Primary finding
Probable cause
Loss of engine power due to the pilot inadvertently selecting engine IDLE mode while in flight, which reduced rotor and engine speed to a level that could not sustain flight.
Investigator assessment
Analysis narrative
The pilot was performing a human external cargo (HEC) long-line qualification exam flight at the training operations facility of a public utility company. After performing the initial maneuvers, the pilot transitioned to the HEC phase of the exam by carrying a lineman on a 60-ft long line. While maneuvering the helicopter at an altitude of about 175 ft above ground level (agl), the helicopter lost engine power. In response, the pilot selected what he thought was the “emergency” throttle detent, but the engine did not respond. The pilot maneuvered the lineman away from the landing helicopter and performed an autorotation. The helicopter landed hard and sustained substantial damage. The pilot sustained serious injuries and the lineman sustained minor injuries. A complete postaccident examination of the airframe, engine, and engine control systems did not reveal any evidence of preimpact mechanical malfunction or failures. The helicopter contained fuel, was loaded within its envelope, and weather was not a factor. The engine was test run in a test cell and it performed nominally. Onboard video recorded a section of the annunciator panel that showed some, but not all, engine warning lights, none of which illuminated at any point during the flight. The recording did capture an audio tone as the descent began that was the same frequency as the engine out and low rotor warning indicators. Review of data recorded by the engine control unit (ECU) revealed that at the time of the loss of engine power, the throttle was in the “IDLE” rather than the “FLY” position as would have been required during flight. A few seconds later, the throttle position moved back into the “FLY” range. The pilot was flying the helicopter solo from the right seat with the door removed so that he could lean outside and observe below. The collective control, which housed the engine throttle twist grip at its end, was mounted on the pilot’s left side. The direction of throttle rotation meant that if the pilot maintained a tight grip on the throttle as he leaned right and out of the helicopter, the direction of throttle rotation was such that his hand could have inadvertently rolled the throttle out of FLY and into IDLE. Under this scenario, the engine would lose power immediately; both audio and engine diagnostic data confirmed this had likely happened. The pilot almost immediately rolled the throttle back into the FLY position but the design of the engine control system was such that it assumed the helicopter was on the ground and therefore limited power application to prevent an over-torque condition. So, even though the pilot immediately remedied the problem, the engine would not have regained power. The pilot stated as soon as he noticed a loss of power he moved the throttle beyond “FLY” and to the “emergency” position. However, although this position could be selected, it had no function on this type of helicopter; it was a hold out from the previous version of the helicopter that the pilot had flown. This mistake had no bearing on the accident and would not have affected the outcome. The HEC operation required that the helicopter be flown at a low height and speed such that it was operating within the “AVOID” range of its Height-Velocity Envelope. In this range, a safe landing could not be guaranteed following an engine failure. Additionally, the nature of the training facility, with multiple training towers and obstacles in the immediate vicinity, further reduced the likelihood of a successful outcome in the event of an emergency. Therefore, although the pilot made the mistake of inadvertently shutting down the engine in flight, he was still able to demonstrate the presence of mind and helicopter control to maneuver the helicopter away from the lineman and into an open area. This was the second time the pilot had taken the HEC exam, having failed the first attempt. The video recording appeared to show the pilot was anxious throughout the flight; however, considering the nature of the work he was performing and the stress he would have been experiencing, this is understandable. It could not be determined if this contributed to the accident.
Source record
Factual narrative
In 2019, the helicopter was converted from its original Rolls Royce 250-C47 engine to a Honeywell HTS900-2-1D. The conversion was performed in accordance with the Eagle Copters, Eagle 407HP supplemental type certificate (STC) SR03496NY. The Honeywell engine was controlled by a Full Authority Digital Engine Control (FADEC) system consisting of two redundant electronic control units (ECUs) along with a fuel metering unit (FMU) assembly, permanent magnet alternator, and other engine and airframe sensors. The FADEC system was designed to provide automatic control of the engine during startup, steady state, and transient operation and ensure that operation stayed within specified envelopes. Pilot-commanded throttle position was monitored by the ECUs through a dual linear variable differential transformer (LVDT) assembly, mounted to the airframe and connected through a series of bellcranks to a conventional twist grip in the collective control column. The helicopter was configured for operation from the right seat, with the collective control on the pilot’s left side. The twist grip had three distinct detent positions: OFF, IDLE, FLY, and an unmarked over-center position beyond FLY. Although it was available, this over-center position had no functional use in the Honeywell engine-equipped version of the helicopter. In the original Rolls-Royce engine-equipped helicopter, the over-center position was marked as the “MaxNG” detent. This position allowed the pilot to manually maintain main rotor/power turbine (Nr/Np) speed by going beyond the FLY (90% gas producer [Ng] speed) detent in the event of a FADEC failure. For both configurations, to reduce engine power, rotating the twist grip from FLY to IDLE