Abstract

Abstract A newly developed methodology is used to support test validation of ball grid array (BGA) solder joint vibration fatigue life prediction model. This model is evolved from an empirical formula of universal slopes, which is derived from high-cycle fatigue test data using a curve fitting technique over 29 different materials of metals. To develop the BGA solder joint vibration fatigue life prediction model, a test vehicles (TV), on which various sizes of BGA daisy-chained packages are soldered, is first designed, fabricated and subjected to random vibration tests with continuously monitoring the solder joint integrity. Based on the measurement results, a destructive physical analysis is then conducted to further verify the failure locations and crack paths of the solder joints. Next, a method to determine the stresses/strains of BGA solder joints resulting from exposure of the TV to random vibration environments is developed. In this method, a 3-D modeling technique is used to simulate the vibration responses of the BGA packages. Linear static and dynamic finite element analyses with MSC/NASTRAN™ computer code, combined with a volume-weighted average technique, are conducted to calculate the effective strains of the solder joints. In the calculation process, several in-house developed Fortran programs, in conjunction with the outputs obtained from MSC/NASTRAN™ static and frequency response analyses, are used to perform the required computations. Finally, a vibration fatigue life model is established with two unknown parameters, which can be determined by correlating the derived solder effective strains to the test data. This test-calibrated model is then recommended to serve as an effective tool to determine the integrity of the BGA solder joints during vibration. Selecting more study cases with various package sizes, solder ball configurations, vibration profiles to further calibrate this model is also recommended. An example of a 313-pin plastic and 304-pin ceramic BGAs is illustrated in the present study.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call