Abstract

Very often need arises for evaluation of solder systems based not just on their reflow temperatures but on their capacity to withstand damage. Is there an inherent advantage from fatigue considerations in choosing one system over another? If so can we predict this difference.

Damage and it’s accumulation resulting in fatigue failure of solder interconnects is difficult to predict because of complex time and temperature dependence of solder behavior. Strain based solder fatigue description, like Coffin Manson, is not adequate as the fatigue life of the solder may also be a function of stresses. To circumvent the inherent drawback of the strain-based approach and for a more accurate prediction of stresses and inelastic strains, a dissipated energy based damage index approach, using a finite strain, nonlinear Finite Element Method (FEM), has been developed by the author for damage and fatigue life prediction. Instead of looking at complex tensorial responses of stress and strain it is proposed to look at the response in terms of the scalar damage.

The damage index method based approach is used here for comparing two different solder systems. The solder systems chosen are a high lead Pb97Sn3 system and the Sn60Pb40 eutectic alloy. In order to avoid biasing the evaluation for the melting point of the solder alloy, a new equivalent temperature cycle method is proposed. The response of the solder system to this equivalent temperature cycle is proposed as a fair methodology to distinguish the inherent differences in solder systems based on their propensity for damage.

This content is only available via PDF.
You do not currently have access to this content.