Issue link: https://iconnect007.uberflip.com/i/1243344
28 SMT007 MAGAZINE I MAY 2020 in Figure 3. The Weibull distribution plots show that HRL1 alloy/SAC305 hybrid solder joints drop shock characteristic life is 82.7% of SAC305 homogeneous solder joints in the BGA84. The LGA84 is used as a quick method to evaluate the drop shock behavior of homo- geneous solder joints. In this case, HRL1 alloy drop shock characteristic life is slightly higher than SAC305. In each case, HRL1 and SAC305 Weibull curves are within the 95% confidence interval. It is also interesting to note that the shape parameter for HRL1 and SAC305 are iden- tical for the BGA84 (1.27) and almost identical for the LGA84 (1.83 and 1.73, respectively). A plot of the respective distribution fits without plotting the sym- bols, as shown in Figure 3 (b), highlights the aspects discussed previously. Thermal reliability tests were performed using a single zone air-air thermal shock chamber, where the samples cycled from -40 to +125°C, with 10 minutes dwell time at each temper- ature up to 2,000 cycles. The electrical resistance of the components was continuously monitored, and failures were defined as an increase of 20% or more in electrical resis- tance for five consecu- tive readings, as described in the IPC 9701-A stan- dard. Figure 4 shows accu- mulated failures after 1,000/1,500/2,000 ther- mal cycles. For the in-situ monitoring, only hybrid LTS/SAC305 solder joints were considered while compared to SAC305 joints. No failures were observed in the first 1,000 cycles. After 1,500 cycles, failures peratures above 200°C, which would defeat the purpose of using a low-temperature alloy in lieu of SAC alloys. Besides that, reducing Bi content from 58 to 40 wt.% increases the drop shock characteristic life (i.e., 63.2% cumula- tive failures) in 77%, but such performance is still 40% lower than the requirement for a SAC305 drop-in replacement. Among dozens of Sn-Bi alloys with various combinations of additives, it was found that HRL1 had the best drop shock performance for both hybrid and homogenous joints, as shown Figure 3: Probability plot of BGA84 drop shock failures. (a) Data point symbols and Weibull distribution fit and (b) Weibull distribution fit omitting symbols.