Issue link: https://iconnect007.uberflip.com/i/989774
78 SMT007 MAGAZINE I JUNE 2018 In Figure 10, T5 solder paste is replaced by T4 and again, the trend continues, but demonstrates a more considerable impact of the reduced powder size. On an uncoated stencil the T5 solder paste produced accept - able volumes but unacceptable variation at 0.56 AR, and acceptable volumes and CVs obtained at 0.63 AR. The nanocoated sten- cil, however, produced acceptable prints with the T5 paste with AR as low as 0.50 even in the worst-case scenario of circular, copper- defined pads. It should be noted that the metal percentage by weight was lowered slightly from 88.5% in the T4 to 88.3% in the T5 to minimize solder powder "packing" during print, resulting in slightly lower viscosity paste. It is possible that the advantage of T5 was a combined effect of both the reduction in mesh size and formula- tion modifications. Peaking Peaking, or the tendency of the solder paste to stretch and snap during release, creates pointed deposits rather than flat top depos- its Figure 12. Pointed deposits are undesirable because they can be the root cause of solder bridging. To check the stencil-solder paste release relationship, the heights of the rectan- gular PTH patterns were analyzed. Rectangles were chosen because they offer two opportuni- ties per deposit to peak, or "dog ear," and the effect of peaking would be more pronounced in the SPI readings than similar measurements on square or circular pads with only one oppor- tunity. Figure 11 shows the average measured height of the mask-defined rectangles for the various pad widths. In this case, with T4 solder paste, the average heights of the deposits are 0.6-0.7 mils higher than those of the coated stencils. T5 paste showed somewhat less peaking, partially closing the gap between coated and uncoated stencils with differentials between 0.4 and 0.5 mils. Copper-defined pads showed even less peaking, with T4 showing a range of 0.3 to 0.4 mils, and the T5 showing a range of 0.2-0.3 mils. The trend was consistent for all pad and powder type combinations: uncoated stencils created greater peaks and dog ears than the coated stencils. In addition to the lower peaking, coated stencils also showed less variation in height than uncoated, cutting it by half or more, and Figure 10: Print performance of Type 5 powder with worst-case scenario pad design.