NTRS - NASA Technical Reports Server

Back to Results
mechanisms of fatigue crack retardation following single tensile overloads in powder metallurgy aluminum alloysIn ingot metallurgy (IM) alloys, the number of delay cycles following a single tensile overload typically increases from a minimum at an intermediate baseline stress intensity range, Delta-K(B), with decreasing Delta-K(B) approaching threshold and increasing Delta-K(B) approaching unstable fracture to produce a characteristic 'U' shaped curve. Two models have been proposed to explain this behavior. One model is based on the interaction between roughness and plasticity-induced closure, while the other model only utilizes plasticity-induced closure. This article examines these models, using experimental results from constant amplitude and single overload fatigue tests performed on two powder metallurgy (PM) aluminum alloys, AL-905XL and AA 8009. The results indicate that the 'U'-shaped curve is primarily due to plasticity-induced closure, and that the plasticity-induced retardation effect is through-thickness in nature, occurring in both the surface and interior regions. However, the retardation effect is greater at the surface, because the increase in plastic strain at the crack tip and overload plastic zone size are larger in the plane-stress surface regions than in the plane-strain interior regions. These results are not entirely consistent with either of the proposed models.
Document ID
Document Type
Reprint (Version printed in journal)
Bray, G. H.
(Virginia Univ. Charlottesville, United States)
Reynolds, A. P.
(Analytical Services and Materials, Inc.; NASA, Langley Research Center Hampton, VA, United States)
Starke, E. A., Jr.
(Virginia Univ. Charlottesville, United States)
Date Acquired
August 16, 2013
Publication Date
November 1, 1992
Publication Information
Publication: Metallurgical Transactions A - Physical Metallurgy and Materials Science
Volume: 23A
Issue: 11
ISSN: 0360-2133
Subject Category
Funding Number(s)
CONTRACT_GRANT: N00014-91-J-1285
Distribution Limits