ERA20FA074 · Robinson R66

The helicopter pilot was conducting a night cross-country flight in visual meteorological conditions. Recorded data showed that while the helicopter was flying at an altitude of about 2,300 ft mean sea level and at an indicated airspeed of about 107 knots, slight pitch and roll oscillations occurred for about 20 seconds followed by a left roll that continued until the helicopter was inverted. A main rotor blade contacted the tailboom, leading to an in-flight breakup over a residential area. Postaccident examination of the airframe, flight controls, and engine assembly revealed no evidence of a preimpact failure or malfunction.

Because the helicopter’s pitch and roll values changed only minimally before the in-flight upset, the autopilot was likely engaged when the pitch and roll oscillations started. The pitch and roll oscillations might have been the result of the autopilot reacting to an external disturbance, specifically, the moderatetosevere turbulence that was reported in the area. Another possible cause of the pitch and roll oscillations was the pilot manually manipulating the cyclic control. Additionally, an in-flight malfunction of the autopilot’s roll servo actuator could not be ruled out as the source for the pitch and roll oscillations. While any of these scenarios might have precipitated the initial oscillations, none would have precluded the pilot from overriding the autopilot and manually flying the helicopter.

Even though the helicopter was equipped with devices that recorded flight and engine data at a rate of at least one parameter every second, the lack of an on-board crash-resistant cockpit imaging system precluded a determination of the pilot’s actions, or lack of actions, in response to the pitch and roll oscillations. About 6 months after this accident, the National Transportation Safety Board issued a safety recommendation requesting that major helicopter manufacturers, including the manufacturer of the accident helicopter, install crash-resistant flight recorders with cockpit imaging systems for existing helicopters. The helicopter manufacturer responded that it incorporated a cockpit camera system as standard equipment on all similar model helicopters manufactured beginning the year after the accident and offered as a kit for existing helicopters the same year. The cameras were not designed to be crash-resistant.

The pilot was not likely impaired or incapacitated by his diagnosed obstructive sleep apnea given the information about his effective continuous positive airway pressure device use. Also, some or all the ethanol detected in the pilot’s specimens might have been from a source other than ingestion. The possibility of an impairing or incapacitating medical event or impairing effects from the pilot’s use of clonazepam (alone or in combination with ethanol effects) could not be determined from the available evidence. However, no operational evidence indicated that the pilot’s performance was deficient during the flight time preceding the oscillations.

Verbatim NTSB analysis and probable cause from the NTSB dataset