After continuously tracking SPI data on the JiajinYuan production line for three weeks, we found that the Solder Paste volume CV values of 0402 and 0603 packaged resistors and capacitors remained consistently above 9.2%, with bridging defects accounting for 37% of single-board defects, while the tombstoning rate stabilized at 0.18‰ after reflow, clearly exceeding the upper limit of the process window. This indicates that the current solder paste supply has exceeded the maximum capacity of the pads, and intervention must start from the source: stencil design.

The Duoxi technical team retrieved stencil usage records from the past six months and corresponding post-reflow AOI data for each batch, confirming that the problem emerged specifically after switching to 30μm laser-cut stencils with an 85% new aperture ratio. This issue is particularly prominent on PCBs with FR-4 substrate and immersion gold surface finish, as these boards have weak solder absorption and a high coefficient of thermal expansion, resulting in lower tolerance for solder paste volume. If the aperture ratio is not adjusted accordingly, solder paste is highly prone to tailing and accumulation during demolding.

Rather than simply adjusting squeegee parameters to mask the problem, we went directly back to the stencil design stage. We uniformly narrowed the aperture dimensions corresponding to 0402 pads from the original 200×120μm to 190×115μm, reducing the aperture area ratio from 85% to 78%. Meanwhile, we fine-tuned the 0603 area from 260×140μm to 250×135μm to ensure that the solder paste release volume matches the pad wetting capacity. This modification does not require replacing the entire stencil; it can be completed through local re-engraving, with the production line changeover cycle controlled within 2 hours.

After reducing the solder paste supply, the SPI-measured average volume dropped from 6.8ng to 6.0ng, and the standard deviation was compressed from 0.72ng to 0.51ng. More importantly, bridging defects decreased to 14% of single-board defects, and the tombstoning rate fell back to 0.07‰, without causing new issues such as cold joints or insufficient solder joint wetting. This indicates that Duoxi's adjustment this time precisely hit the optimal threshold range for solder paste volume.

We simultaneously tightened controls on solder paste warm-up and stirring, requiring all JiajinYuan solder paste to be left at a constant temperature for more than 4 hours before printing, and using a centrifugal mixer with fixed speed and duration to perform 60 seconds of pre-stirring. This prevents fluctuations in Solder Paste Viscosity from amplifying the sensitivity introduced by aperture ratio adjustments; after all, no matter how precise the stencil design is, it cannot withstand the impact of uncontrolled material conditions.

On-site verification found that when the viscosity difference between solder paste batches exceeds 150Pa·s, even with an optimized aperture ratio, local demolding defects still occur. Therefore, we strictly require that each batch of solder paste must be accompanied by a viscosity test report before warehousing; if the data deviation exceeds the limit, an isolation process is immediately initiated to ensure that abnormal materials never enter the printing workstation.