This solder sample test selected a new batch of lead-free Solder Paste from Jia Jinyuan and the old batch currently in use on the production line for parallel comparison. All tests were completed on the same DEK Horizon 03i printer and BTU Pyramax 125 reflow oven to avoid interference variables introduced by equipment differences. The stencil uniformly adopted 0.10mm thick SUS304 material with laser cutting + electropolishing treatment. The corresponding aperture size for the new solder was reduced by 5% according to the IPC-7527 standard, with the aspect ratio maintained above 1.5, while the old solder still used the original designed aperture. In the printing parameters, the squeegee pressure was set to 4.2kg, the speed was 45mm/s, and the angle was 60 degrees, all three of which were kept unchanged synchronously to ensure variable uniqueness.

For temperature profile collection, a KIC Navigator six-channel temperature profiler was used, with 3 thermocouples inserted in each test group, attached to the center positions of QFN48 (0.4mm pitch), 0201 resistor, and PCB board respectively. The measured peak temperature of the new solder was 238℃, the time above liquidus was 86 seconds, and the heating slope was 1.8℃/s. The corresponding values for the old solder were 241℃, 92 seconds, and 1.6℃/s. Although the difference was small, it had already affected the tendency of solder ball formation, especially 3 micro solder beads appeared in the 0201 area, while the new solder only had 1, indicating that its Flux activity release was more moderate and volatilization was more sufficient.

In terms of stencil parameters, since the new solder had slightly lower viscosity (measured 220Pa·s@shear rate 100s?1) and higher metal content (90.2wt%), the aperture area ratio was increased from the original 0.62 to 0.65. At the same time, the QFN pad opening was indented inward by 0.015mm to suppress trailing. SPI inspection after printing showed that the average volume deviation of the new solder was ±6.3%, and that of the old solder was ±8.7%. Moreover, the residual amount at the bottom of the stencil after continuous printing of 200 pieces was reduced by about 40% for the new solder, indicating that its anti-clogging ability was stronger.

After reflow, the number of cold solder joints detected by AOI was 7 per thousand points for the new solder and 12 per thousand points for the old solder. Among them, the average void rate at the bottom of QFN was 18.5% for the new solder and 24.1% for the old solder. X-ray sampling confirmed that the voids were concentrated at the four corners of the thermal pad, which was directly related to the temperature plateau at the end of the reflow soaking section being 0.8℃ lower. The new solder, due to its accelerated wetting and spreading speed, had higher actual filling efficiency under the same reflow temperature settings, which was equivalent to a disguised increase in energy input in the thermal pad area.

All solder sample test data were recorded into the MES system and bound to batch numbers. The new solder did not trigger any welding defect line stoppage in the trial production of 1200 pieces in 3 consecutive batches, while the old solder had a 17% increase in AOI false alarm rate due to solder beads in the same period test, requiring an additional 2.3 seconds per board for manual re-inspection. This shows that solder sample testing is not only a material replacement verification, but also a key input node for recalibrating the parameters of the entire SMT line.