was consistent with a lack of fusion that resulted from molten regions not bonding during the welding process. High-magnification optical and scanning electron microscope examinations of the aft sidewall fracture showed the presence of small fatigue cracks that had propagated from the weld discontinuity region toward the outer diameter, which slightly increased the total penetration of the preexisting defect region. Visual examinations of the intact welded joints on the first officer seat revealed that gaps existed between the mating surfaces of both welded joints, which was consistent with incomplete weld penetration. The forward sidewall region had about a 2.9-inch-long gap (the total joint length was 4.2 inches). The aft sidewall region had two slightly overlapping gaps that extended about 3 inches. A section of the forward sidewall welded joint was excised and fractured open, revealing that the gaps at the inner diameter of the welded joint had features similar to the preexisting weld discontinuity (lack of fusion) regions on the inner diameter of the welded joints on the captain seat. The maximum penetration of the weld discontinuity was about 0.045 inch in the section that was fractured open for examination. 1.16.3.2 FAA Vertical Drop Test of ATR-42 Cockpit Seats In May 2003, the FAA Technical Center in Atlantic City, New Jersey, conducted a vertical drop test of an ATR-42 airplane that had cockpit seats with the same design and part number as the cockpit seats on the accident airplane. According to the test report, vertical loads in the cockpit area were measured at 30 Gs for the left cockpit floor and 34 Gs for the right cockpit floor.40 Neither of the cockpit seats broke during the test. Because of the similarities between the accident and test airplanes’ cockpit seats and vertical loading, the cockpit seats from the FAA drop test were sent to the Safety Board’s Materials Laboratory for comparison to the cockpit seats from the accident airplane. 39 An Ipeco representative stated that the manufacturing specification for the sleeve assembly called for 100 percent weld penetration; therefore, the accident end cap and sleeve assemblies did not meet manufacturing specifications. 40 Federal Aviation Administration, The Vertical Drop Test of an ATR-42-300, FAA Technical Center (Atlantic City, New Jersey: FAA, 2004). Factual Information 25 Aircraft Accident Report Visual examination of the test airplane’s captain seat parts revealed that gaps existed at the inner diameter between the end cap and sidewall portions of the sleeve on both the forward and aft welded joints. No evidence indicated that the inner diameter surface had become molten during the welding process. Both the forward and aft welded joints were fractured in the laboratory by bending the sidewall portion outward relative to the end cap, revealing that the gaps at the inner diameter of the welded joints corresponded to weld discontinuity regions. These weld discontinuity regions extended along the entire circumferential length of the joints and penetrated about 50 to 60 percent of the joints’ depth. Visual examination of the test airplane’s first officer seat parts revealed gaps at the inner diameter that extended along the circumferential length of the welded joint. 1.17 Organizational and Management Information Executive Airlines began service as Executive Air Charter in 1982 and began scheduled passenger operations in 1985. American Eagle Holding Corporation bought Executive Air Charter in 1989. In October 2002, the company reorganized, and Executive Airlines began operations as a separate entity doing business as American Eagle and AMR Leasing Corporation. Executive Airlines is a regional airline that provides Part 121 scheduled passenger service to 40 island locations in the Caribbean. The company operates 130 flights daily from its major hubs at SJU and Miami International Airport, Miami, Florida. At the time of the accident, Executive Airlines was the largest ATR-42 and -72 operator in the United States and had a fleet of 8 ATR-42 and 41 ATR-72 airplanes. 1.17.1 Flight Crew Training Executive Airlines’ pilots attend training at the American Eagle training center located at Dallas/Forth Worth International Airport (DFW), Dallas/Fort Worth, Texas. All first-time pilots at Executive Airlines attend a basic indoctrination course, where they are taught general information on company operations. Pilots then attend initial and/or transition ground school and simulator training. According to the Executive Airlines ATR-42/72 Ground Training Instructor Lesson Plan, the typical initial and transition training consists of 10 days of ground school and 20 hours of IOE. As stated previously, new and upgrade pilots perform their IOE in the presence of a check airman. In addition, all pilots are required to attend recurrent training (a 4-hour, line-oriented flight training [LOFT] session)41 and perform a 1.5-hour proficiency checkride every year. Recurrent training and the proficiency checkrides are conducted by company check airmen. 