On February 5, 2021, about 1400 mountain standard time, a Hughes OH-6A helicopter, N6639R, was substantially damaged when it was involved in an accident near Horseshoe Bend, Idaho. The pilot sustained serious injuries and a passenger sustained minor injuries. The helicopter was operated as a public use flight. The purpose of the accident flight was to conduct wildlife damage management operations. While maneuvering at low altitude over hilly terrain, the pilot noted a loss of main rotor rpm. The pilot maneuvered the helicopter into a draw to reduce the engine power demand and regain rotor rpm. The helicopter impacted sloped terrain, rolled over, and came to rest on its left side. The pilot reported that, two days before the accident flight, he experienced a similar loss of rotor rpm while over flat terrain and was able to safely land the helicopter. While on the ground, he utilized the “BEEP” switch (N2 governor control switch) on the collective to increase rotor rpm, but it had no effect. He then changed the position of the throttle from fly to idle and observed the engine instruments for any abnormalities. With no abnormalities identified, he again changed the position of the throttle from idle to fly and regained main rotor rpm. He then utilized the “BEEP” switch and all engine controls returned to normal operations. While remaining on the ground, he repeated the process several times attempting to duplicate the loss of main rotor rpm with no success. As there had not been a prior malfunction and unable to duplicate the loss of rotor rpm, the pilot believed the event as an isolated anomaly and did not notify maintenance personnel. Postaccident examinations of the helicopter revealed substantial damage to the fuselage and tail boom, as well as an air leak at the fuel cell outlet valve that connected the fuel supply line from the fuel shutoff valve and the frangible in-line fuel valve on the self-sealing engine fuel inlet hose located at the firewall. (see Figure 1.) The fuel cell outlet valve along with the engine fuel inlet hose were shipped to the NTSB (NTSB) Materials Laboratory for further examination. Figure 1: view of fuel line leak (photo provided by MD Helicopters) Flight control continuity was confirmed from the cyclic pitch control and collective pitch control systems to the main rotor swashplate. Anti-torque pedal control continuity was established from the pedals up to tail rotor pitch control tube. The engine was removed and shipped to the manufacturer for additional examination and testing. The engine was installed in a test-run cell, where it was started and ran for about an hour with no mechanical anomalies identified that would have caused the reported sudden loss of engine power. Of note were two decelerations and a singular brief engine stall; both events were within the manufacturers experience of a mid-life engine and would not have contributed to the reported power loss. According to the engine manufacturer, the accident helicopter had a non-pressure-boosted fuel system that was utilized for normal operation. The engine fuel pump generated a vacuum at its inlet to draw the fuel from the tank to the engine fuel delivery system. A leak in the fuel delivery system prior to the engine can produce air entrainment in the fuel being supplied which could result in the interruption of homogenous fuel flow. Maintenance records provided by the operator indicated that a 100-hour inspection was completed on May 19, 2020. Correspondence with the maintenance facility revealed that a 100-hour inspection was conducted in accordance with the Rolls Royce maintenance and operation manual, 250-C20 Section 72-00-00. No other fuel system troubleshooting or repairs were completed. A vacuum check of the fuel system was conducted during the 100-hour inspection, with no anomalies noted. The Rolls Royce maintenance and operation manual, 250-C20 Section 72-00-00, titled “Engine Fuel and Control System -- Maintenance Practices,” contained multiple warnings that, “Air leaks in the fuel system or the pneumatic sensing system can cause flameouts, power loss or overspeed.” The fuel cell outlet valve was examined by the NTSB Materials Laboratory. A pressure test utilized soapy water as part of its testing protocols; soap bubbles formed at the barrel of the pipe and fuel outlet valve, which indicated the presence of an air leak. Visual examination of the fitting had circumferential scratch marks on the barrel of the pipe, as seen in Figure 2. A visual examination of the fitting threads revealed that the first two threads were damaged due to cross-threading. The O-ring was removed with both outer and inner diameter surfaces flattened/compressed. Metal chips were found embedded in the O-ring; when the O-ring was removed, a metal chip dislodged and revealed an impression in the rubber. Figure 2: View of optical microscope images showing circumferential scratch marks and pile-up material on the outer diameter surface of the pipe barrel (photo provided by NTSB) A review of the helicopter’s maintenance records could not determine the age of the fuel cell outlet valve, or if any maintenance had been performed on the valve.