would have required a rotation in the clockwise direction when facing forward. Additionally, a lockout button was included that needed to be pressed to move the control from IDLE to OFF. The pilot reported about 3,500 hours of total flight experience including about 1,000 hours as a flight instructor and agricultural pilot. He started working for the helicopter operator about 15 months before the accident. He reported a total of almost 175 hours flight experience in the Bell 407, 10 of which were in the Honeywell engine-equipped 407HP version, all in the accident helicopter. The pilot’s total FAR Part 133 external load time before the accident totaled about 87 hours. Most of the flight time took place in the Eurocopter AS-350, with 2.3 hours in the Bell 407, and 6.7 hours in the 407HP. Two of the flights involved HEC, with all other flights involving training or carriage of cargo or buckets, all flown from the right seat. The accident flight was a demonstration flight to PG&E so he could be signed off to perform HEC flights for them. According to the helicopter operator, this was the second time he had taken the check ride, having not passed on the first attempt. The night before the accident he stayed in a hotel and went to bed about 2100. He awoke at 0630 and reported having a normal night sleep and was well rested. The pilot was 6 ft tall. On May 11, 2022, about 0930 Pacific daylight time, a Bell Helicopter Textron Canada 407, N999GH, was substantially damaged when it was involved in an accident near Livermore, California. The pilot was seriously injured and the lineman crew member sustained minor injuries. The helicopter was operated as a Title 14 Code of Federal Regulations Part 133 rotorcraft external-load flight. The pilot was performing a HEC long-line qualification exam flight at the Livermore Electric Safety Academy, a training operations facility owned by the Pacific Gas and Electric Company (PG&E). The pilot was flying the helicopter solo from the right seat with the door removed so that he could lean outside and observe below. The maneuvers were observed and monitored on the ground by a group of examiners. The pilot stated that the first 20 minutes of the flight were normal and the accident occurred during the final phase while carrying a PG&E lineman on a 60-ft long line. The pilot was maneuvering the helicopter at an altitude of about 175 ft agl, positioning the lineman against a static line marker ball midway along a 90-ft wire that spanned an 88-ft lattice tower and a 65-ft wood pole. The pilot reported that the engine then lost power and he rolled the throttle out of the “FLY” detent and into the “emergency” range, but the engine did not respond. He attempted to maneuver the lineman into a clearing and partially raised the collective once the lineman was about 15 ft agl to cushion the lineman’s landing. The pilot stated that he then maneuvered the helicopter to a landing spot away from the linemen and, after reaching an altitude of about 30 ft agl, he raised the collective and dropped the HEC line and cargo hooks; the helicopter landed hard. The pilot stated that the engine was still running after landing and he was able to then shut off the main fuel valve and battery. He was extricated from the helicopter about 30 minutes later. Multiple witnesses recounted observations that matched the pilot’s statement. All stated that the helicopter appeared to be operating without issue throughout the maneuvers until they heard a change in engine tone, with some then observing the main rotor blades slowing as the helicopter began to rapidly descend. A video provided by a witness showed the helicopter descending just before impact. It appeared intact and was not emitting smoke or vapors. The helicopter came to rest on its belly and was tilted slightly to the left (see Figures 1 and 2). The landing skids had spread outwards on both sides. The primary airframe structure was largely intact but had buckled in the roof area below the main transmission, which had moved out of position and was tilted slightly down. All four main rotor blades remained attached to the hub assembly; one blade was essentially undamaged and the remaining three had sustained delamination, aft bending, and buckling damage from midspan outwards. The tailboom remained attached to the aft bulkhead, and both tail rotor blades had detached with their roots remaining on the hub. Figure 1– Helicopter at the accident site (photo courtesy of the FAA). Figure 2 - Accident location within training facility (photo courtesy of PG&E). The engine was still attached to its mounts and appeared to have sustained minimal damage. The engine output drive coupling had separated where it joined the main transmission. The damage to the coupling appeared to be consistent with movement of the main transmission at impact. There was no evidence of catastrophic engine failure or foreign object ingestion. The transmission drivetrain was continuous from the engine output drive coupling through to the main gearbox and both the tail and main rotor blades; the freewheeling clutch was operational. The flight controls sustained damage consistent with impact at the base of the main transmission assembly and the pilot’s collective control but were otherwise intact. Video The helicopter was equipped an Appareo Vision 1000, which was a self-contained image, audio, and parametric data recorder. The parametric data included GPS aircraft time, position, altitude and speed, along with 3-axis acceleration with derived pitch, roll, and yaw. The unit was mounted on left side of the aft cabin bulkhead, facing forward. The field of view was focused to the left and downward. As such, only a portion of the instrument panel, the left segment of the annunciator panel, and some externally visible features through the helicopter's left chin bubble and left windscreen were visible. The pilot's body was not captured, nor were any of his control inputs. The recording contained a stereo audio track. One audio channel was of a cockpit area microphone, and the other was recorded from the helicopter's intercom. The intercom track recorded continuously and capt