Bombardier Global 7500 incident near Riyadh, SA, March 31, 2022

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

Post-event inspection of a GE Aviation Passport 20-19BB1A turbofan engine found light sooting aft of the fuel nozzles, along with minor thermal damage to an air duct thermal sleeve consistent with an undercowl engine fire. This corroborates the pilot’s report of an in-flight engine fire during takeoff climb and during the throttle increases to troubleshoot the fire warning. The engine was removed and sent to GE for evaluation and a fuel system leak check was performed but the source/location of the fuel leak could not be identified. At several fuel nozzle locations, Nos. 12, 14, 16 and 18, the fuel nozzle pigtail-to-fuel manifold b-nut connections were shiny, lacked sooting, and were thought to be possible leak locations. The fuel pressure and nitrogen check pressure used during the leak tests were well below the fuel pressure the engine experienced during the event and were determined to be insufficient to replicate the fuel leak. A torque check of all the fuel manifold b-nut connections found five locations, all located on the left fuel manifold, which had very low torque values, less than 200 inch-pounds, when compared to the other b-nut connection locations and the required installation torque of 285 inch-pounds nominal. Two of the five - Nos. 16 and 18 - were identified as possible leak sources during the initial examination. Detailed analysis of the fuel nozzle No. 18 pigtail-to-fuel manifold b-nut connection revealed evidence of fuel manifold outer diameter surface galling and elliptical and intermediate contact marks indicative of misalignment contact with the fuel manifold female ferrule sealing surface and signs of contaminants and imbedded material (hard particles) on the fuel nozzle male bullnose sealing surface; no other b-nut connection location exhibited this type of surface distress. The source of the hard particle damage could not be positively identified but the observed diagonally orientated abrasive material marks/material removal was consistent with what appears to be a previous repair; there is no record of a repair operation being performed on this specific nozzle. Since the post-event fuel system leak checks performed by GE could not induce a fuel leak, the exact location/source of the fuel leak could not be determined by testing and observations. However, the combination of low fuel manifold b-nut torque, misalignment between fuel nozzle No. 18 pigtail-to-fuel manifold b-nut connection and sealing surface distress created the conditions, and sealing surface distress created the conditions by which there was insufficient sealing allowing fuel to leak during high engine power and high fuel pressures. No other fuel nozzle locations had this combination of factors that would have allowed a fuel leak.

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.