TITLE:
Study on Fatigue Lifetimes and Their Variation of Mg Alloy AZ61 at Various Stress Ratios
AUTHORS:
Kenichi Masuda, Sotomi Ishihara, Minoru Ishiguro, Hiroshi Shibata
KEYWORDS:
Fatigue, Stress-Fatigue Life (S-N) Curve, Probabilistic Stress-Fatigue Life (P-S-N) Curve, Crack Growth, Extruded Magnesium Alloy, Intermetallic Compound, Extreme-Value Statistics
JOURNAL NAME:
Materials Sciences and Applications,
Vol.9 No.13,
December
3,
2018
ABSTRACT: In this
study, fatigue tests under different R ratios were conducted on the AZ61 Mg alloy to investigate its fatigue lifetimes
and fatigue crack growth (FCG) behavior. The fracture surface of the failed
specimens was investigated using a scanning electron microscope to study the
size of the intermetallic compounds from which the pioneer fatigue crack
initiated and led to the final failure of the specimen. To determine the
maximum size of the intermetallic compounds existing within the cross section
of the specimen at higher risk, Gumbel’s extreme-value statistics were
utilized. In the present study, the intermetallic compounds contained within
the specimen were assumed to be the initial cracks existing in the material
before the fatigue tests. A modified linear elastic fracture-mechanics
parameter, M, proposed by McEvily et al., was used to analyze the short
FCG behavior under different stress ratios, R.
The relation between the rate of FCG and M parameter was found to be useful and appropriate for predicting the fatigue
lifetimes under different R ratios.
Moreover, the probabilistic stress-fatigue life (P-S-N)
curve of the material under different R ratios could be predicted with this method, which utilizes both the FCG law and
a statistical distribution of sizes of the most dangerous intermetallic
compounds. The evaluated results were in good agreement with the experimental
ones. This correspondence indicates that the estimation method proposed in the
present study is effective for evaluation of the probabilistic stress-fatigue
life (P-S-N) curve of the material under different R ratios.