After ten years operation, one L-1 blade of steam turbine with large output power suddenly fell down. All 6 LP rotors (two machines) were checked and many cracks on L-1 blades were found. Due to economic reason, new blades were made with same geometry, but with better material in term of yield limit. Some L-1 stages were made from blades of original material, the rest was made from new material blades. Also the tip-timing measurement was installed on two L-1 stages to monitor blades. The investigation of blade cracks reason started. After one year of smooth operation there was an inspection. Surprisingly, the cracks were indicated again. The blades from new material had relatively more cracks than the original blades. This part describes some computational analyses which were done. For example, the modal analysis and forced vibration of mistuned bladed disk and of whole rotor train with attached L-1 blades, CFX and CFD calculations were done. The measurement of rotor torque, damping, Tip-Timing, experimental modal analysis of blades and rotor train and material test are presented in Part 1 – [7]. Application of both approaches (experimental and computational) rejected some hypotheses and revealed which hypotheses should be analyzed in a deeper way. The unstalled flutter has been identified as the most probable blade failures.

This content is only available via PDF.
You do not currently have access to this content.