Methodology for Automated Design of Microelectronic Packages

Abstract

A general design methodology for the optimal design of electronic packages is presented. The design problem is cast as a nonlinear programming (NLP) problem. Design for critical packaging issues such as thermal and mechanical performance, is automated by combining thermal and mechanical finite element analysis with an NLP optimizer. The major issues such as design problem formulation, model generation or preprocessing, nonlinear finite element analysis, design variable identification, design sensitivity analysis and software implementation are addressed. A case study involving the design of an RF packaging component in communication products illustrates the design methodology implemented in the Automated Design of Electronic Packaging (ADEP) software system.