Location, location & size: defects close to surfaces dominate fatigue crack initiation

Abstract

Metallic cast components inevitably contain defects such as shrinkage cavities which are inherent to the solidification process. Those defects are known to significantly alter the fatigue life of components. Yet very little is known, quantitatively, on the dangerosity of internal casting defects compared to surface ones. In this study, fatigue specimens containing controlled internal defects (shrinkage pores) are used to foster internal cracking. In situ fatigue tests monitored by X ray synchrotron tomography revealed that the internal nucleation and propagation of cracks was systematically overran by surface cracking initiated at castings defects up to ten times smaller than the internal ones. These findings indicate that the presence of internal defects in cast components can be tolerated to a larger extent than is allowed by nowadays standards

In situ synchrotron ultrasonic fatigue testing device for 3D characterisation of internal crack initiation and growth

“…However, this manufacturing process produces casting defects (such as pores, oxides, or shrinkages) inducing local stress concentrations and consequently a lower fatigue resistance compared with wrought alloys. But only very few works either in low cycle fatigue (LCF) 9,10 or high cycle fatigue (HCF) 11,12 have been published on internal crack initiation and propagation monitoring with a nondestructive technique. Numerous studies have used X-ray microtomography for studying crack propagation mecha-Nomenclature: A, area of the defect projected on the plane perpendicular to the loading axis; HCF, high cycle fatigue; LCF, low cycle fatigue; N, number of cycles; N i , number of cycles in the ith loading sequence; N Tot , total number of cycles; U 1 , amplitude of the fundamental frequency of the displacement of the specimen extremity; U 2 , amplitude of second harmonic of the displacement of the specimen extremity; VHCF, very high cycle fatigue; β, non-linearity damage parameter; ΔT, temperature variation nisms 6,7 or damage development 8 .…”

“…Surface cracks prevail over internal ones in LCF and HCF regimes, that is the main reason why only restricted observations of internal crack propagation 9,11,12 have been done so far but this is different in very high cycle fatigue (VHCF) regime where internal cracks dominate. Surface cracks prevail over internal ones in LCF and HCF regimes, that is the main reason why only restricted observations of internal crack propagation 9,11,12 have been done so far but this is different in very high cycle fatigue (VHCF) regime where internal cracks dominate.…”

“…Numerous studies have used X-ray microtomography for studying crack propagation mecha-Nomenclature: A, area of the defect projected on the plane perpendicular to the loading axis; HCF, high cycle fatigue; LCF, low cycle fatigue; N, number of cycles; N i , number of cycles in the ith loading sequence; N Tot , total number of cycles; U 1 , amplitude of the fundamental frequency of the displacement of the specimen extremity; U 2 , amplitude of second harmonic of the displacement of the specimen extremity; VHCF, very high cycle fatigue; β, non-linearity damage parameter; ΔT, temperature variation nisms 6,7 or damage development 8 . But only very few works either in low cycle fatigue (LCF) 9,10 or high cycle fatigue (HCF) 11,12 have been published on internal crack initiation and propagation monitoring with a nondestructive technique.…”

“…These studies show that (i) both internal and external cracks can initiate in the same specimen and (ii) surface crack growth rates are higher than internal ones. Surface cracks prevail over internal ones in LCF and HCF regimes, that is the main reason why only restricted observations of internal crack propagation 9,11,12 have been done so far but this is different in very high cycle fatigue (VHCF) regime where internal cracks dominate. Furthermore, there are only very limited experimental data of internal crack growth rate available at the moment.…”

“…Furthermore, there are only very limited experimental data of internal crack growth rate available at the moment. 9,11,12 Several authors [13][14][15] have already underlined that internal failure can be promoted by low stress loadings leading to a very large number of cycles to failure. Such tests, in the VHCF regime, can be performed in a realistic testing time using ultrasonic testing machine.…”