1.17.1.1 Simulator Flight Training According to the Executive Airlines ATR-42/72 Simulator Training Syllabus, all new pilots and pilots qualifying for upgrades receive nine 4-hour simulator lessons (for a 41 LOFT facilitates the transition from simulator to line flying. Factual Information 26 Aircraft Accident Report total of 36 hours of simulator training)42 and one 2-hour simulator checkride. In addition, pilots receive 2 hours of ATR-42 training and a 1.5-hour airplane checkride. Further, pilots receive 5 hours of ATR-72 differences training.43 Simulator training is conducted by a company check airman and a simulator instructor. Executive Airlines uses an ATR-42 simulator located at DFW for most of its pilot training. The company also uses an ATR-42 simulator located at FlightSafety International, Houston, Texas, when not enough training time is available in the DFW simulator. The ATR-42 simulators used for Executive Airlines’ training are Level C simulators.44 1.17.1.1.1 Observations of Simulator Sessions On June 20, 2004, the Safety Board observed various flight profiles and procedures conducted by Executive Airlines pilots in the ATR-42 simulator at DFW, including the following: • a visual approach to the runway with the first officer as the flying pilot, 090° winds at 10 knots, gusts at 15 knots, and ATC assigning various airspeeds during the approach; • an initiation of a go-around after touchdown; and • a visual approach to the runway with the first officer as the flying pilot, the captain taking control at 80 feet, and a pitch uncoupling during a go-around— making the trim system inoperative.45 Executive Airlines’ procedures for operation of the ATR-72 allow the flying pilot to position the condition levers46 to 100 percent for landing or to leave the levers positioned at 86 percent. During a go-around, if the condition levers are positioned at 86 percent, the propellers automatically position to 100 percent when the throttles are advanced and the power management selector47 is in the TO position. This automatic 42 The hours scheduled for the nine simulator training periods are based on the pairing of two students. A student who does not have a partner is scheduled for only 18 hours of simulator training. 43 The Executive Airlines FAA-Approved Training Manual states that differences training is required for ATR-42 captains and first officers before they can serve as flight crewmembers on the ATR-72 during revenue operations. The manual states that the required crewmember emergency training and operating experience for either initial or recurrent training may be accomplished in either the ATR-42 or -72 and that the crewmember will be considered trained for both airplane models. 44 Level C simulators may be used for specified light operational task training for Part 121 and 135 transition, upgrade, recurrent, and requalification training. These simulators may also be used for initial new hire and initial equipment training on specified events for individuals who have previously qualified as PIC or second-in-command with the training operator or who meet the FARs for advanced simulator training. 45 An emergency was declared, and the appropriate checklist was completed. 46 The condition levers, which are located next to the power management selector, control propeller rpm. 47 The power management selector, which is located on the center instrument panel, provides maximum torque limit indications on each torque indicator for the selected mode of operation. The selector has four labeled operating modes: TO (takeoff), MCT (maximum continuous), CLB (climb), and CRZ (cruise). Factual Information 27 Aircraft Accident Report feature is not available in the ATR-42. During the simulations, investigators observed that the flying pilot always positioned the condition levers to 100 percent before landing. 1.17.1.2 Crew Resource Management The Executive Airlines FAA-Approved Training Manual, chapter 1, section 2, page 24, outlined the curriculum for the 2.5-hour, stand-alone crew resource management (CRM) training segment, which all new-hire pilots were required to attend. The manual states the following: Material presented in this subject area acquaints the crewmember with the principles and importance of effective crew resource management. Emphasis shall be placed on precise communication, crewmember interaction, crewmember assertiveness, and delegation of cockpit duties. The CRM curriculum contains instruction on methods of fostering crew input, maintaining situational awareness, crew coordination during an emergency or abnormal situation, cockpit discipline, and proper cockpit procedures. During day 5 of basic indoctrination training, all first officers attend a 4-hour First Officer Duties and Responsibilities Program, which is based on CRM techniques and R.E.A.C.T procedures48 and includes instruction on flight crew communications, decision-making, and stress and fatigue management. When pilots transition from first officer to captain, they attend a 6.5-hour Captain Leadership Duties and Responsibilities Program, which emphasizes flight crew roles and responsibilities, leadership development, and conflict resolution. 