Printing offset or solder draw spikes occur frequently, usually caused by the low viscosity of Domestic Solder Paste leading to Solder Paste sticking to the stencil during demolding. The practical solution is to first reduce the squeegee pressure from 4.5kg to 3.8kg, simultaneously lower the printing speed from 40mm/s to 25mm/s, and use a viscometer to actually measure whether the viscosity of the new batch of solder paste at 25℃ falls within the 650±50Pa·s range. The key preventive measure is that the matching verification between stencil apertures and solder paste particle size must be completed before each product changeover, with special attention to whether the aperture aspect ratio corresponding to large pads above 1206 still meets the 1.5:1 requirement.
After reflow, a large number of tombstone defects are concentrated on 0402 and 0603 resistors and capacitors. The root cause is that the surface tension attenuation curve of the rosin-based Flux in Domestic Solder paste has a time difference of about 2.3 seconds compared with imported materials. The practical solution is to fine-tune the temperature of the second preheating zone of the reflow oven from 150℃ to 147℃, extend the soaking zone time by 18 seconds, and use thermocouples to actually measure whether the temperature difference on the PCB surface is controlled within ±2.5℃. The preventive measure is that in the small batch verification stage, PCBs from the same batch must be passed through the oven 5 consecutive times to observe the trend change of the tombstone rate.
Uneven thickness of the solder joint IMC layer and blackened edges are mostly seen on the bottom pads of QFN devices. The cause is the fluctuation of halogen content in domestic solder paste leading to reduced oxide film removal efficiency. The practical solution is to take 3 PCBs that have been mounted but not reflowed, and test the wetting angle in environments with nitrogen concentrations of 95%, 98%, and 99.5% respectively, to confirm that when the nitrogen purity is ≥98.2%, the wetting angle is stable at 28°±3°. The key preventive measure is that before each shift starts, a portable dew point meter must be used to detect whether the dew point value at the end of the nitrogen pipeline is lower than -40℃, and the binding relationship between the corresponding solder paste batch number and nitrogen parameters must be recorded.
The stencil life drops sharply to 1200 pieces before serious aperture clogging occurs, which is mainly caused by the slow complexation reaction between some activator components in domestic solder paste and the nickel alloy stencil. The practical solution is to use an ultrasonic cleaning machine with a 5% citric acid solution to clean the stencil every 400 pieces, and use a 100x microscope to spot-check whether there are tan deposits on the inner wall of the apertures. The preventive measure is that when introducing a new brand of domestic solder paste for the first time, a laser one-step forming stencil must be replaced synchronously, and mixing cleaning agents with chlorine-containing solvents is prohibited.
The proportion of cold solder joints rises to 0.7% after batch switching. The root cause is that the wide melting point range of domestic solder paste compresses the peak temperature window. The practical solution is to use a KIC thermocouple to actually measure the actual peak temperature at the center of the BGA device; if it is lower than 235℃, the temperature of the reflow oven peak zone should be increased by 3℃ as a whole, and the heating slope should be slowed down from 1.8℃/s to 1.5℃/s. The preventive measure is that after each batch of solder paste arrives, 50g of sample must be taken for DSC differential scanning calorimetry analysis to confirm whether its liquidus temperature is strictly controlled at 217.5±0.3℃.
The ICT test missed detection rate increases, and a semi-transparent residual film is visible on the solder joint surface under a microscope, which is caused by insufficient volatilization of the new polymer resin carrier in domestic solder paste in the low temperature section. The practical solution is to reduce the air speed of the reflow oven cooling zone from 1.2m/s to 0.9m/s, and add a 30-minute constant temperature baking process at 65℃ after the oven. The preventive measure is that in the verification stage, an ion contamination tester must be used to measure whether the NaCl equivalent on the soldered board surface is ≤1.5μg/cm2, and the data must be cross-compared with the solid residue index in the solder paste TDS file.
Solder beads with diameter >0.15mm account for more than 12%, concentrated at the root of QFP pins. The main reason is that the thixotropic recovery time of domestic solder paste is 1.7 seconds slower than that of imported materials, leading to increased collapse of the paste after printing. The practical solution is to uniformly adjust the stencil thickness to 0.12mm, change the squeegee angle from 60° to 58°, and add a 30-second vacuum standing step before component placement. The preventive measure is that all verification boards must use the same SPI device to collect the volume coefficient of variation CV value, ensuring it is stable within the range of 9.2%±0.8%.
