Finite element prediction of high cycle fatigue life of aluminum alloys
- PDF / 913,851 Bytes
- 9 Pages / 594 x 774 pts Page_size
- 74 Downloads / 213 Views
I.
INTRODUCTION
THE fatigue design of automotive components
includes a material specification in the form of either a stress-life (S-N) curve or the closely related strain-life curve, i.e., the relationship between the applied cyclic stress, or cyclic strain, and the number of cycles required to cause failure. In either form, these fatigue life curves are obtained from a large number of tests on smooth specimens. Over the years, these curves have been measured for a wide range of alloys, but whenever a new material is introduced, its fatigue properties must be characterized by the generation of its fatigue life curve. This is a timeconsuming procedure, particularly when the primary interest is in low cyclic stresses and long fatigue lives extending to l0 s or e v e n 109 cycles, a regime known as the fatigue endurance limit, or fatigue strength, of a material. In fact, such high cycle fatigue data do not abound in the literature because of the extensive testing involved. So, not surprisingly, attempts have been made to estimate the high cycle region of the S-N curve and, particularly, the fatigue strength of a metal. The ideal is to base the estimate upon tensile mechanical properties, since the latter can be obtained from a single quick test. Complex empirical relationships tll have been proposed, but the only useful rule of thumb is that the fatigue strength is in the range of 0,3 to 0.5 of the tensile strength, the ratio depending upon the alloy. Some attempts have also been made to express the fatigue life curves mathematically. The most notable of these are the empirical laws of Basquin, t21 Coffin, t31 and Manson t41 which are described in the next section. However, there is still no viable alternative to the tedious experimental determination of a fatigue life curve. A major obstacle to calculating the form of a fatigue life curve, particularly in the high cycle (low stress) regime, has been the incomplete understanding of the basic
WILLIAM J. BAXTER, Senior Staff Research Scientist, and PEI-CHUNG WANG, Staff Research Scientist, are with General Motors Research Laboratories, Warren, MI 48090-9055. Manuscript submitted August 17, 1989. METALLURGICAL TRANSACTIONS A
mechanisms involved and the inability to specify the ratecontrolling process. This is not due to a lack of effort, as testified by the extensive investigations which have identified the important stages of metal fatigue as the sequential formation of (1) persistent slipbands (PSB's), (2) microcracks (recently rechristened as "short ~ cracks) either in the PSB (stage I cracks) or in adjacent grain boundaries, (3) microcrack coalescence to form a "long" crack, and (4) crack growth (stage II) to final fracture. But which of these processes is dominant in determining the fatigue life? It has long been generally accepted that at high stress levels, fatigue life is determined primarily by crack growth, while at low stress levels, most of the life is consumed by the process of crack initiation. However, the distinction between these two regimes depend
Data Loading...
