CEN13FA196

The private pilot's inadequate response to the dual engine shutdown during cruise descent, including his failure to adhere to procedures, which ultimately resulted in his failure to maintain airplane control during a single-engine go-around. An additional cause was the pilot's decision to allow the unqualified pilot-rated passenger to manipulate the airplane controls, which directly resulted in the inadvertent dual engine shutdown.

According to the cockpit voice recorder (CVR), during cruise flight, the unqualified pilot-rated passenger was manipulating the aircraft controls, including the engine controls, under the supervision and direction of the private pilot. After receiving a descent clearance to 3,000 feet mean sea level (msl), the pilot told the pilot-rated passenger to reduce engine power to maintain a target airspeed. The cockpit area microphone subsequently recorded the sound of both engines spooling down. The pilot recognized that the pilot-rated passenger had shutdown both engines after he retarded the engine throttles past the flight idle stops into the fuel cutoff position. Specifically, the pilot stated "you went back behind the stops and we lost power." According to air traffic control (ATC) radar track data, at the time of the dual engine shutdown, the airplane was located about 18 miles southwest of the destination airport and was descending through 6,700 feet msl. The pilot reported to the controller that the airplane had experienced a dual loss of engine power, declared an emergency, and requested radar vectors to the destination airport. As the flight approached the destination airport, the cockpit area microphone recorded a sound similar to an engine starter spooling up; however, engine power was not restored during the attempted restart. A review of the remaining CVR audio did not reveal any evidence of another attempt to restart an engine. The CVR stopped recording while the airplane was still airborne, with both engines still inoperative, while on an extended base leg to the runway. Subsequently, the controller told the pilot to go-around because the main landing gear was not extended. The accident airplane was then observed to climb and enter a right traffic pattern to make another landing approach. Witness accounts indicated that only the nose landing gear was extended during the second landing approach. The witnesses observed the airplane bounce several times on the runway before it ultimately entered a climbing right turn. The airplane was then observed to enter a nose low, rolling descent into a nearby residential community. The postaccident examinations and testing did not reveal any anomalies or failures that would have precluded normal operation of the airplane. Although the CVR did not record a successful engine restart, the pilot was able to initiate a go-around during the initial landing attempt, which implies that he was able to restart at least one engine during the initial approach. The investigation subsequently determined that only the left engine was operating at impact. Following an engine start, procedures require that the respective generator be reset to reestablish electrical power to the Essential Bus. If the Essential Bus had been restored, all aircraft systems would have operated normally. However, the battery toggle switch was observed in the Standby position at the accident site, which would have prevented the Essential Bus from receiving power regardless of whether the generator had been reset. As such, the airplane was likely operating on the Standby Bus, which would preclude the normal extension of the landing gear. However, the investigation determined that the landing gear alternate extension handle was partially extended. The observed position of the handle would have precluded the main landing gear from extending (only the nose landing gear would extend). The investigation determined that it is likely the pilot did not fully extend the handle to obtain a full landing gear deployment. Had he fully extended the landing gear, a successful single-engine landing could have been accomplished. In conclusion, the private pilot's decision to allow the unqualified pilot-rated passenger to manipulate the airplane controls directly resulted in the inadvertent dual engine shutdown during cruise descent. Additionally, the pilot's inadequate response to the emergency, including his failure to adhere to procedures, resulted in his inability to fully restore airplane systems and ultimately resulted in a loss of airplane control.

