|
|
|
|
|
Backward Compatible Solder Joint Reliability Under Accelerated Conditions
Production Floor |
|
Authored By:Richard Coyle, Peter Read, Richard Popowich, Debra Fleming Alcatel Lucent, Murray Hill, NJ Vasu Vasudevan, Raiyo Aspandiar Intel Corporation, Hillsboro, OR Iulia Muntele Sanmina Corporation, Huntsville, AL Steve Tisdale Intel Corporation, Chandler, AZ SummaryAccelerated temperature cycling (ATC) was used to assess the thermal fatigue reliability of a Pb free, 37.5 mm fully populated, 1284 I/O ball grid array (BGA) package assembled with backward compatible, mixed alloy (Pb free BGA/SnPb paste) assembly processes. Two different temperature cycling profiles were used in the evaluation. The baseline profile was the standard accelerated temperature test cycle of 0 to 100 C as prescribed by the IPC-9701 attachment reliability guideline (TC1). The second profile was a mildly accelerated cycle using a less aggressive temperature extremes and a smaller T with a resultant range of 20 to 80 C. The limited temperature extremes provided by the latter cycle result in lower strains more typical of service conditions. The surface mount assembly was done using custom SnPb eutectic soldering profiles designed to optimize the complete (full) mixing of the Pb and to create two additional test cells with levels of Pb mixing in the Pb-free BGA balls defined as low and medium. To complete the reliability comparisons and provide experimental controls, SAC405-SAC405 and SnPb- SnPb assemblies were included. The results indicate that backward compatible, mixed alloy assemblies should have acceptable reliability under service conditions. Complete or full Pb mixing is preferred in order to achieve consistent and acceptable solder joint reliability. ConclusionsThe results from this experimental study show that the thermal fatigue performance measured by characteristic lifetime of the large-body 1284 I/O full array BGA with Full Pb Mixed solder joints, is comparable to that of the 1284 I/O BGA with Pb-free SAC405 solder joints. The Full Pb Mixed and the SAC405 Pb free assemblies perform equally well when tested with the standard 0/100 C thermal cycle as with the mildly accelerated 20/80 C cycle. Additionally, the mixed and Pb-free assemblies outperform the SnPb eutectic assemblies by a factor of 2.5 to 3 in 0/100 C testing and by a factor of almost 5 in 20/80 C testing. The latter result indicates that the acceleration factor for mixed alloy assemblies is comparable to or greater than Pb-free or SnPb. There is no indication from the thermal cycling data or the metallurgical failure analysis that the Pb introduced by the mixed alloy assembly has a significant impact on thermal fatigue performance. The fracture features and failure modes of thermally fatigued mixed microstructures are the same as the fracture features of Pb-free SAC failures and the common fracture characteristics are consistent with their virtually identical fatigue performance. Although these experimental results are consistent with some previous studies it is important to recognize that in practice, mixed alloy assembly is a custom process and its level of risk always should be assessed with respect to specific package construction, product design, and assembly parameters. Initially Published in the SMTA Proceedings |
|
Comments
|
|
|
|
|
|
|
|
|