Early Stages of Gear Contact Fatigue Detection Thougth Residual Stress variation Using Magnetic Barkhausen Noise

Tuesday, October 1, 2024: 9:00 AM
25 B (Huntington Convention Center)
Mr. Bruno Henrique Lima , Aeronautics Institute of Technology, Sao Jose dos Campos, Sao Paulo, Brazil
Prof. Ronnie R. Rego , Aeronautics Institute of Technology, Sao José dos Campos, São Paulo, Brazil
This study addresses the Magnetic Barkhausen Noise (MBN) technique as a non-destructive testing method to evaluate the surface integrity characteristics in the early stages of contact fatigue in gears. The primary objective is to evaluate the MBN signal evolution during the lifetime of gears, specifically aiming to detect contact fatigue failures before any visible damage appear at the flank surface. Fatigue testing was conducted on five gear samples, inducing a natural evolution of gear contact fatigue. Monitoring MBN signals at regular intervals during testing cycles allowed for correlation with surface integrity degradation. Furthermore, the study delves into microstructural aspects related to contact fatigue, exploring various stages in the MBN evolution curve. The MBN technique was employed to characterize magnetic response variations during the initiation of contact fatigue mechanisms. In-depth analyses of residual stresses and microhardness provided a comprehensive understanding of surface degradation. A substantial increase in the MBN signal was identified before fatigue failure, indicating microstructural alterations affecting magnetic properties. Early contact fatigue stages were characterized by surface softening in the near-surface region, up to approximately 40 µm depth, accompanied by a less compressive residual stress region at 20 µm depth. The study also observed a lower influence of microstrains on the diffractogram, suggesting higher dislocation annihilation during the initial stages of contact fatigue. Results revealed a significant variation in MBN signals influenced by operational loads during tests, with a noteworthy increase observed just before gear failure. Using a scale from 0% (manufactured condition) to 100% (failure), the study successfully detected failures at 17% of the gear's lifespan, providing valuable insights for early failure detection in industrial applications.
See more of: Residual Stress I
See more of: Residual Stress