Novel board level drop test simulation using implicit transient analysis with input-G method

• Main Authors: Jing-en Luan, Tong Yan Tee
• Format: Conference Proceeding
• Language: English
• Subjects: Analytical models; Finite element methods; Integrated circuit modeling; Integrated circuit packaging; Load modeling; Numerical models; Strain measurement; Testing; Transient analysis; Vibration measurement
• DOI: 10.1109/EPTC.2004.1396693

In recent several years, there are many efforts to study the reliability performance of IC package under drop impact loading by numerical modeling. Most of the models utilize LS-DYNA or ABAQUS types of explicit solvers. However, there is a lack of efficient implicit model for the organization without access to explicit solvers. This paper presents a novel non-linear transient dynamics model with implicit algorithm for board level drop test modeling using conventional finite element modeling framework such as ANSYS implicit solver. According to the JEDEC board level drop test standard, the impact pulse is compulsorily monitored during drop test qualification of components. This impact pulse includes the effects of velocity before and after impact, and the effects of contact surface and material, as well as drop tester's characteristics such as friction of guiding rods, which are too complicated to be simulated by conventional free-fall model. A lumped model with one degree of freedom mass-spring-damping system is proposed to describe the dynamic behaviors of PCB assembly under drop impact. Based on this model, the Input-G method using implicit transient dynamic analysis is derived and proven both theoretically and numerically. This method is more accurate and much faster, and bypasses many technical difficulties in conventional free-fall drop model such as adjusting the parameters of contact surfaces, defining contact type, and etc. The model established has good correlation with experimental measurement of dynamic strain, PCB center displacement, and vibration frequency as observed in actual drop test. This implicit method is comparable with explicit method developed in previous study in terms of solution time and accuracy. Its solution time is about 30% longer than explicit Input-G method, but much shorter than explicit free-fall model and implicit free-fall model. It is a good cost-saving alternative for organizations without access to explicit solver to perform drop test simulation.