|
SMTA Pan Pacific Symposium
|
|
|
Assembly and Process Challenges for High Density PoP (Package-on-Package) Utilizing SOP (Solder-on-Pad) Technology
Joanna Wildhart, Moody Dreiza |
|
|
This paper covers the assembly and process challenges related to high density Package-on-Package (PoP) utilizing solder-on-pad (SOP) technology. PoP architecture integrates logic (bottom package) and memory (top package) into a single 3D packaging solution. This packaging configuration makes it a popular solution for consumer handheld devices. As popularity for PoP increases, especially in the cellar and MP3 markets, so does the need for increased functionality, performance and memory. This trend towards more complex memory is resulting in the need for finer pitch on the top package. On the bottom package, the trend towards more functionality and extended logic function is driving the need for stacked die. These two trends are conflicting since the finer pitch on the topside reduces the stand-off height between the top and bottom package. On the other hand, the need for stacked die in the bottom package drives the need for increased stand-off height between packages. Thus, a solution is required that will provide additional top package standoff while keeping the same small package profile. That is, allowing high I/O count between the packages at a fine pitch while clearing a thicker moldcap on the bottom package. This challenge is addressed by solder-on-pad (SOP). SOP technology works by placing a solder ball on the top of the bottom package. The solder ball on the bottom of the top package is then aligned to the solder ball on the top of the bottom package and reflowed together. While this technology addresses the problem of increasing stand-off height, the resulting ball-to-ball stacking presents new challenges in terms of surface mount process. This paper will analyze several material and process conditions for SOP PoP assembly. First the assembly process and test vehicle will be described. Next, the optimized parameters for placement of the top package will be explained. Three different stacking materials (standard PoP flux, standard FBGA dipping PasteA and PoP specific dipping PasteB) and three reflow profiles (standard, longer warm-up time and increased peak temperature) will be analyzed in detail. The parts from each leg will be inspected through visual inspection and x-ray analysis for failure analysis. Joint integrity will then be studied by cross-section with SEM inspection and dye-pry analysis. Finally, stand-off height will be measured. The best material and process conditions will be determined based on final yield and stand-off height. This paper will show that stacking material and, to a lesser degree, reflow profile have a significant effect on the stacking yields. |