MECHANICAL TESTING OF SOLDER JOINT ARRAYS VERSUS BULK SOLDER SPECIMENS
Author: Robert Darveaux Company: Arizona State University and Amkor Technology, Inc Date Published: 9/24/2006
Abstract: Mechanical testing of bulk solder specimens was compared to solder joint arrays for the case of lead free alloys. The bulk specimen data was taken from the literature. Solder joint array data was 1.5X to 2.5X higher strength than bulk specimen data. Both bulk reflowed and solder joint reflowed samples showed a 20% decrease in strength with aging. Quenched bulk solder specimens had the largest effect with a 40% reduction. Finite element analysis was used to show the stress and strain distribution in solder joint array samples under various loading conditions. Misalignment of the sample results in a significant impact to the loaddisplacement response. For a well aligned sample, the load distribution from joint-to-joint becomes very uniform once inelastic deformation starts to occur. The stresses within a given joint are not uniform, however, the practice of using the pad area for shear stress calculation does result in adequate steady state constitutive properties. For thermal cycle simulations, using bulk specimen properties will generally result in higher predicted strain range and lower stress range than that obtained using solder joint array properties. Calculated acceleration factors between different thermal cycle conditions strongly depend on the method used to obtain the solder alloy properties.
Key words: Mechanical Testing, Lead Free Solder, Creep, Finite Element Analysis, Fatigue Life Prediction