Reliability investigation of system in package devices toward aeronautic requirements: Methodology and application

Abstract

System in Package components (SiP) are nowadays widely used in commercial telecommunication applications. Based on More than Moore approach, SiP components consists in increasing the number of integrated circuits/packages enclosed in a single module. The main advantage of those technologies remains the possibility to densify the number of Commercial Off-The-Shelf (COTS) subsystems at component level rather than at board level. However, if the asset of SiP components seems to be obvious in term of equipment design, their reliability is still questioning regarding aeronautic environments, which are characterized by harsh and long term stresses. Furthermore, due to their quite novelty, there is no feedback about SiP devices. As a consequence, it becomes mandatory for aeronautic end users to assess in phase lead, the reliability of those devices prior to use them in electronic equipment. On this view, the goal of this paper is based on two pillars. First of all, a general methodology devoted to the component selection as a function of the mission profile and the technology will be presented. In order to underline the method, two configurations of SiP components will be studied according to two aeronautic mission profiles. Those profiles typically concern airplane and helicopter applications and will be described by the means of thermo-mechanical and vibration stresses. Then, some finite element modelling and simulations will be performed at the 2nd reliability level in order to evaluate the impact of SiP architectures on the solder joints integrity. The targeted parameters investigated in this study will be the equivalent stress, the strain and deformation. Next, either the number of cycles or the time to failure will be determined for all SiP configurations and compared to the aeronautic reliability specifications at 2nd reliability level. This approach enables an end-user component to select a suitable SiP device for his application if reliability results got by simulations successfully passes the aeronautic specifications. However, reliability results are strongly linked to the design and process parameters such as solder joints, packaging dimensions and board design or the choice of process assembly for instance. Considering both the lack of design standards and the high number of SiP configurations available on market, guidelines do not exist for end user to evaluate the reliability of commercial SiP. Based on the studied configurations, this paper will supply some design rules of SiP ball grid array architectures at 2nd level of reliability in order to fulfil the reliability specifications. This study has been performed in the framework of the ENIAC project entitled ESiP and is a result of a European collaboration between EADS, Siemens and Murata Electronics Oy.