Although not required, the airplane was equipped with an L-3/Fairchild model FA2100-1010 CVR, serial number 446023. The CVR recording contained about 31 minutes of digital audio, which was stored in solid-state memory modules. The CVR was not damaged during the accident and the audio information was extracted from the recorder normally. The recording consisted of four channels of audio information, ranging from good to excellent quality. The recording began at 1545:31 with the airplane established in cruise flight at 41,000 feet (FL410), and the recording stopped about 1616:32 while the airplane was maneuvering toward the destination airport with both engines inoperative. A transcript of the CVR audio information is included with the docket materials associated with the investigation. The airplane was not equipped with a flight data recorder, nor was it required to be so equipped. The accident flight was on an activated instrument flight rules (IFR) flight plan. A review of available ATC information indicated that the accident flight had received normal air traffic control services and handling. A transcript of the voice communications recorded between the accident flight and South Bend Approach Control are included with the docket materials associated with the investigation. The South Bend Airport (SBN), a public airport located approximately 3 miles northwest of South Bend, Indiana, was owned and operated by the St. Joseph County Airport Authority. The airport was a certificated airport under 14 CFR Part 139 and had on-airport fire and rescue services. The airport field elevation was 799 feet msl. The airport had three runways: runway 9R/27L (8,414 feet by 150 feet, asphalt/grooved); runway 18/36 (7,100 feet by 150 feet, asphalt/grooved); and runway 9L/27R (4,300 feet by 75 feet, asphalt). --- Sound Spectrum Study --- A study was performed to evaluate the sound spectrum of audio recorded by the cockpit area microphone after the loss of engine power at 1614:27. The CVR audio was compared with audio recorded during ground testing of an exemplar Hawker Beechcraft model 390 (Premier IA). The sound spectrum study indicated that, at 1615:02, the pilot engaged a starter motor in attempt to restart one of the engines. The study further established that the electrical noise from the engine igniters was not present at any point during the CVR recording, including the attempted engine air start. (The air start procedure required that the igniter switches be switched to the "ON" position before attempting any engine air start) A review of the remaining CVR audio did not reveal any evidence of another attempt to restart an engine. --- Surveillance Video Study --- There were several surveillance videos of the accident airplane during the two landing attempts, and the final descent and impact. A study of airport surveillance footage was completed to determine an average ground speed of the airplane during the second landing attempt. The study determined that the airplane's average ground speed was 127 knots (+/- 4 knots) during the 3.75 seconds of camera footage of the second landing attempt. Additional information concerning the surveillance videos can be found with the docket materials associated with this investigation. --- Mobile Device Examinations --- Several mobile devices were recovered from the wreckage and sent to the National Transportation Safety Board (NTSB) Vehicle Recorder Laboratory for examination. The pilot's tablet mobile device contained several aviation related applications; however, none of the applications contained flight track data for the accident flight. One application, ForeFlight, depicted the planned route-of-flight for the accident flight. Additionally, the ForeFlight application also contained 160 file-and-brief entries for previous flights. Another application, LogTen Pro, contained a partial flight history log. The pilot's mobile phone was reviewed and no information pertinent to the investigation was recovered. The pilot-rated-passenger's mobile phone contained a text message, dated March 13, 2013, concerning a previous flight that he had in the accident airplane with the pilot. No additional information was recovered that was pertinent to the investigation. Another passenger's mobile phone contained multiple out-going text messages with timestamps between 13:45 and 13:53 central daylight time. These text messages noted that the accident flight was about to takeoff and provided the expected time en route to South Bend. At 1505 eastern daylight time, a multi-media text message was sent with a photograph from inside the airplane cabin looking toward the cockpit. At 1612, another photo was taken from inside the cabin looking outside through a cabin window. No additional information was recovered that was pertinent to the investigation. --- Starter-Generator Examinations --- An initial visual examination of both starter-generators determined that their drive shafts were intact and the armatures rotated. The brush covers were removed and the brushes were observed to be in a good condition. The starter-generators were examined and tested at the manufacturer and no failures or anomalies were noted that would have prevented normal operation. --- Generator Control Unit Examinations --- Visual examination revealed the outer dust sleeve for the left generator control unit (GCU) was dented; however, further disassembly revealed no internal damage. The right GCU appeared to be undamaged. Both devices were examined and tested at the manufacturer and no failures or anomalies were noted that would have prevented normal operation. --- Battery Examinations --- During the on-site investigation, the no-load voltage of the main battery was 25 volts. Additional examination, at the manufacturer, confirmed that the battery was electrically intact and exceeded the acceptance test standards for a new battery. The standby battery was visually inspected at the accident site and no additional testing was completed. --- Throttle Quadrant Assembly Examinations --- The throttle quadrant assembly was removed from the airplane and examined at the manufacturer. A visual inspection revealed that both throttle levers were bent to the right and the fuel cutoff pull-up locks were jammed. There was foreign object debris, mostly loose attic insulation, found within the throttle quadrant assembly. To facilitate additional testing, the throttle arms were straightened to a vertical position. A partial Acceptance Test Procedure was completed because of existing damage to the throttle quadrant assembly. An electrical continuity check confirmed proper function of the throttle quadrant at each switch location. --- Engine Electronic Control Unit Examinations --- Both engine electronic control units (ECU) were examined and tested at the manufacturer on a Williams FJ44-2A engineering test cell. After a successful bit check at power-up, the contents of the ECU's non-volatile memory were downloaded. The examination of the recorded fault codes from each ECU determined no faults were recorded during the last flight in memory. Additionally, neither device contained any information regarding the engine operation during the last recorded flight. Additional component examination summaries are included with the docket materials associated with the investigation. One of the surviving passengers was interviewed by two NTSB Human Performance and Survival Factors investigators. The passenger reported that he loaded his luggage and computer gear on the airplane between 1330 and 1345 central daylight time. After loading, he and the other passenger boarded the airplane and waited for the pilots. Around 1350, the pilot and pilot-rated-passenger boarded the airplane. The passengers were not provided a safety briefing. He stated that the takeoff and cruise portion of the flight appeared to be normal; however, while the airplane was on a