Sikorsky S76 accident near Islamorada, Florida, December 15, 2017

The NTSB's narrative final · quoted from the NTSB record

The pilot landed the helicopter in an open field to pick up a patient. He reported that he positioned the engine controls to idle, and the main rotor speed (Nr) was less than 60% before he applied the rotor brake to stop the rotors. The rotors stopped, and the paramedic and flight nurse exited the helicopter. After they departed, the pilot noted that the No. 1 engine temperature was fluctuating with an increase in the inlet turbine temperature, so he shut down the No. 1 engine. The flight nurse then noticed dark smoke, sparks, and then flames emanating from the main rotor gearbox cowling area of the helicopter. The pilot was notified of the fire, shut down the No. 2 engine, and discharged both fire bottles, but the fire continued to burn until fire department personnel extinguished it. The thermal damage observed in the main rotor gearbox compartment appeared centered around the rotor brake. The rotor brake control components did not exhibit evidence of anomalies that would have either led to its uncommanded engagement with rotors turning or sustained engagement after disengagement of the rotor brake. Raised material found on the rotor brake disk surface was consistent with fused material from the rotor brake pads. Both forward and aft brake pucks extended when hydraulic pressure was applied to the brake calipers. The forward pucks of the left and right brake calipers did not automatically retract when hydraulic pressure was removed, but they were manually retracted without difficulty. The rotorcraft flight manual supplement (RFMS) for the accident helicopter mentioned the possibility of a rotor brake fire if pressure has been applied to the rotor brake system that resulted in a rotor brake puck dragging against a brake disk. However, with hydraulic pressure removed, it is unlikely the forward pucks of the left and right brake calipers would have asserted enough force on the rotor brake disk to increase friction between the puck and brake disk to cause these components to heat up enough to result in a fire. The No. 1 engine did not exhibit evidence of anomalous damage, and its FCU functioned normally on the test bench. There was no evidence of thermal damage to both engines’ exteriors. However, the gas generator turbines and thermocouples from the No. 2 engine exhibited thermal damage. Because thermal damage was not observed on the compressor stages, it is unlikely the thermal damage internal to the No. 2 engine was caused by the fire from the main gearbox compartment. Thus, it is likely the internal thermal damage to the No. 2 engine was caused by excess fuel flow into the combustion chamber of the engine. The No. 2 engine fuel control unit (FCU) functioned normally on the test bench; the FCU governed power turbine speed (Np) when its throttle was set to flight and stopped governing Np when its throttle was set to idle. A check of the rigging between the cockpit engine control levers (ECL) and the FCU throttles revealed no anomalies. Given that the engine FCU bench test was normal, it is likely that, after landing the helicopter, the pilot activated the rotor brake while the No. 2 engine FCU throttle position remained in the flight position and the No. 2 ECL was set to a condition to govern Np. As Nr, and consequently Np, dropped with the application of the rotor brake, the No. 2 engine FCU increased fuel flow to compensate, likely to its maximum rate, which resulted in thermal damage to the engine. The application of the rotor brake increased friction and heat as the additional fuel continued to provide power to the engine; thus, with the rotor brake force applied and with the No. 2 engine attempting to overcome it, the friction increased until a fire ignited near the rotor. The specific origin of the fire could not be determined; however, given the pilot's report that the area around the rotor brake assembly generally can accumulate slung grease and oil, it is possible that these materials, when exposed to the hot rotor brake disk, could have ignited and exacerbated the risk and spread of fire. A decal on the rotor brake lever cautioned that both engines must be shut down or one engine can be at idle for rotor brake application. The RFMS applicable to helicopters equipped with Arriel 1S1 engines installed indicated that rotor brake application was limited to one or two engines operating at idle or both engines shut down. Although a discrepancy existed between the decal and the RFMS, the Arriel 1S1 engine installation allows both engines to be at idle when applying the rotor brake; thus, the discrepancy was not a factor in this accident.

The complete narrative as the NTSB published it. The NTSB's docket holds the report as a PDF and any photographs, statements and other documents from the investigation.