Stencil Printing Process Optimization to Control Solder Paste Volume Transfer Efficiency

This research aims to optimize the stencil printing parameters to control the solder paste volume TE and increase the first-pass-yields of printed circuit boards (PCBs). Stencil printing process (SPP) factors considered in this research are related to solder paste composition, product configuration, printer setup, and stencil design. Support vector regression (SVR) with different kernels and regression tree (RT) are used to estimate the solder paste volume transfer efficiency (TE) and capture the nonlinear relationships between the SPP factors and volume TE. A novel mixed-integer nonlinear programming (MINLP) model is proposed to optimize the printing parameters for different PCB configurations of pads. The epsilon-constraint approach is used to solve the proposed MINLP model using metaheuristics, such as simulated annealing (SA) and evolutionary strategies (ES). The experimental results indicate relatively high R-2 values of 81% and 72% for average and standard deviation volume TE models that use SVR and RT, respectively. Then, the optimal printing parameters for certain pads' configuration are retrieved for different volume TE spec limits. The results show that SA and ES can find optimal solutions for the spec limits of 50%-150%, 70%-130%, and 80%-120%.