球栅阵列
材料科学
有限元法
模具(集成电路)
倒装芯片
复合材料
热膨胀
弯曲
曲率
结构工程
几何学
数学
工程类
焊接
胶粘剂
图层(电子)
纳米技术
作者
Abdellah Salahouelhadj,Mario González,Kris Vanstreels,G. Van der Plas,G. Beyer,Eric Beyne
标识
DOI:10.1016/j.mee.2023.111947
摘要
The objective of this study is to investigate warpage of a Flip-Chip Ball Grid Array (FCBGA) package experimentally and numerically. Projection Moiré technique was used to measure full-field warpage during heating experiment from room temperature up to 250 °C. As temperature changes within the FCBGA package, different thermal strains are generated in the package resulting from the mismatch in the Coefficient of Thermal Expansion (CTE) between the constituent material layers. A finite element model and analytical equations based on the classical bending theory were used to analyze the thermally induced deformations of package under test. The results show that a maximum warpage value of 67 μm with a concave or smiling shape occurred at the neighborhood of 70 °C during heating and 87 μm warpage with a convex crying shape at 250 °C by assuming a stress-free temperature at 175 °C. Warpage analysis showed that the studied FCBGA package exhibits a different curvature in the die region at the center of the package compared to the region outside the die. At temperatures higher than 175 °C, a concave or smiling warpage shape was shown in the die region while a convex or crying shape in the region outside the die. In contrast, at room temperature, the die area has a convex shape warpage, while outside this area presented a concave shape. Furthermore, the good agreement between warpage of the studied package from finite element simulation and measurements was achieved thanks to the use of the measured material's thermal-mechanical properties of the substrate and Molding Compound (MC) rather than using datasheet data. In-plane CTE and Young's modulus of the substrate were measured using stereo-digital image correlation. Moreover, temperature dependent Young's modulus of the MC was measured by nanoindentation.
科研通智能强力驱动
Strongly Powered by AbleSci AI