Authored By:
Zachary Slater, MSc. and Martin K. Anselm, Ph.D.
Rochester Institute of Technology
NY, USA
Summary
This study investigated the cause of tombstoning for 0402 capacitors in surface mount technology (SMT). It serves as a follow-up investigation to the study conducted by Feng in SMTAI 2022 and sought to determine the cause of the high occurrence of tombstoning seen in Sample Group 4, even though the printing quality fell within the standards set by IPC-7527. Statistical analysis methods were used to determine if there was any significant difference between the target experimental printing offset and the additional SPI measured offset of the printed solder paste.
Additional analysis was conducted to determine if there was a difference in solder paste accuracy between components that showed defects vs. those that did not. Finally, two test boards were subjected to reflow while oriented 90˚ counterclockwise to the original design to check the effects of changing the offset orientation of the solder paste. The analysis did reveal a significant difference in actual paste position from the target. However, the difference in the solder paste’s measured positions did not show any correlation to defects. This study further indicates that reflow orientation can lead to even more defects depending on the direction of the paste offset.
Conclusions
The statistical analysis revealed the positional accuracy of the printed solder paste. Though the analysis did show a significant error in positional accuracy, the two-sample t-tests revealed no significant differences in accuracy between the components that tombstoned vs. the ones that did not. This was consistent when both the real and absolute value of the offsets were compared. In other words, there is no evidence that indicates that random positional inaccuracies produced during printing were the cause of the high rates of tombstoning produced when given a 25% offset in the component’s Y-axis.
Similarly, though Figures 13 and 15 show a compelling trend in solder paste precision, the difference in position from the nominal value averages out at only about 6%. And regardless of that, this trend in offset was present in all of the components, whether or not they possessed a defect.
Merely changing the orientation of the PCB when sending it through the reflow oven, and thus changing the reflow direction, had a surprisingly strong effects on the results of the process. Originally, the components oriented at 0˚ displayed no tombstones at all. However, when those same components were subjected to reflow with a 90˚ orientation, tombstone defects were observed in that same section of the PCB, even though every other factor had remained the same. Even more intriguing is that when components were reflowed at 0˚, but with solder paste offset directed in the opposite direction almost all of them displayed defects.
This study strongly supports the observation that reflow direction is a major factor in determining whether or not components printed with a bowtie solder paste shape will fail, even if the print offset still technically lies within acceptable standards. It is even more important to note that this effect can be both helpful or harmful, as it led to almost no components failing at the 0˚ orientation in the original study and nearly every component at the same reflow orientation failing in this study.
Printing bowtie deposits can result in one end of a component having a larger degree of contact with the solder paste than the other end. This significantly raises the likelihood of defects, regardless of which way the offset is biased.
Initially Published in the SMTA Proceedings
|
