ERA15FA088

The pilot's failure to properly set the left engine fuel selector before takeoff and to recognize the incorrect setting during the flight, which resulted in fuel starvation and a loss of engine power on both engines.

The commercial pilot departed on a cross-country flight in night instrument meteorological conditions with the airplane's fuel tanks full, providing an estimated fuel endurance of 4 hours 50 minutes. Two hours 50 minutes into the flight, the pilot reported a loss of engine power on the right engine, which was followed by a loss of engine power on the left engine. The pilot attempted to land at a nearby airport; however, the airplane impacted trees about 8 miles short of the airport. A review of weather information revealed no evidence of in-flight icing or other weather conditions that may have contributed to the accident. Postaccident examination of the airframe and engines revealed no preimpact failures or malfunctions that would have precluded normal operation. All fuel tanks were compromised; however, an undetermined amount of fuel spilled from the left fuel tank during recovery of the wreckage. The left engine fuel selector valve was found in the "X-FEED" (crossfeed) position, and the corresponding cockpit fuel selector switch was found in an intermediate position, which was likely the result of impact damage. The right engine fuel selector valve and the corresponding cockpit fuel selector switch were found in the "ON" position. With the valves in these positions, both the left and right engines would have consumed fuel from the right fuel tank. Review of performance charts and fueling records indicated that if the flight was conducted with the valves in the as-found positions, exhaustion of the fuel in the airplane's right fuel tank would have occurred about the time the pilot reported the dual engine failure. In addition, the yaw trim was found in the full nose-right position. It is possible that the pilot used nose-right yaw trim to counteract an increasing left-turning tendency during the flight as fuel was burned from only the right wing's fuel tank making it relatively lighter than the left wing. According to the expanded checklist in the pilot's operating handbook for the airplane, during taxi, the pilot was to move each fuel selector to "X-FEED" for a short time, while the other selector was in the "ON" position, before returning both fuel selectors to the "ON" position before takeoff. According to a checklist found in the airplane, the fuel selectors were to be set to "X-FEED" during taxi and then to "ON" during engine run up. GPS data recovered from onboard devices indicated that the pilot taxied from the ramp and onto the active runway without stopping in about 3 minutes, indicating that it is unlikely he performed a complete run up of both engines before takeoff. He likely failed to return the left engine fuel selector from the "X-FEED" to the "ON" position, where it remained throughout the flight and resulted in fuel starvation and a loss of engine power on both engines. Toxicological testing revealed that the pilot was taking citalopram (an antidepressant) and rosuvastatin (a statin); however, it is unlikely these drugs contributed to the accident. Review of medical and pathological information revealed no evidence of any medical condition that may have contributed to the accident.

