An advanced, systematic simulation approach for studying warpage drivers of an assembled printed circuit board in early development stage

Digitalization, autonomous driving and the internet of things are pushing electronics industries towards modular build-ups for the latest generation of electronic devices. The warpage potential of such electronic devices during a temperature cycle is the major key performance parameter for assembled printed circuit board (PCBa) reliability and must be reduced to a minimum. Considering the actual printed circuit board (PCB) designs, the related materials or the different component types assembled, various parameters are available to influence the warpage of the final PCBa. In development of electronic systems, finite element analyses can be a key engineering tool to identify warpage drivers and reduce the warpage in an early development stage. Nevertheless, in order to fulfil the demand for short and efficient development cycles, it is necessary to provide standardized simulation modelling techniques and material test protocols.The presented use case demonstrates the application of the AT&S simulation approach to asses PCBa warpage. In order to model the PCBa, all relevant material parameters were determined by utilizing a combination of mechanical test procedures and micromechanical methods. The glass fiber reinforced dielectric material parameters are considering the orthotropic behavior of the laminates as well as the temperature dependency of the epoxy-based resin. Tensile tests of the thermosetting resin system were also performed at elevated temperatures to evaluate nonlinear stress-strain characteristics. With respect to the copper material model, plasticity was considered. These methods are already on a certain level of standardization for the daily application of material models in the simulation environment. The PCB model itself was generated by a homogenization approach, considering the actual PCB layer design as well as the related material properties. Active components mounted to the PCB were modelled based on cross sections of the used components.In the first step, for all individual components as well as for the PCB bare board, warpage was calculated. In the second step, the PCBa warpage was numerically determined. Based on the results for components, bareboard and PCBa, the main drivers of the system could be identified and quantified in a subsequent step. All simulations were done in ABAQUS
^®
by using a nonlinear static approach. In parallel, a validation build of the investigated module was done. The occurring PCBa warpage was measured over a temperature cycle by the utilization of digital image correlation (DIC). In the last step, numerical and experimental findings were compared to each other.Based on the presented procedure, the warpage to be expected for the PCBa after reflow could be calculated, and the influence factors for the PCBa warpage from PCB and large active components could be quantified. It can be concluded that the PCB and the component layer for the specific use case are contributing to a similar amount to the overall occurring warpage of the PCBa.A measurement of the PCBa warpage, carried out utilizing a 3D digital image correlation system, showed very good alignment of the simulation results and warpage values measured for the hardware PCBa. Additionally, the lead times of the simulation and the sample build are compared and subsequently the potential of the use of a standardized, systematic finite element simulation approach in product development is highlighted.