A mechanistic model of damage evolution in lead free solder joints under combinations of vibration and thermal cycling with varying amplitudes

Abstract

A broad range of electronics applications involve the long-term exposure to combinations of thermal excursions and vibration with constantly varying amplitudes, and the ultimate life is very often limited by fatigue of the solder joints. The assessment of this is usually based on a combination of simplified accelerated tests and the assumption of simple acceleration factor expressions or models. Neither of the latter can, however, even account for observed effects of accelerated test parameters. To make matters worse common constitutive relations may be strongly misleading. Also, current damage accumulation rules cannot account for major interactions between thermal cycling and vibration or even just for ongoing variations in vibration amplitudes or thermal cycling parameters, respectively. Even if we are ‘just’ aiming to decide between alternative material, design, or process parameters, or to compare a new product to a previous (supposedly similar) one, we should be concerned with the relative performance in service and not just in accelerated tests.The main objective of the present work is to combine two recent mechanistic models into a more comprehensive one that includes interactions between effects of thermal cycling and vibration with varying amplitudes. Important parameters in the resulting model still remain to be determined but a detailed understanding of the underlying mechanisms and the consequences of their interactions for realistic Tin-Silver-Copper (SnAgCu) joints allows for the prevention of surprises, the definition of appropriate test protocols, and the proper interpretation of results. This includes the identification of ‘worst case’ real life scenarios to test for, as well as justification for the use of shear testing in thermal cycling when event detection is not an option. Extension of the model to new alloys is briefly discussed as well.