A Hybrid Finite Element Modeling: Artificial Neural Network Approach for Predicting Solder Joint Fatigue Life in Wafer-Level Chip Scale Packages

Abstract Fatigue life prediction of electronic devices is of great importance in both research and industry. Traditionally, fatigue tests and finite element modeling (FEM) are the two main methods. This paper presents a new hybrid approach (FEM combined with artificial neural network, (ANN)) for fatigue life prediction. Finite element models on wafer-level chip scale packages (WLCSP) with different chip thickness, PCB thickness, and solder joint pitches were created to evaluate the effect of structure parameters on the increase in maximum creep strain under thermal fatigue load. Modified Coffin–Manson equation was then employed to estimate the corresponding fatigue life. ANNs were built, and then trained, tested, and optimized with the datasets from modeling to predict increase in maximum creep strain and fatigue life. For the ANN built for strain prediction, prediction accuracy of the optimal network was 97% in accuracy tests and 93% in generalization tests. Accuracy of the other ANN for predicting fatigue life was 94.2% in accuracy tests and 88% in generalization tests. This hybrid method shows very promising application in fatigue life estimation of electronic devices which requires much less time and lower cost.

Download Full-text

Application of Artificial Neural Network for Fatigue Life Prediction

Key Engineering Materials ◽

10.4028/www.scientific.net/kem.297-300.96 ◽

2005 ◽

Vol 297-300 ◽

pp. 96-101

Author(s):

Ishak Abdul Azid ◽

Lee Kor Oon ◽

Ong Kang Eu ◽

K.N. Seetharamu ◽

Ghulam Abdul Quadir

Keyword(s):

Neural Network ◽

Finite Element ◽

Fatigue Life ◽

Solder Joint ◽

Finite Element Simulation ◽

Life Prediction ◽

Fatigue Life Prediction ◽

Element Simulation ◽

Parametric Studies ◽

Solder Joint Fatigue

An extensively published and correlated solder joint fatigue life prediction methodology is incorporated by which finite element simulation results are translated into estimated cycles to failure. This study discusses the analysis methodologies as implemented in the ANSYSTM finite element simulation software tool. Finite element models are used to study the effect of temperature cycles on the solder joints of a flip chip ball grid array package. Through finite element simulation, the plastic work or the strain-energy density of the solder joints are determined. Using an established methodology, the plastic work obtained through simulation is translated into solder joint fatigue life [1]. The corresponding results for the solder joint fatigue life are used for parametric studies. Artificial Neural Network (ANN) has been used to consolidate the parametric studies.

Download Full-text

The Effect of Intermetallic Compound in Solder Joint Fatigue Life Prediction Using Finite Element Modeling

2003 International Electronic Packaging Technical Conference and Exhibition, Volume 2 ◽

10.1115/ipack2003-35281 ◽

2003 ◽

Author(s):

Mohammad Masum Hossain ◽

Dereje Agonafer ◽

Puligandla Viswanadham ◽

Tommi Reinikainen

Keyword(s):

Finite Element ◽

Fatigue Life ◽

Intermetallic Compound ◽

Solder Joint ◽

Finite Element Modeling ◽

Life Prediction ◽

Temperature Cycling ◽

Fatigue Life Prediction ◽

Element Modeling ◽

Solder Joint Fatigue

The life-prediction modeling of an electronic package requires a sequence of critical assumptions concerning the finite element models. The solder structures accommodate the bulk of the plastic strain that is generated during accelerated temperature cycling due to the thermal expansion mismatch between the various materials that constitute the package. Finite element analysis is extensively used for simulating the effect of accelerated temperature cycling on electronic packages. There are a number of issues that need to be addressed to improve the current FEM models. One of the limitations inherent to the presently available models is the accuracy in property values of eutectic 63Sn/37Pb solder or other solder materials (i.e. 62Sn/36Pb/2Ag). Life prediction methodologies for high temperature solders (90Pb/10Sn, 95Pb/5Sn, etc.) or lead-free based inter-connects materials, are almost non-existent due to their low volume use or relative infancy. [1] Another major limitation for the models presently available is excluding the effect of intermetallic compound (Cu6Sn5, Cu3Sn) formation and growth between solder joint and Cu pad due to the reflow processes, rework and during the thermal aging. The mechanical reliability of these intermetallic compounds clearly influences the mechanical integrity of the interconnection. The brittle failures of solder balls have been identified with the growth of a number of intermetallic compounds both at the interfaces between metallic layers and in the bulk solder balls. In this paper, the effect of intermetallic compound in fatigue life prediction using finite element modeling is described. A Chip Scale Package 3D Quarter model is chosen to do the FE analysis. Accelerated temperature cycling is performed to obtain the plastic work due to thermal expansion mismatch between the various materials. Solder joint fatigue life prediction methodologies were incorporated so that finite element simulation results were translated into estimated cycles to failure. The results are compared with conventional models that do not include intermetallic effects. Conventionally available material properties are assumed for the eutectic 63Sn/37Pb solder and the intermetallic material properties. The importance of including intermetallic effect in finite element modeling will be discussed.

Download Full-text