ASSESSMENT OF NEED FOR SOLDER IN MODELING POTTED ELECTRONICS DURING GUN-SHOT

摘要

Smart projectiles use electronic components such as circuit boards with integrated circuits to control guidance and fuzing operations. During gun-launch, the electronics are subjected to 3-dimensional g-forces as high as 15000-gs. The US Army uses finite element analysis to simulate electronics with high-g, dynamic loads. Electronics are difficult to model due to the large variation in size, from large circuit boards, to very small solder joints and solder pads. This means that to accurately model such small features would require very large models that are computationally expensive to analyze, often beyond the capability of resources available. Therefore small features such as solder joints are often not included in the finite element models to make the models computationally tractable. The question is: what is the effect on model accuracy without these small features in the model. The purpose of this paper is to evaluate the effect solder joints and solder pads have on the accuracy of structural analysis of electronic components mounted on circuit boards during gun shot. Finite element models of simplified circuit boards, chips and potting were created to do the evaluation. Modal analysis and dynamic structural analysis using typical gun loads were done. Both potting at high temperature (soft) and potting at low temperature (stiff) were used in the dynamic analysis. In the modal analysis there was no potting. All of these models were run with and without solder. Comparisons were done using the general purpose finite element package ABAQUS [4], dynamic analyses were nonlinear. A cross section of a finite element model is shown in Fig. 1. This figure shows the outer can, potting material, chips, and circuit board. Solder is faintly shown between the chips and the circuit board.