1.17.1.3 Bounced Landing Recovery Training Executive Airlines’ manager of training and standards stated that the company did not provide formalized bounced landing recovery techniques to pilots before the accident and that none of the company manuals contained any information about bounced landing recovery.49 The manager stated that he would not want to conduct bounced landing recovery techniques in the simulator because it is very difficult to demonstrate a bounce. The manager stated that bounced landing recovery techniques could be addressed during pilot briefings. The manager stated that, after the accident, Executive Airlines’ president and vice president of operations asked him to look into the feasibility of conducting bounced landing recovery training and incorporating bounced landing recovery techniques in the company manuals. One simulator instructor stated that, if the airplane landed hard enough to bounce, the pilot should execute a go-around. He added that, if an airplane bounced 15 to 20 feet 48 According to company documentation, a first officer can use R.E.A.C.T. (review and reconfirm, evaluate, advise, challenge, and take) procedures to challenge the captain if the first officer believes that the captain’s actions might jeopardize the safety of the operation. 49 A Safety Board review of Executive Airlines’ operations and training manuals verified that the company had no documentation regarding bounced landing recovery techniques. Factual Information 28 Aircraft Accident Report into the air after touchdown, the pilot should power up and get back to the flare position. Another simulator instructor stated that a pilot should add power to recover from a bounce. A third simulator instructor stated that, if sufficient runway existed, the pilot should add power and land, and, if sufficient runway did not exist, the pilot should execute a go-around. A company line check airman stated that, if a first officer were to bounce the airplane on landing, he would take control of the airplane, apply power, and go around. Another company line check airman stated that, if a bounced landing could be corrected safely, the pilot should proceed with the landing, and, if a bounced landing could not be corrected safely, the pilot should execute a go-around. This line check airman added that he would allow the first officer to execute the go-around after the bounce but that he would take control of the airplane if he felt the need to do so. On September 25, 2004, Executive Airlines incorporated bounced landing recovery techniques in its Airplane Operating Manual (AOM). The bounced landing recovery guidance states the following: In the event the aircraft should bounce after landing, hold or re-establish a normal landing attitude and immediately add power as necessary to control the rate of descent. When using this recovery technique, exercise extreme caution not to increase the pitch attitude above normal as this will only increase the height of the bounce and may cause entry into stall warning. DO NOT push over, as this will only cause another bounce and damage the nose gear. If there is any doubt as to a safe recovery, the captain will call for and conduct an immediate go-around. Apply go-around power and fly the Missed Approach/Rejected Landing Profile. DO NOT retract the Landing Gear until a positive rate of climb is established because a second touchdown may occur during the recovery. The Safety Board informally surveyed six airlines, an airplane manufacturer, and a pilot training facility to determine if bounced landing recovery techniques were typically contained in industry flight manuals. The survey revealed that only some of the companies included bounced landing recovery techniques in their flight manuals and discussed these techniques during training. Most of the companies indicated that bounces commonly occurred during IOE checkrides and that, when a bounce did occur, the check airman would provide verbal guidance to the pilot on how to recover the airplane. 1.17.2 Operational Guidance 1.17.2.1 Approach Airspeed Guidance Executive Airlines’ ATR-42/72 AOM, Volume 1, Performance, “Landing Speeds” (dated April 1, 2004), contains initial approach airspeed guidance. The manual states that Vapp, which is the initial approach airspeed, is to be selected at the pilot’s discretion and must be more than Vref plus VmGA15 (minimum go-around airspeed, 15° of flaps, low bank) and less than or equal to VmLB0 (minimum low bank, 0° of flaps) or VmHB0 (minimum high bank, 0° of flaps). The manual adds that Vapp is to be maintained until Factual Information 29 Aircraft Accident Report 500 feet, at which point, power and attitude should be adjusted to ensure that the airplane crosses the runway threshold within +10/-0 knots of Vref. According to the airspeed flip cards and the airspeed guidance contained in Executive Airlines’ AOM, assuming an airplane weight of 37,000 pounds, the flight crew should have selected an initial approach airspeed more than 107 KIAS and less than or equal to 128 KIAS. Executive Airlines’ ATR-42/72 AOM also contains wind additive to approach airspeed guidance. The manual states that Vref is determined by correcting VmHB30 (minimum high bank, 30° of flaps) for wind. The manual states that the wind factor should be the greater of one-third of the headwind component or the full gust factor50 and should not exceed 15 knots of correction. The manual also states that the pilot should set the calculated Vref on the airspeed bug. The manual states that the airspeeds are published in the AOM and in the flip cards51 for various landing weights and that the airspeeds to be used for approach and go-around are based on actual airplane weight and are rounded to the next heaviest increment. The AOM and the flip cards showed that the Vref for the accident flight, assuming a landing weight of 37,000 pounds, would have been 95 KIAS.52 