Comparative Sensitivity Analysis for μBGA and QFN Reliability

摘要: The estimation of product reliability during the design is one key questions in microelectronics. assembly lifetime a function such parameters as geometry, material properties and loads. All these influences exhibit systematic stochastic variations. effect variability can be analysed by probabilistic FE-simulation statistical methods. This paper presents an approach for prediction thermal fatigue life two CSP types, μBGA QFN. Besides geometry parameters, also cycle temperatures are used variable inputs. Based on preliminary study input were defined normal distributed. Sensitivities to computed ranked after FE-simulations both QFN packages had been performed. solder joints this work based Coffin-Manson model it was performed using stress-strain data extracted from FE-simulations. As there exists dependency deformation behaviour, correct choice lead-free alloys important aspect work. Summarising, simulation method developed compute realistic failure distributions types.