E. O. Pershin, A. S. Shalumov, D. B. Solovyov

Modern electronic equipment must be able to survive harsh environments. The endurance of such equipment is defined by the durability of their internal sensitive components. Mechanical failures of electronic components such as microprocessors, capacitors, transformers, resistors, etc., are major roadblocks to design cycle time and product reliability. Fatigue failures are failures caused in components under the action of fluctuating loads. They are estimated to be responsible for 80% of all failures of electronic components. Fatigue failures occur when components are subjected to a large number of cycles of the applied stress. With fatigue, components fail under stress values much below the ultimate strength of the material. What makes fatigue life of component leads a very important mechanical characteristic. In the current programming tools for electronic equipment analysis, the outputs of random vibration analysis are limited to spectral densities and root mean-square values of the stress with no phase information (which is required to be able to calculate the von Mises equivalent stress). This information is insufficient to calculate fatigue. There are no analytic models of electronic components for fatigue life prediction taking into account arbitrary mounting types on printed circuit boards and geometry of component leads. For this purpose corresponding methodology and mathematical models are required. This article presents a general methodology of fatigue life prediction of these electronic components within printed circuit assemblies (PCA) under random vibration loads. Mechanical characteristics of components are determined using finite element modeling approach. Dynamic analysis is performed within ANSYS environment. The vibration analysis provides system characteristics such as modal shapes and, and dynamic responses including displacements, accelerations, and stresses. Using the results of analysis, fatigue life is predicted based material durability information. Automatization of PCA finite element model generation is realized as follows: Using PCA models constructed in ASONIKA-TM (automated subsystem for the complex analysis of constructions of printed circuit assemblies at thermal and mechanical effects), geometrical and material properties of electronic components are transferred to ANSYS preprocessor. The database of ASONIKA-TM contains a large number of parametrical models of electronic components with arbitrary mounting types. A program was developed that permits automatic transformation of ASONIKA-TM parametrical models into ANSYS solid model, and automatic finite element mesh generation. The program analyzes models information and generates a set of ANSYS macro commands which create finite element model. After finite element model is created, dynamic analysis can be preformed.