WPR16LA105

The pilot's decision to conduct an upslope, downwind takeoff combined with an improper flap setting, which resulted in the airplane's inability to clear trees beyond the runway end. The reason for the improper flap setting could not be determined.

The passenger, who was a student pilot, recently purchased the airplane in an estate sale. He and the airline transport pilot, both of whom lived in Mississippi, had traveled to California to retrieve the airplane and fly it back to Mississippi. Before the purchase, the airplane had not been maintained, operated, or flown in almost 11 years. Following the purchase, the owner contracted with a mechanic in California to ensure the airplane was in airworthy condition, which the mechanic reportedly did. The day before the accident, the pilot and owner took the airplane for its first flight after its dormant period and flew one uneventful circuit in the airport traffic pattern, as planned. The following day, the pilot and owner planned to fly the airplane for some systems evaluations. During that takeoff attempt from runway 9, the airplane became airborne but failed to climb and struck trees and terrain beyond the runway end. Although the pilot believed that he was taking off into the wind, witness statements and other evidence indicated that the takeoff was attempted with an approximate 5-knot tailwind. The first 1,000 ft of the runway was level, but the remaining 2,000 ft was sloped uphill. Although the Pilot's Operating Handbook specified using 15° flaps for takeoff, and the pilot reported that he used that setting and did not alter the flap position during the flight, the flaps were found to have been fully retracted at impact. Surveillance camera imagery captured about 2 seconds of the flight, when the airplane was about midfield and 4 ft above ground level (agl). Review of that imagery and audio data indicated that the ground speed was about 68 knots and that the engine speed was about 2,640 rpm; both values were consistent with normal takeoff values. However, the exact winds (and thus airspeed) were unknown, and because the propeller was a constant-speed model, nominal takeoff rpm could be achieved even if the engine was not developing full-rated power. Detailed examination of the airplane, including the engine, revealed that, although its condition was not in accordance with Federal Aviation Administration and manufacturer guidance, none of the observed deficiencies could have caused or contributed to the loss of climb performance, except for one magneto that was found to be mistimed to the engine by 7°. Evidence suggested that this was likely a result of the accident but that could not be determined with certainty. Performance calculations conducted by the airplane manufacturer, which accounted for most of the known takeoff conditions, including fully retracted flaps, indicated that the distance to 50 ft agl was slightly more than the available runway. The estimated airplane takeoff weight was about 300 lbs (11%) below the maximum takeoff weight that was used in the calculations, which would yield better performance than the calculated results. However, those calculations did not account for off-nominal values of the many other variables that could adversely affect takeoff performance, including pilot technique, airframe and engine deterioration, and inaccurate or improperly set instrumentation and controls. Thus, although a successful downwind takeoff with no flaps was unlikely, it might have been possible, but there were too many other unknowns to determine its likelihood with greater certainty. The reason(s) for the retracted flaps could not be determined. It is possible that the pilot forgot to extend them or that they were inadvertently and unknowingly retracted. Given the location of the flap control switch and its design (momentary, paddle-type), it is possible that the pilot extended the flaps to the proper takeoff setting of 15° but that they were subsequently retracted when the nonpilot passenger inadvertently contacted and actuated the flap control. The size and location of the flap position indicator gauge, combined with the location of the flaps (behind the pilot on the low-wing airplane), minimized the possibility that the pilot would notice that they had been retracted.

