ERA14FA282

The failure of the pilots to maintain airspeed while maneuvering, which resulted in the airplane exceeding its critical angle of attack and experiencing an aerodynamic stall. Contributing to the accident was the loss of engine power for a reason that could not be determined by the postaccident examination, which was limited due to impact damage.

According to a friend of the airplane's owner, the purpose of the flight was for the owner to check out another pilot in the airplane. It could not be determined which pilot was flying the airplane at the time of the accident. Review of uncorrelated radar data indicated that the flight departed and maneuvered in the local area for about 26 minutes before the accident occurred. One witness stated that the airplane was about 200 ft above ground level when it entered a turn, and then its nose dropped and it descended to the ground. Two other witnesses reported seeing the airplane descending in a nose-down attitude. The witnesses provided conflicting information as to whether or not the airplane's engine was producing power. Examination of the accident site indicated that the airplane impacted in a steep descent. The witness observations and the impact geometry are consistent with the pilots failing to maintain adequate airspeed while maneuvering, resulting in the airplane exceeding its critical angle of attack and experiencing an aerodynamic stall. The propeller blade signatures were consistent with the engine not producing power at impact. Engine parameter data downloaded from the full authority digital engine control's (FADEC) data recording device revealed normal rpm, cylinder head temperature, and fuel pressure readings from takeoff to the end of the recorded data, and no FADEC fault codes were recorded. However, the recorded data ended before the loss of control occurred. Postaccident examination of the engine powertrain, fuel distribution block, and fuel injectors revealed no evidence of preimpact failure or malfunction. Both of the engine's electronic control units sustained impact damage that precluded operational testing. Although the auxiliary fuel pump was determined to have been inoperative for a long period of time before the flight due to separation of one electrical wire near the pump, the engine-driven fuel pump was operational and capable of providing adequate fuel to the engine to sustain engine power. The investigation could not determine the reason the engine was not producing power at impact.