As noted previously, ATIS Juliet, which was current at the time of the accident, reported winds of 060° at 17 knots with gusts up to 23 knots. According to the wind data, one-third of the headwind component (1/3 x 17 knots) would have been 6 knots, and the full gust factor (23 - 17 knots) would also have been 6 knots; therefore, the correct Vref for the accident flight would have been 101 KIAS. During postaccident interviews, the captain stated that he could not recall what airspeeds were used throughout the approach. Further, he could not remember if he made a correction to the airspeed during the accident flight or if he used the airspeed that was specified on the airspeed flip card (95 KIAS). He stated that airspeed corrections were needed most of the time when he landed at SJU. The first officer stated that he set the bugs on his airspeed indicator according to the airspeeds read to him by the captain. He stated that the captain did not mention if he had made an airspeed correction. Neither pilot could recall the airspeed when the airplane crossed the approach end of runway 8. 1.17.2.2 Before Landing Checklist Executive Airlines’ ATR-42/72 AOM, Volume 1, Normals, “Before Landing Checklist,” states that the power management selector should be set to the TO position by the nonflying pilot before landing. The power management selector was found in the CLB position. During postaccident interviews, the captain stated that he could not recall if the power management selector was in the CLB or TO position before landing. The captain thought all of the appropriate checklists were performed. The first officer stated that the 50 The full gust factor is the gust speed minus the headwind speed. 51 Flip cards are carried on board all company airplanes and list takeoff and approach airspeeds for various landing weight and flap configurations. 52 During postaccident examinations, Safety Board investigators found the flip cards in the airplane opened to the page that specifies the airspeed for a landing weight of 37,000 pounds. Factual Information 30 Aircraft Accident Report power management selector was in the TO position before departing MAZ. The first officer stated that he did not notice if the power management selector was positioned to CLB when he called for the climb checklist after departure from MAZ. He thought that the captain had moved the power management selector to the CRZ position when he called for the cruise checklist. Shortly after recording the captain stating that he had completed the before landing checklist, the CVR recorded him stating that he was going to position the power management selector to CLB. The CVR did not record the captain stating that he was repositioning the selector to TO. Executive Airlines’ ATR-42/72 AOM states that the condition levers should normally remain at 86 percent for landing and that the minimum setting for landing is 86 percent. The AOM adds that the maximum rpm could be set at the captain’s discretion. The AOM notes that, if the power management selector is set to TO before retarding the power levers, the condition levers might automatically advance to 100 percent. During the cruise portion of the flight, the CVR recorded the captain discussing the use of the condition levers with the first officer. The captain stated that he always landed with the condition levers set at 86 percent. He added that some pilots landed with the condition levers set at 100 percent but that he did not see any reason to land with the levers set at 100 percent. During the before landing checklist, the CVR recorded the captain stating that he was setting the condition levers to 86 percent. On September 25, 2004, Executive Airlines standardized its procedures and required that the condition levers on all of its ATR airplanes be set to 100 percent before landing. 1.17.2.3 Evacuation Procedures Executive Airlines’ ATR-42/72 AOM, Volume 1, Emergency/Abnormals, states that the alternate evacuation signal if the PA system becomes inoperative is the following: SEATBELT SIGN ........................................................................... OFF EMER EXIT LT [light] .....................................................................ON ATTND [attendant] Calls pb [push button] ................................Depress Executive Airlines’ ATR-42/72 AOM, Volume 1, Emergency/Abnormals, contains the following ground evacuation checklist: AIRCRAFT PARKING BRAKE ..........................................STOP/SET ATC (TIME PERMITTING) (F/O) .......................................... NOTIFY CL [Condition Levers] (BOTH) ................. FEATHER then FUEL S/O MIN [Minimum] CAB LIGHTS........................................................ON FLIGHT ATTENDANT (PA) ................................................. NOTIFY SEATBELTS SIGN ......................................................................... OFF EMERGENCY EXIT LIGHTS .........................................................ON ENGINE FIRE HANDLES (both) ................................................PULL FIRE AGENTS ......................................................................... AS RQD ENG [Engine] START Selector ...................................................... OFF Factual Information 31 Aircraft Accident Report FUEL PUMPS (both)....................................................................... OFF BAT [Battery] (before leaving A/C [aircraft]) ................................ OFF VOICE RECORDER CB [Circuit Breaker] 42/Row D 72/Row ...PULL After the airplane came to a complete stop, the CVR recorded a discussion between the first officer and the captain about which checklist should be initiated. The CVR recorded the captain instructing the first officer to perform the emergency evacuation checklist, which was on placards on both the first officer and captain control wheels. The first officer stated that he did not know “where to start,” and the captain then told him to perform the fire on ground checklist, which was located on the yellow emergency procedures checklist in the cockpit. During postaccident interviews, the captain stated that he tried to perform the emergency evacuation checklist but that he could not recall specifically how many steps were accomplished or if the battery switch was turned off. He stated that he tried to shut down the right engine and feather53 the right propeller. The captain added that, after shutting down the engines, his main concern was evacuating the passengers. The captain stated that he could not remember who pulled the engine fire handles or whether the fire bottles were activated. The first officer stated that he performed the fire on ground checklist first and then the emergency evacuation checklist. The first officer stated that he recalled that the captain attempted to feather the propellers and that he was able to feather the left propeller but not the right propeller because the condition lever in the cockpit had jammed. He stated that battery power was available but that the radios and the PA system were inoperative. 1.18 Additional Information 1.18.1 Additional Information About Ipeco Cockpit Seats 1.18.1.1 General Information An Ipeco representative stated that, since 1983, when the cockpit seat design found on the accident airplane went into service, the company had produced 1,420 seat assemblies and had never received a report that a seat had broken similarly to the accident seat. The representative stated that, since 1983, the company had manufactured 125 sleeve assemblies as spare parts, which were provided to FAA-authorized repair stations. He stated that he did not know how many of these assemblies had been used or were still in parts inventories. He also stated that Ipeco did not maintain records indicating why the parts were replaced. 53 Feathering means to rotate the propeller blades so that the blades are parallel to the line of flight (streamlined to the airflow) to reduce drag in flight and prevent further damage to an engine that has been shut down. Factual Information 32 Aircraft Accident Report An Ipeco representative from the company’s subsidiary in Torrance, California, stated that sleeve assemblies are periodically replaced on seats sent to its facility for overhaul. The representative stated that, during overhaul, the entire cockpit seat is inspected and that sleeve assemblies are replaced when evidence of cracking in the metal adjacent to the weld at the base of the sleeve is found. He stated that he thought that the damage on these assemblies was caused when mechanics dropped or dragged them during the removal process. According to the representative, during the last 5 years, the company had overhauled 131 ATR seats and had replaced the sleeve assemblies on 45 of the overhauled seats. 1.18.1.2 Ipeco Cockpit Seat Tensile Strength Testing As a result of the Safety Board’s metallurgical findings and Ipeco’s U.S. subsidiary’s reports of damaged sleeve assemblies, Ipeco conducted tests in December 2004 to determine the effects of weld discontinuity and cracking on the tensile strength of the sleeve assemblies.54 The Ipeco test report stated that, of the six sleeve assemblies that were used during the tests, five had been removed from in-service seats by Ipeco’s U.S. subsidiary because the assemblies had cracks, which were found during overhaul, and exhibited a lack of weld penetration that was similar to that found on the accident sleeve assemblies. The sixth sleeve assembly was newly manufactured and had 100-percent weld penetration. According to Ipeco’s report, the tests were conducted on a tensile/compression machine. Fixtures were used to mount the sleeve assemblies as they would be when normally installed on a cockpit seat. Tensile loads were then applied to the assemblies until they failed or stopped reacting to the loads. The highest load value attained before the assemblies failed or stopped reacting was recorded and was considered the ultimate load for each assembly. All of the test assemblies exceeded the certification strength requirements, including the 6-G vertical load requirement, contained in 14 CFR 25.561. 1.18.1.3 Ipeco’s Postaccident Actions Since the accident, Ipeco has taken several actions to address the issues found during the investigation. According to an Ipeco representative, Ipeco changed the type of weld it used on the sleeve assemblies to allow easy visual inspection of the weld penetration. In May 2005, Ipeco performed tensile strength tests (identical to the December 2004 tensile strength tests) on a number of parts welded with the new weld material to ensure that the parts met the seat’s strength requirements, and the weld was found to meet the requirements. The company started using the new we...
Purchase answer to see full attachment