Impact of boundary conditions, glass fiber orientation and stress-free temperature on the performance of a plastic-enclosed electronic product in a thermal cycling environment
التفاصيل البيبلوغرافية
العنوان:
Impact of boundary conditions, glass fiber orientation and stress-free temperature on the performance of a plastic-enclosed electronic product in a thermal cycling environment
Automotive electronic products generally consist of a glass filled plastic housing, aluminum cover, and a laminated PCB with several components such as chip capacitors, resistors and leaded chips mounted on it. Typically all electronic products are subjected to validation testing to evaluate their reliability under vibration, shock and thermal cycling environment. Reliability of such electronic products/components poses a huge challenge for design engineers, particularly under severe automotive thermal cycling profiles. Typical failures under thermal cycling such as component lead breakage, wire bond breakage at the heel near bond pad and shearing of the solder joint result in discontinuity of electrical connections and thereby fail to qualify the product level functionality test. Finite Element Analysis (FEA) is a powerful computational technique to evaluate the physical behavior of a product in a given environment. However, boundary conditions or constraints in a system level model significantly affect the behavior of the product in thermal cycling analysis. A poor choice of boundary conditions can sometimes fail to capture the real physics of the problem, thereby resulting in incorrect conclusions. In this paper, effects of system level boundary conditions on an automotive electronics package are studied for a thermal cycling profile. A coefficient of thermal expansion (CTE) mismatch analysis is performed to simulate the given thermal cycling profile. The study is focused on the reliability of gold wire bond on an integrated chip (Ie) making electrical connection to an adjacent substrate. FE analysis results indicate that boundary conditions on the system-level model significantly affect wire bond stresses noticed at the heel of the bond and these in turn, affect the number of survival thermal cycles. Hence it is extremely critical to use appropriate constraints on the product that will result in the same physical behavior as seen on the test sample inside the thermal chamber. Results from selected FE analyses are correlated with laboratory thermal durability tests on wire bond. This paper also addresses how fiber orientations in glass-filled plastic and the appropriate choice of stress-free temperature are critical to end result. We caution the FE analyst from using temperature dependent coefficient of thermal expansion (CTE) data without transformation to a stress-free condition in a cooling cycle of thermal mismatch analysis.
