Analysis of Thin Flip Chip Chip-Scale Package Warpage Causes and Variations

The objective of this paper is to present results of an investigation on variations in materials properties, package designs, and assembly processes affecting final warpage measurement of thin flip chip chip-scale package (FCCSP). Past shadow-moire measurements have revealed unexplained warpage measurements shifts and variations for the same package assembled in different lots. Variations in warpage measurements can be attributed to package unit-to-unit differences within a lot, as well as differences between assembly lots. Variability due to within-lot differences is a consequence of intrinsic variances in mold compound properties, substrate dimensions, assembly processes, and shadow-moire metrology. To support repeatable productions, variability between-lots should ideally be non-existent or at its minimum when all material properties are held constant. Package warpage can be predicted with high accuracy once all within-lot variability due to material and process variances have been characterized and included. However, variability due to between-lot differences is more intractable and unpredictable making it difficult to characterize and validate the models. To illustrate the uncertainties in simulated warpage prediction and experimental warpage results arising from between-lot variabilities, finite element analysis (FEA) simulation of six different FCCSP test vehicles are compared to experimental warpage data. It is found that within-lot warpage ranges for all the package options evaluated are consistent with past data but warpage ranges between-lots are not reproducible. It is determined that apart from the well documented sources of variations leading to package warpage variations, there are still other factors to be unraveled that will explain warpage measurement shifts between lots. This paper provides findings completed to-date and concludes with recommendations on next steps of this investigation. Readers are encouraged to further pursue their own investigations based on these findings.