Computational & Technology Resources
an online resource for computational,
engineering & technology publications
PROCEEDINGS OF THE FOURTEENTH INTERNATIONAL CONFERENCE ON CIVIL, STRUCTURAL AND ENVIRONMENTAL ENGINEERING COMPUTING
The Applicability of Multiaxial Fatigue Approaches using Spectral Analysis Only
L. Ben Fekih, G. Kouroussis and O. Verlinden
University of Mons, Belgium
L. Ben Fekih, G. Kouroussis, O. Verlinden, "The Applicability of Multiaxial Fatigue Approaches using Spectral Analysis Only", in , (Editors), "Proceedings of the Fourteenth International Conference on Civil, Structural and Environmental Engineering Computing", Civil-Comp Press, Stirlingshire, UK, Paper 208, 2013. doi:10.4203/ccp.102.208
Keywords: multiaxial fatigue, random vibrations, tin-lead solder, spectral data.
The aim of the research described in this paper is to assess rapid and trustworthy fatigue prediction methodologies intended for aerospace electronic boards design. The motivation for this study comes from the feedback of former launch missions revealing the vulnerability of tin-lead solder interconnects subject to random vibrations. A case study consists of an electronic chip soldered to a support board using a ball grid array. The assembly is intended to undergo an out-of-plane random acceleration field. A selection among many prospective multiaxial high cycle fatigue methodologies has been made. Since all fatigue life estimations require the transformation of the multiaxial stress state to an equivalent stress scalar in order to extract cycles and therefore accumulate fatigue damage, a search of effective stress predictors was undertaken. By use of the spectral stress responses at a critical point of the solder-board interface, fatigue models having the most suitable estimators have been retained. Furthermore, the original frequency-domain formulations have been brought to some time-domain fatigue approaches. This allows investigation of the simulation stress data without recourse to sampling thus contributing, in some cases, to alleviate the computational cost.
purchase the full-text of this paper (price £20)