An Improved Automated SPI Data Analysis Report Generator for Printed Circuit Board Assembly

Optimizing solder paste deposition and transfer efficiency is essential for successfully assembling printed circuit boards (PCBs) in Surface Mount Technology (SMT). In a previous effort, we introduced an innovative automated approach that combines Design of Experiments (DOE) and C
pk
analysis with Solder Paste Inspection (SPI) data to optimize printing parameters, evaluate SPI data after printing, and generate automatic reports for the results. This work updates our automated SPI data analysis tool to include C
pm
offering more accurate analyses as it circumvents C
pk
centrality requirement between specification limits. Additionally, we have introduced heat maps as visual representations of solder paste volume transfer efficiency in this upgraded version, offering a more robust understanding of solder paste deposition across different locations. The first part of our study involved a quick full-factorial DOE around three stencil printing parameters: squeegee speed, squeegee pressure, and snap-off speed, each with two levels to determine machine setup values for optimizing solder paste deposits on complex PCB components, like high I/O ball grid array packages. This approach rapidly generates an 8-treatment (2
³
) design, optimizing parameter values based on response variable outputs such as mean volume deposition, standard deviation, and C
pm
. In the second part, our focus shifts to an automated setup that evaluates mean volume deposition, standard deviation, and C
pm
values for transfer efficiency per component for all boards in a build. This setup extracts solder paste volume percentages from the SPI machine and calculates response metrics values for each component, providing a comprehensive assessment of transfer efficiency across the entire assembly. Historically, stencil printing DOEs and process capability analysis for transfer efficiency have been labor-intensive and time-consuming, often performed manually on a sample basis. Our proposed fast, automated setup reduces human error, saves time, and enhances process control. Results from an actual SMT process were used to validate the effectiveness and reliability of this setup in improving transfer efficiency and optimizing process capability. These results are significant for enhancing process efficiency, ensuring quality, and reducing costs in SMT assembly.