Fracture of glass seals in metallic hermetic electronic packaging is a significant failure mode because it may lead to moisture ingress and also to loss of load carrying capacity of the glass seal. Seal glasses are intrinsically brittle and their fracture is governed by the stresses generated. This study investigates stresses in lead seals caused by handling, testing, mechanical vibration, and thermal excursions. Loads considered are axial tension, bending, and twisting of the lead. More general loading can be handled by superposition of these results. Factorial techniques, commonly used in multi-variable Design of Experiments (DoE), are used in conjunction with finite element parametric simulations, to formulate closed-form regression models which relate the maximum principal stress within the glass seal to the type of loading and geometry. The accuracy of the proposed closed-form equations are verified through analysis of residuals. The analysis reveals the sensitivity of the magnitude of the seal stress to design variables such as the materials and geometry of the seal, lead, and package. Manufacturing-induced problems such as defects and flaws are not considered. An additional purpose for presenting this study is to illustrate the use of design of experiment methods for developing closed-form models and design guidelines from simulation studies, in a multi-variable problem.

This content is only available via PDF.
You do not currently have access to this content.