Multistage Fatigue Modeling of Cast A356-T6 and A380-F Aluminum Alloys

From EVOCD
Jump to: navigation, search

Contents

Abstract

This article presents a microstructure-based multistage fatigue (MSF) model extended from the model developed by McDowell et al.[1,2] [1] to an A380-F aluminum alloy to consider microstructure-property relations of descending order, signifying deleterious effects of defects/discontinuities: (1) pores or oxides greater than 100 lm, (2) pores or oxides greater than 50 lm near the free surface, (3) a high porosity region with an area greater than 200 lm, and (4) oxide film of an area greater than 10,000 lm 2 These microconstituents, inclusions, or discontinuities . represent different casting features that may dominate fatigue life at stages of fatigue damage evolutions. The incubation life is estimated using a modified Coffin–Mansion law at the microscale based on the microplasticity at the discontinuity. The microstructurally small crack (MSC) and physically small crack (PSC) growth was modeled using the crack tip displacement as the driving force, which is affected by the porosity and dendrite cell size (DCS). When the fatigue damage evolves to several DCSs, cracks behave as long cracks with growth subject to the effective stress intensity factor in linear elastic fracture mechanics. Based on an understanding of the microstructures of A380-F and A356-T6 aluminum alloys, an engineering treatment of the MSF model was introduced for A380-F aluminum alloys by tailoring a few model parameters based on the mechanical properties of the alloy. The MSF model is used to predict the upper and lower bounds of the experimental fatigue strain life and stress life of the two cast aluminum alloys.

Introduction

Figure 1 Completely reversed strain-life curves for A356-T6 and A380-F Al alloys

CAST aluminum alloys of various compositions and casting conditions have been developed as major structural materials for automobiles and industrial equipment. These structural components are subjected usually to cyclic loading in service. The components, in general, are designed in such a way that the stress remains in the elastic regime, except perhaps for an occasional overload. However, the stresses around microstructural discontinuities, such as hard and brittle Si- or Fe-bearing particles, oxides, casting pores, and dendrite features, may exceed the elastic limit, even if the macroscopic response is elastic. Due to the complex microstructure of cast aluminum alloys, the fatigue life exhibits two to four orders of variation in the high-cycle fatigue (HCF) regime, in which the fatigue life is sensitive to microstructures. Therefore, a practical fatigue life prediction model must incorporate the effects of the microstructure features on fatigue life to capture the scatter due to stochastic discontinuities in the component. A multistage fatigue (MSF) model that incorporates microstructural effects on fatigue damage incubation and growth was developed for a cast A356-T6 aluminum alloy by McDowell et al., [2] based on micromechanics simulations [3,4] and small-scale experiments. [5–9] The MSF model and concepts of fatigue growth stages were extended to a 7075-T651 aluminum alloy for airplane frame applications [10,11] and lightweight AM50 and AE44 magnesium alloys. [12,13] A few MSF model parameters specifically developed for A356 aluminum alloy are represented by the static mechanical properties indicated by Xue et al. [10,12] In this article, we extend the MSF model developed by McDowell et al. [2] for a cast A356-T6 aluminum alloy to a die-cast A380-F aluminum alloy to obtain HCF life predictions. The two alloys have different amounts of primary alloy element Si (7 wt pct and above 9 wt pct) and a much larger difference in the secondary alloy elements of copper and iron. However,the fatigue life limiting features for both alloys are large casting pores especially for HCF loading cases. Only a few MSF model parameters for A356-T6 were modified based on the macroscopic mechanical properties and common fatigue model constants of the A380 aluminum alloy. The MSF model predictions correlate very well with the experimental results. The results of this article show that the MSF model can be used for other aluminum alloys, and as such can be used as a microstructure-sensitive engineering model to support material process design and component-specific tailoring of materials.

Results

Figure 3 Comparison of experimental data and calibrated MSF models for completely reversed strain-life curves of an A356-T6 Al alloy under uniaxial strain control (noted as A356-T.EXP) and of A380-F under stress-controlled.
Figure 4 The MSF model prediction for the breakdown of regimes of incubation, microstructurally and PSC growth, and long crack growth for A356-T6, assuming the fatigue damage forms at a 120-lm-diameter casting pore.
Figure 5 The MSF model predictions for aluminum A356-T6 that give the upper and lower bounds of the fatigue life in accordance with experimental data, assuming the cracks form at casting pores of diameter 20 lm (upper bound) and 1200 lm (lower bound), respectively.

The MSF models for A356-T6 and A380-F aluminum alloys are shown in Figure 3 assuming that the fatigue damage formed at a pore of 120 lm in diameter. The MSF model for A380-F correlates very well with the experimental results. This successful application of the MSF model indicates that the fatigue crack growth, both MSC/PSC and long crack growth, may be similar.

To explore the partitioning of the total life into incubation, small and long crack growth, a breakdown of the lives corresponding to MSF model predictions for incubation, microstructural small crack growth, and long crack growth are shown in Figure 4 for A356-T6. It is clearly demonstrated that incubation takes about 80 or 90 pct of the total life when the fatigue life is greater than 105.

Therefore, by only adjusting the incubation parameters, a fairly good MSF model correlation was obtained for the A380 aluminum alloy. To understand the large scatter of fatigue life in the strain-life tests, we simulated various sizes of the incubation pores for A356-T6 using the MSF model. A pore having a size of 1200 lm gave a lower bound while a pore having a size of 20 lm gave the upper bound of the fatigue life in the strain-life tests shown in Figure 5.

References

  1. D.L. McDowell, J. Fan, and M.F. Horstemeyer: AFS Trans., 1999, vol. 109, pp. 703–12


Citation: Multistage Fatigue Modeling of Cast A356-T6 and A380-F Aluminum Alloys Y. Xue, C. L. Burton, M. F. Horstemeyer, D. L. McDowell and J. T. Berry

Personal tools
Namespaces

Variants
Actions
home
Materials
Material Models
Design
Resources
Projects
Education
Toolbox