Here is the abstract you requested from the IMAPS_2010 technical program page. This is the original abstract submitted by the author. Any changes to the technical content of the final manuscript published by IMAPS or the presentation that is given during the event is done by the author, not IMAPS.
|Validated High Speed Pull and Shear Test Methodologies to Evaluate Pb-Free BGA Mechanical Strength|
|Keywords: BGA, Mechanical Strength, High Speed Pull/Shear|
|The conversion to lead free Ball Grid Array (BGA) packages has raised several new assembly and reliability issues. Lead-free solder joints are generally stiffer than tin-lead solder joints, and mechanically induced failures have become more prevalent in lead-free solder assemblies. Traditionally, assembly level bend and shock testing is performed to evaluate mechanical assembly reliability. However, bend and shock tests are expensive, cumbersome and not feasible for evaluating lot-to-lot variations in mechanical strength. Consequently, there is a need for a validated, component level test method that can be used as an accurate indicator of assembly level mechanical strength. Several papers have been written to evaluate the efficacy of high speed pull and shear testing as a viable reliability indicator. However, so far, previously published data has not demonstrated clear trend correlation with assembly level bend and shock testing. In addition, the comparisons reported in the literature do not clearly articulate what specific machine parameters should be controlled to what accuracy to obtain clear trends. In this study, a comprehensive Gage R&R study is performed to evaluate the accuracy of the test equipment, including high speed video calibration. Then, test studies were performed to compare the accuracy of the results, spread across different package constructions, solder metallurgies, ball pitch and surface finish. In addition, the effect of parameters like multiple reflows and aging on specific metallurgies and surface finishes was studied. The results were generated over more than 2000 test runs. Finally, the study rank orders all critical test parameters and articulates what precise steps can be taken to generate relevant data for standard and custom devices during early evaluation stages and during high volume manufacturing.|
|Mudasir Ahmad, Staff Engineer, Interconnect Design and Reliability
Cisco Systems, Inc.
San Jose, CA