Issue link: https://iconnect007.uberflip.com/i/699765
July 2016 • SMT Magazine 73 convex (+) at room temperature and changed to concave (–) at elevated temperature. One sample behaved in a unique way—it was con- cave at room temperature and less concave at elevated temperature. Other samples retained their shape of either concave or convex across the temperature range with minimal magnitude changes. Depending on the construction of the package, they exhibited different room temper- ature warpage magnitude ranges from ~50 µm to 150 µm, while at high temperature warpage floated more than –100 µm. The effect of Bake and MET at room tem- perature and 260°C are shown in Figure 3. It was determined that the package warpage was technically comparable among the precondi- tioning considered. There was some indication that baking the package reduced some variation of the warpage magnitude, but a more statistical approach is required to confirm this finding. In general, the data suggested that these ultra-thin packages were not sensitive to the preconditions considered, but this may not be generally applied to all PoP packages. This is because the package construction and material used can play signifi- cant roles in determining the dynamic warpage behavior and the impact of preconditioning. As for the PoP memory package, the dynamic warpage of the individual memory with differ- ent attributes is shown in Figure 4. Depending on the memory package attributes, the dynamic warpage ranged from ~100 µm to –100 µm and exhibited different dynamic warpage character- istics depending on the construction and mate- rial sets used. Some packages showed predomi- nantly convex shape at room temperature and became concave at higher temperature, while others could remain predominantly concave across the temperature range. The range of dy- namic warpage seemed to be higher for thinner and larger package size. The reduction of ball pitch with thin package design may suggest a greater challenge in PoP component board as- sembly in the near future. Figure 4: Dynamic warpage of PoP memory package warpage. Figure 3: Effect of 'as is', Bake and MET 9 days to package warpage at room and peak reflow temperature for PoP. PACKAGE-ON-PACKAGE WARPAGE CHARACTERISTICS AND REQUIREMENTS