E45 was situated at an elevation of 2,933 feet above mean sea level, and was equipped with a single paved runway designated 9/27, which measured 3,624 ft by 50 ft. The western-most 1,000 ft segment of the runway was relatively level, but then the runway sloped uphill beyond that (towards the east). The slope of that uphill portion was not constant; a maximum up slope of 1.8 per cent was present for the segment between 2,500 and 3,000 ft from the threshold of runway 9. The overall average gradient was 1.1 per cent. There was a 100 ft gravel overrun at the east end of the runway. Beyond that, the terrain descended about 30 feet, but that region was populated with numerous trees as high as about 100 feet. Mechanic and Maintenance Record Information According to the pilot and the owner, a few weeks prior to the accident, the owner had contracted with a mechanic at E45 to conduct an annual inspection on the airplane, and to perform the maintenance necessary to render the airplane airworthy for its return to service. They also reported that subsequent to the maintenance, and prior to the accident flight, the mechanic made airframe and engine logbook entries that indicated that the airplane had been inspected in accordance with an annual inspection, was in airworthy condition, and that the mechanic's signature block denoted that he was an IA. Subsequent to the accident, the mechanic refused to provide the logbooks to the owner. The mechanic claimed that the owner owed him $6,000 for the maintenance that he had performed, and that he was retaining the logbooks for security until he was paid. FAA and NTSB attempts to convince the mechanic to release the logbooks to the FAA or NTSB were unsuccessful; again the mechanic stated that he was holding the logbooks as security until he was paid by the owner. The mechanic eventually allowed an FAA inspector to examine and photograph the two most recent entries in each logbook. The FAA inspector, and his photographs, indicated that portions of the original airframe and engine logbook entries by the mechanic had been altered with "whiteout" and overwritten. The revised airframe and engine entry text indicated that the airplane had been inspected in accordance with a "ferry inspection," and the revised mechanic's signature block indicated that he was an "A&P." "Ferry inspection" is not a term that is defined, referenced, or otherwise recognized by the FAA. FAA and NTSB conversations with other aircraft owners at E45 revealed that subsequent to March 31, 2015, the mechanic had continued to represent himself as a valid IA holder, and that he had conducted and signed off numerous aircraft as an IA. Airplane Performance Takeoff performance distance data (ground roll, and total over 50 ft obstacle) for the airplane were presented in table form in the POH. The performance table values were predicated on the following fixed conditions: - Gross weight: 2,750 lbs - Engine/propeller rpm: 2,700 - Engine leaned "to field elevation" - Flaps: 15º - Landing gear retracted after lift-off - Runway: paved, level, dry surface - Takeoff speeds: lift off, 71mph; 50 ft height, 75 mph The table provided for variations in the following parameters: - Headwind (no tailwind accountability) - Pressure altitude - Ambient temperature Because the POH performance data did not account for runway slope, tailwind, or 0º flaps, the manufacturer provided calculated performance estimates that accounted for variations in those parameters for two example cases. The first case used the prescribed takeoff flap setting of 15º, and the second used the actual takeoff setting of 0º. Both cases use the calculated pressure altitude, a 1.1% runway upslope, a 5 knot tailwind, and all other fixed parameter values specified above. The 15º flap case resulted in an estimated ground roll distance of about 1,900 ft, and an estimated distance to 50 ft agl of about 3,250 ft. The 0º flap case resulted in an estimated ground roll distance of about 2,300 ft, and an estimated distance to 50 ft agl of about 3,700 ft. It should be noted that these results do not represent certificated performance, and should not be construed as such. The pilot estimated that the airplane actually weighed about 2,460 lbs for the takeoff. Although the POH performance table included a "NOTE" that provided a means to account for weights below 2,750 lbs, there was insufficient data to substantiate application of that correction factor to these two performance cases. Other factors that can adversely affect takeoff performance, but whose specific values and effects could not be determined for this accident, included: - Pilot techniques (engine leaning, airspeed, attitude) for the takeoff - Airspeed indication system accuracy - Engine, propeller, and airframe deterioration due to age, use, and care - Engine rpm (tachometer, governor) - Propeller blade pitch - Ambient conditions (wind, temperature) Airport Surveillance Video There was a fixed-view surveillance camera mounted on a building on the north side of the runway at E45. The image and audio data from the camera was recorded, and the accident takeoff was captured. The data recording of the takeoff was provided to the investigation for review. The camera view was oriented perpendicular to the runway, looking south. The camera was situated approximately 2,500 feet along the runway from the west (9) threshold. Its field of view encompassed the runway segment approximately 155 feet to either side, for a total field of view of about 310 feet of the runway. The airplane traversed the image from right to left. The airplane was already fully in the frame at the beginning of the image file. When it first appeared, it was airborne, with its landing gear extended, and the landing gear appeared to be about 2 feet above the runway. When it exited the frame, the gear altitude appears to have increased to about 6 feet. When the airplane was in mid-frame, perpendicular to the camera, the airplane attitude was measured to be approximately 11º nose up. The flaps appeared to be retracted, but the image resolution was insufficient to positively ascertain the flap position. Analysis by NTSB Recorders Laboratory personnel indicated that the airplane operated at an average ground speed of about 68 kts during the nearly 2 second period that the airplane was visible in the image. During the first 40% of that time, the estimated average speed was about 66 kts, and during the last 60% it was about 70 kts. A frequency analysis of the audio recording of the engine/propeller was conducted by NTSB Recorders Laboratory personnel. The analysis indicated that the engine speed was 2,640 rpm. E45 was not equipped with any official weather sensing or recording equipment. Resident and eyewitness reports indicated that about the time of the accident, the temperature was about 75 degrees F (23 C), and there was a light wind from the west. An individual who was a flight instructor and FAA-designated pilot examiner estimated that the tailwind component along runway 9 was about 5 knots. Calculations using the available information indicated that the temperature was about 14 degrees C above the standard atmosphere value ("ISA"), and that the resulting density altitude was 4,686 feet. Pilot The pilot reported that for both flights, he was seated in the left front seat, and was the sole manipulator of the controls. He held an airline transport pilot certificate, and reported about 22,800 total hours of flight experience, including about 4,310 hours in single engine airplanes. Prior to his flight in the airplane the day before the accident, the pilot had no experience in the accident airplane make and model. His most recent flight review was completed in May 2015, and his most recent Federal Aviation Administration (FAA) third-class medical certificate was issued in January 2015. Owner The owner was seated in th