HISTORY OF FLIGHT
On November 29, 2016, about 0850 central standard time, a GoJet Airlines Bombardier CRJ700, N367CA, equipped with two General Electric (GE) CF34-8C5B1 turbofan engines, experienced an uncontained No. 1 (left) engine failure during cruise (FL 340). The crew reported hearing a loud noise from the back of the airplane immediately followed by a No. 1 engine roll back and flame out. The crew declared an emergency, diverted to Lambert-St. Louis International Airport (STL), and made an uneventful single engine landing. The flight was being operated in accordance with 14 Code of Federal Regulations Part 121 and was a regularly scheduled flight from Cincinnati/Northern Kentucky International Airport (CVG) Covington Kentucky to Denver International Airport (DEN) Denver, Colorado as a Delta Air Lines connection. There were no reported injuries to the passengers or crew.
DAMAGE TO AIRPLANE
There was no damage to the airframe. A penetration was observed in the No. 1 engine low pressure turbine (LPT) case, but the released debris did not penetrate the engine nacelle.
TEST AND RESEARCH
Engine Disassembly
The incident engine, engine serial number (ESN) E965352, was shipped to Delta TechOps in Atlanta, Georgia for examination and disassembly. The LPT case had a penetration located at the 8 o'clock position, about 1 inch aft of the LPT flange, coincident with the high pressure turbine (HPT) 2nd stage rotor plane of rotation. The case material around the penetration was curled radially outward, away from the engine. Seven HPT 2nd stage blades were found separated below the blade platform and the remaining 61 blades exhibited impact damage and were separated above the platform at random locations along the blade span. Impact and thermal damage were observed on all four LPT rotor and nozzle stages.
Metallurgy
The complete set of 68 HPT 2nd stage rotor blades were sent to the GE materials laboratory in Lynn, Massachusetts for analysis. The seven blades that separated below the blade platform were all fractured at the shank upper radius and the fracture surfaces exhibited visual evidence of high cycle fatigue originating on the convex side from areas of corrosion. The 61 blades that were separated above the platform all failed due to overload, consistent with secondary damage. Of the 61 blades separated above the platform, multiple blades had corrosion and fatigue cracks present below the blade platform in the same general area as the blades that separated below the platform. The blade material composition and hardness levels met drawing specification.
The LPT case was sent to the GE materials laboratory in Evendale, Ohio for visual examination and material analysis. The case material composition and hardness levels met drawing specification. The LPT case penetration fracture surfaces had ductile dimples consistent with overload.
The NTSB materials laboratory reviewed both GE reports and concurred with all findings.