Fatigue of casting flaws at a notch root under an SAE service load history

Citation
Aa. Dabayeh et Th. Topper, Fatigue of casting flaws at a notch root under an SAE service load history, FATIG FRACT, 23(12), 2000, pp. 993-1006
Citations number
19
Categorie Soggetti
Material Science & Engineering
Journal title
FATIGUE & FRACTURE OF ENGINEERING MATERIALS & STRUCTURES
ISSN journal
8756758X → ACNP
Volume
23
Issue
12
Year of publication
2000
Pages
993 - 1006
Database
ISI
SICI code
8756-758X(200012)23:12<993:FOCFAA>2.0.ZU;2-4
Abstract
Smooth and notched specimens of a 319 cast aluminium alloy were fatigue tes ted under a Society of Automotive Engineers service load history in the as- cast and hipped conditions. The hipping process, which includes subjecting the cast material to a high pressure at high temperature and then slowly co oling down to eliminate internal flaws, decreased the flaw size and improve d the fatigue life of cast Al 319 smooth specimens. A 0.6-mm-diameter hole was drilled at the notch root of notched specimens to simulate a natural fl aw at the notch root. Specimens with two different notch sizes were tested. Circular edge notches reduced the fatigue strength and a 0.6-mm-diameter d rilled hole at the notch root resulted in a further reduction. The fatigue lives of smooth specimens, notched specimens and notched specim ens with a flaw at the notch root subjected to the service load history wer e predicted using the strain-life approach, an effective strain-life approa ch and a strain-based intensity factor crack growth model. In crack growth modelling of the fatigue life of smooth cast aluminium specimens the flaw w as modelled as a circular edge notch having the same diameter as the flaw. However, in the case of a flaw at a notch root the flaw was modelled as a t hree-dimensional cavity subjected to the notch stress field and the crack l ength was predicted in the longitudinal and transverse directions of the sp ecimen cross-section. The strain-life approach was unconservative for all s pecimen geometries studied. The effective strain-life approach gave good pr edictions for smooth and blunt notched specimens but gave very conservative predictions for the specimens with flaws in the notch roots. The crack gro wth calculations gave accurate predictions for all the specimen geometries.