The reliability of component boards studied with different shock impact repetition frequencies

Abstract

As a result of the ever-shortening time to market of new electronic products there is a clear need to make the currently employed reliability assessment procedures more efficient. However, while trying to achieve these cost savings it is important to pay attention to how much reliability tests can be accelerated without producing misleading lifetime statistics or irrelevant failure modes and mechanisms. Therefore, the primary objective of this work was to clarify the effects of impact repetition frequency on the lifetime and failure mechanisms of electronic component boards. The reliability assessments were conducted with four package types (BGA288, BGA144, QFN72, and μSMD36) with the intention of including different package dimensions and interconnection shapes and sizes in the scope of evaluation. Three shock impact repetition frequencies (0.01Hz, 0.1Hz, and 1.6Hz) were used to study the sensitivity of their lifetime to the time between the shock impacts. The results showed that the impact repetition frequency has a significant effect on the average lifetime of the studied packages: the average number of impacts-to-failure increased with increased impact repetition frequency. A change in the impact repetition frequency did not, however, change the primary failure mode and the failure mechanism. In order to explain this observation, the formation of residual stresses in the solder interconnections from shock impacts and their relaxation during the time between the shock impacts were evaluated by employing FEM, which showed that at room temperature the stress relaxation by creep of solder is significant even at less than 10s time frame. The effect of the stress relaxation on the lifetime of the component boards was studied experimentally by repeating the impact tests at elevated temperature with the intention of influencing rate of stress relaxation. The results showed that the relaxation of the residual stresses has a significant effect on the component boards shock reliability (at less than 1% risk level). The results presented in this paper indicate that the impact repetition frequency should be taken into account, especially when reliability assessments are being carried out with different testers, since the interval between the shock impacts varies from one test system to another.