PSI - Issue 7

A. Brueckner-Foit et al. / Procedia Structural Integrity 7 (2017) 36–43 A. Brückner-Foit / Structural Integrity Procedia 00 (2017) 000–000

38

3

100,000 cycles prior to failure, after which the post-mortem specimens were analyzed in order to identify the fracture behavior.

Fig. 1 Typical microstructure of the studied alloy (A): 1) eutectic silicon, 2) ß-Al 5 FeSi, 3) Al 2 Cu.

Fig. 2 Volume element with gas pores surrounding the shrinkage pores.

The three-dimensional measurements of the casting defects was performed using a high resolution lab computer tomograph CT (Zeiss Xradia Versa 520) at 70 KV and a power of 7 W with the 0.4X objective, achieving a resolution of 3.84 μ m/voxel for the pores segmentation and 2.89 μ m/voxel for Fe-rich inclusions. The image analysis was carried out with the rendering software Avizo [FEI], in which the casting pores were segmented with the Interactive Thresholding tool after applying the Non-local means filter, whereas the Fe-rich inclusions were segmented manually. 3. Results 3.1 The role of gas and shrinkage pores Fig. 3 shows the results of the fatigue tests performed with the previously described specimens. The large scatter in the fatigue life in each stress amplitude indicates the strong influence of the casting defects on fatigue as well as the high variability of the defectology associated to this casting process. An exhaustive analysis of the fracture surfaces both by scanning electron microscopy and micro-computed tomography revealed that there are distinct failure behaviors related to different fatal defects.