The airplane was equipped with an Engine Date Interface (EDI), TCM P/N 657230 Rev A, P/N 14049 B, S/N 0643005. The EDI was retained by NTSB for read-out by the NTSB Vehicle Recorder Division, located in Washington, D.C. According to the NTSB Vehicle Recorder Division Report concerning the EDI, the file structure on the card was found to be corrupted due to the rapid removal of electrical power as a result of the accident. The file structure was rebuilt using data recovery software and the recorded data was extracted normally. The last file recorded identified as "edi0036.dal" contained what appeared to be three flights. The 1st flight recorded from takeoff to landing was approximately 3426 seconds, or 57.2 minutes long. The 2nd flight in the file was about 2126 seconds, or 35.4 minutes long and the 3rd flight was approximately 98 seconds long. Without having the date and time accurately recorded in the data files no positive determination could be made as to what flight segment was the accident flight. There were no engine or engine controller faults recorded during the last recorded data file (edi0036.dal) which included the accident flight. Correlation of the engine data from elapsed time to the event local time, EDT, was performed with an offset to the time of the accident. The recorded data in approximately 1 second increments associated with the last file revealed engine start occurred about 2017:04, and the data continues without interruption until 2036:00, resulting in approximately 19 minutes of recorded data. The report also indicates correlation of data associated with engine start, taxi, and run-up. For about 1 minute 3 seconds after takeoff, or to the end of the recorded data, the rpm was noted to be between approximately 2,500 and 2,600, the fuel pressure increased from 39 psi to about 56 psi (within normal green arc range), and all readings for cylinder head temperature were above the minimum reported values. A copy of the report and downloaded data as an attachment are contained in the NTSB public docket. The Merritt Island Airport is a publically owned uncontrolled field with a published common traffic advisory frequency (CTAF)/UNICOM of 122.975. The frequency is not recorded. A fixed base operator (FBO) at COI has a VHF transceiver; however, they closed at 1900 hours that day, and the CTAF/UNICOM frequency is not recorded. Correlation of the permanent maintenance records with the hand written entries made in an "Aircraft Flight Log" pertaining to the FADEC entries indicates there was no maintenance record entries regarding any troubleshooting that was performed. It was also noted there were several entries in the "Aircraft Flight Log" which indicate there were no FADEC annunciations. Excerpts from the airframe and engine permanent maintenance records as well as the "Aircraft Flight Log" are contained in the NTSB public docket. Examination of the auxiliary fuel pump was performed at the manufacturer's facility with FAA oversight which revealed the electrical wires and non-Weldon supplied connector were protected by heat-shrink tubing. No damage to the connector, heat shrink, or wiring was noted. The pump was then electrically connected but the device exhibited results consistent with an open electrical circuit (it did not draw current when energized). Continuity testing between the positive and negative terminal of the electrical connection of the pump-motor revealed an infinite resistance consistent with an open electrical circuit. Hand movement of the wires with the meter connected did not result in any resistance reading. Both electrical wires of the pump-motor were then cut along their length, and power was applied to the pump/motor but a dead short electrical circuit/locked rotor condition was noted. The motor was unbolted from the pump assembly, and there was "alarmingly little evidence of normal operational wear in the motor to drive interface", and there was no evidence of the drive cavity having been exposed to anything out of the ordinary. Electrical power was then applied to only the motor assembly via the cut wires and the unit was found to operate normally. The pump was then attempted to be rotated using a hand tool and was unable. Disassembly inspection of the pump revealed internal contamination consistent with long-term inactivity of the pump. As part of the inspection the shrink wrap was carefully cut parallel to the wires to expose the lead wires. As the shrink wrap was moved away, the negative (black colored wire) fell freely away. Cursory inspection of the negative lead wire revealed no evidence of any heat damage consistent with the high current degrading the termination. Following the inspection of the pump, it was subsequently returned to the storage facility, and then sent to the NTSB Materials Laboratory for examination. A report from the manufacturer and concurring statement from the FAA inspector are contained in the NTSB public docket. According to the NTSB Materials Laboratory Factual Report concerning the inspection of the broken conductor strands of the black wire of the auxiliary fuel pump, the ends of wire strands were still visible inside the connector. The insulation around the wire had evidence of material flow, and ridges were observed on one side of the wire. The side of the wire diametrically opposite the ridges was thinned due to the material flow. Smearing covered portions of the fracture surfaces on the strands, and the fracture surfaces on a few strands were completely obscured by damage. The undamaged areas of the strand fracture surfaces had ductile dimples consistent with overstress separations resulting from tension; the ductile dimples were elongated in one direction on several fracture surfaces consistent with overstress in shear. No melting was noted on any of the strand ends. The terminals for the black (negative) and orange (positive) colored wires were removed from the connector for further examination using a Liberty Aerospace technical document for reference. The terminal for the black wire was determined to have the crimp too far forward and an oversize bellmouth, while the terminal for the orange also had an oversize bellmouth and a Banana crimp, or bending of the terminal. Additionally a pinched strand was observed in the terminal for the orange colored wire. The NTSB Materials Laboratory Factual Report is contained in the NTSB public docket. Examination and bench testing of the engine-driven fuel pump was performed at the manufacturer's facility with NTSB oversight. The testing was performed to the standards for a new production fuel pump, and did not take into account adjustments for the airframes fuel system. Following inspection which documented the damage, the pump was placed on a test bench and a small leak (approximately 2 drops every 3 seconds) was noted originating near the diaphragm. Less than specified discharge fuel pressure was noted at Test Points, 3, 4, and 5. The fuel discharge pressure at Test Point 4 which equates to full RPM of the pump was 7.33 psi less than the minimum specified; however, according to personnel of the engine manufacturer, the out of tolerance condition is adjustable and likely would have no effect. A copy of the inspection and bench test results are contained in the NTSB public docket. Correlation of the "Engine Runup" plastic covered checklist that was found in the wreckage with the version from the FAA Approved Airplane Flight Manual (FAA Approved AFM) revealed omitted items, incorrect sequences, abbreviations, and incorrect information. For example, item 5 of the checklist found in the wreckage indicates the "Fuel Boost Pump Mode Switch" be turned "ON", while the FAA Approved AFM indicates for step 6 that the "Fuel Boost Pump Mode Switch" be turned off. Both checklists then indicate to verify that the "HSA Fuel PMP Annunciator" to "CHECK ON." A search of the internet produced an exact copy of the "Engine Runu