Meeting thermal performance and reliability challenges for a thermally enhanced ball grid array package (TEBGA)

Abstract

For devices with challenging power management requirement, thermally enhanced ball grid array package (TEBGA) offers a good solution, where the device is attached to a heat spreader, usually made of copper, with a thermally conductive epoxy to ensure a good conductive path for heat to escape from the die. The top die surface and bonding wires are covered with an overmolding compound for environmental protection such that heat dissipation is typically limited in that direction. However, TEBGA is not without its unique challenges. In this paper, we present a study on the challenges of meeting the thermal performance and reliability requirements for a ASIC packaged with TEBGA. A localized deformation or ldquodimplerdquo of the TEBGA package is discovered during the package assembly process, where the heat-spreader is noted to have deformed under the die shadow, which results in a circular shaped indentation. This raises concerns about the impact on the thermal performance of the subsequent package to heat sink interface when it is integrated into the system. Solution to this potential problem rests on balancing thermal performance, reducing package stress level & understanding potential long term package reliability. Deformation of the package with each process step will be first described and particular attention will be given to the change of package profile after the die attach process; then a finite element analysis of the stress and deformation of the die attach process is discussed and important parameters affecting the deformation and stress are shown; moreover, a thermal resistance model assessing the thermal budget for this package in a system environment is reviewed and confirmation with numerical analysis & validation by experimental analysis are highlighted; furthermore, an interactive analysis is subsequently performed based on the FEA model for package stress/deformation and thermal resistance model to optimize the packaging solution; finally, balanced solution through this interactive optimization process is summarized and demonstrated in the manufacturing process.