The pilot contacted Memphis Air Route Traffic Control Center at 1747:10, and reported descending to 6,000 feet msl, from 6,700 feet msl. At 1750:48, the pilot reported "…I've got problems" and requested vectors to the nearest VFR airport. At 1752:01, the pilot transmitted "I don't know what's wrong…both engines are malfunctioning – everything's forward it was running perfect I have fuel I just don't know the right engine is out." At 1754:14, the pilot stated the "engines are not producing power I don't know what's up." At 1754:43, the pilot asked if there were any nearby suitable landing areas, and at 1755:06, radar contact was lost. There were no further communications with the airplane. Non-Volatile Memory Devices A Garmin 696 GPS, a Garmin 496 GPS, and an Apple iPhone 5s were retained from the wreckage and sent to the NTSB Recorders Laboratory for data download. No data pertinent to the accident was obtained from the Apple iPhone 5s. Both the Garmin 696 GPS and Garmin 496 GPS recorded data from the accident flight. According to GPS data, the units began recording, and then about 5 minutes later, the airplane departed TLH at 1500. It was noted that about 3 minutes elapsed between the time the airplane began to taxi and when the takeoff roll began. In addition, in that 3-minute timeframe the airplane was in constant motion. At 1750:16, at a recorded altitude of about 5,000 feet msl, the airplane slowed from about 160 knots to 142 knots groundspeed and began a descent. The descent continued, and the airplane slowed to about 100 knots for the remainder of the flight. At 1751:25, the airplane turned to a westerly heading and continued to descend. Both GPS receivers stopped recording data at 1754 and approximately 700 feet GPS altitude. Normal Procedures Checklist According to a checklist found at the accident site, the checklist items associated with the fuel system included: After Engine Start, item number 12, "Fuel Selectors – Crossfeed." Then, during the engine run up, item number 1 stated "Fuel Selectors – ON." Finally, the Top of Climb/Cruise Checklist indicated that item number 2 stated "Mixtures – Lean." Fuel Performance Calculations According to the POH, fuel usage for engine start, taxi, and takeoff was 4.2 gallons of fuel. Interpolation of performance charts revealed that the engines burned approximately 20 gallons per hour. Furthermore, the fuel consumption calculated on the previous flight was about 20 gallons per hour. An estimate of the airplane's fuel consumption during the accident flight revealed that the 2 hours and 55 minutes that had elapsed from the time of departure to the time of the accident would have consumed about 60 gallons of fuel, which included the fuel needed for engine start, taxi, and takeoff. The Office the Chief Medical Examiner for the State of Kentucky performed an autopsy on the pilot in Louisville, Kentucky. The pilot's autopsy report indicated the cause of death was "multiple blunt force injuries." The FAA's Bioaeronautical Sciences Research Laboratory, Oklahoma City, Oklahoma, performed toxicological testing of the pilot. Fluid and tissue specimens from the pilot tested negative for carbon monoxide and ethanol. However, the testing detected citalopram and its metabolite, n-desmethylcitalopram, in the urine and blood. Also, rosuvastatin was detected in the urine. Citalopram was an antidepressant with selective serotonin reuptake inhibiting action. Rosuvastatin was a member of the drug class of statins, used to treat high cholesterol and related conditions, and to prevent cardiovascular disease. The wreckage was examined at the accident site on January 4, 2015. The airplane impacted trees in a wooded area, about 8 miles east of M34. There was a strong odor of fuel, and all major components of the airplane were accounted for at the scene. The wreckage path was oriented 228 degrees, was approximately 300 feet in length, and at an elevation of 480 feet. The airplane came to rest inverted with the landing gear retracted. The fuselage and empennage were largely intact, but heavily damaged by impact. All fuel tanks were compromised during the accident and evidence of a small postcrash fire was observed at the right wing outboard fuel tank. As the airplane was being moved for recovery, an undetermined amount of fuel was noted flowing out of the inboard section of the left wing. Flight control continuity was confirmed from all flight control surfaces to the cockpit through tensile overload breaks and cuts to control cables made by recovery personnel. Examination of the cockpit and cabin areas revealed that both control yokes were attached to their respective columns and that the throttle, mixture, and propeller levers were intact in the throttle quadrant, and in the full forward position. The yaw trim actuator was observed in the full nose-right position, and the stabilator trim actuator was observed in the neutral position. The airplane was not equipped with aileron trim. The cockpit fuel selectors indicated that the left engine was in the "X-FEED" position and the right engine was in-between the "ON" and "OFF" position. The fuel selector valve positions in the wings were examined and indicated that the left engine was in the "X-FEED" position and the right engine was in the "ON" position. The seats were anchored in their mounts, the seatbelts were buckled, and all were cut by rescue personnel with the exception of the forward-facing right aft seat belt, which was intact and unbuckled. The left engine was separated from its engine mounts but remained attached to the left wing through wires and cables. The left engine turbocharger was removed from the engine and examined. Rotational scoring was noted on the interior of the turbocharger near the turbine vanes. The left propeller was separated from the left engine and was in the vicinity of the main wreckage. One propeller blade exhibited a slight s-bend, and the other propeller blade exhibited chordwise scratching. The spinner was impact damaged. The right engine remained attached to its engine mounts and was attached to the right wing. The engine cowl was removed to facilitate further examination. All major engine components remained attached to the engine. The right propeller was separated from the right engine and was located forward of and in the vicinity of the right engine. The right propeller spinner exhibited impact damage and both propeller blades were bent in the aft direction. A detailed examination of the airplane was conducted at a recovery facility in Springfield, Tennessee. In addition, each engine was retained for further examination at Continental Motors Inc., Mobile, Alabama, under the supervision of an NTSB investigator. Compressed air was applied into the fuel system from the base of each wing toward to outboard section of the wing, and there were no blockages noted in either wing. In addition, the fuel selectors, crossfeed functions, and fuel lines in the fuselage were tested with compressed air and no blockages were noted. There were no blockages or anomalies identified in the fuel system that would have precluded normal operation prior to the accident. Subsequent examination of the left engine revealed impact damage to the exhaust system, engine driven fuel pump, turbocharger, and the No. 5 cylinder, which precluded functional testing of the engine in a test cell. Both magnetos were rotated and produced spark on all towers. The fuel pump was disassembled with no anomalies noted. The fuel manifold valve was examined with no anomalies noted. The vacuum pump was disassembled and all vanes were intact. Overall, examination and testing of the engine and its components revealed no preimpact anomalies that would have precluded normal operation of the engine prior to the accident. The right engine was functionally checked in a test cell where it started immediately, accelerated smoothly, an