Boeing 767-338 incident near Ontario, California, January 29, 2021

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

The low pressure turbine stage 5 nozzle segment No. 3 was the only item identified that had any indication of primary fatigue, and no damage or distress was found upstream (forward) of the low pressure turbine that would have accounted for the damage observed throughout the low pressure turbine, the most likely initiating event was the fatigue fracture and liberating of the two missing stage 5 nozzle airfoils. The low pressure turbine stage 5 blades are located directly downstream (aft) of the stage 5 nozzle; thus, the loss of the two stage 5 nozzle airfoils would have traveled downstream in the airflow direction contacting and damaging the stage 5 blades resulting in stage 5 blade airfoil fractures. The damage inflicted on the stage 5 blades by the ingestion of the stage 5 nozzle airfoils would have created an imbalance in the low pressure turbine rotor. This low pressure turbine imbalance was confirmed by the flight data recorder data. Shortly after takeoff during climb the low pressure turbine vibration started to climb and within a few seconds reached the value of 5 cockpit units which is the maximum value that can be recorded; a cockpit unit is a dimensionless scaler unit, a magnitude with no unit of measure attached. GE performed a rotor imbalance and deflection analysis to determine the amount of radial deflection each stage of the low pressure turbine would be anticipated to experience based on the condition of the hardware as documented during the engine exam. Since the exact condition of the low pressure turbine hardware for any point in time during the event was not known the analysis does not provide the exact vibration levels or deflections at the time of the event or during the engine deceleration but instead a general assessment. The results of the analysis showed that the amount of low pressure turbine rotor expected radial deflection was several times greater than the nominal running blade radial clearances at takeoff for each of the low pressure turbine rotor stages. This is consistent with all the observed gouging and heavy wear down to the backing strip of the honeycomb on the blade outer shroud segments and the accompanying loss of blade tips for each stage of the low pressure turbine. The analysis also predicted that the low pressure turbine stage 5 would be the stage that experienced the most deflection and that the deflection would be the greater not at takeoff/climb low pressure turbine rotor speed but as it decelerates from the accumulation of damage and the pilot’s action to shut down the engine. This would indicate that the overall damage observed throughout the low pressure turbine was initially caused by the loss of the low pressure turbine stage 5 nozzle airfoils due to fatigue and impacting the stage 5 blades fracturing them creating imbalance and deflection in the low pressure turbine rotor. The low pressure turbine rotor imbalance and deflection progressively increased due to the accumulation of additional stage 5 blade damage and the accompanying loss of rotor speed that eventually led the entire low pressure turbine rotor to lose radial blade clearance. This loss of radial blade clearance throughout the low pressure turbine resulted in blades contacting static structure, fracturing, and causing additional downstream damage. Since the flight data recorder does not record vibration levels or amplitudes above 5 cockpit units, any vibration values higher than 5 cockpit units is capped to 5 cockpit units, the exact vibration the right engine experienced was unknown; however, the results of the GE imbalance and deflection analysis indicated that the loads experience by the right engine would have been sufficient to fracture the oil supply tube and the turbine exhaust sleeve. Since neither of these items showed signs of a pre-existing anomalies, their failures were as a resulted of the high vibrational loads from the right engine during the failure sequence. The low-grade thermal/fire damage observed on the inside of the right engine core cowls and thrust reverser halves and on the outside of the right engine was due to the oil from the fractured oil supply line contacting hot engine cases and smoldering and igniting. The fuel system, forward of the left and right fuel manifold where the thermal/fire damage was most pronounced, was leak checked and no leaks were found; thus, the only source of a flammable fluid would have been the fractured oil supply line.

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.