Molded underfill process for the SiP
ity. Those MUF voids are from whole Internal void or delamination is anoth- this case, resin does flow very fast. On the
Legs except 4, 5, and 8. er main cause for low reliability. A SAT test other hand, the pressure is not high. There
Figure 5 shows the cross-section images is used for this analysis. Only Leg 8 shows possibility exists a flow speed difference
of each Leg sample. From cross-section the internal voids and delaminations. So, between the under chip and path areas.
analysis, the sample results can be seen. Leg 8 EMC is also eliminated. This is expected for the void to be trapped
The void is located at the center of chips. Another important factor for EMC under the chip.
Void shape is trapped. It may be formed selection is warpage. The warpage is mea- Figure 9 shows the second test sample
during epoxy transfer. During the epoxy sured after PMC (post mold cure). Results and SAT analysis results. Only Leg 5 shows
transfer, there are many registers. In the are clarified by measuring window warpage. MUF void while other Legs show no trap
window, there are 16 units. Each unit Figure 8 shows warpage while Legs 4 and void. So, it confirms the study. Transfer
is located side by side. During the resin 6 show least warpage. From above EMC mold transfer time and pressure are critical
transfer, the flow is accelerated along paths screen test, Leg 4 is selected. Legs 5 and 8 to mold quality. Those two parameters
compared between top and under areas of show no void. Nevertheless, external visual control the resin flow condition.
the dies due to thin spacing. However, un- defect, internal void, and warpage property Faster transfer speed (short transfer
der the die, resin flow is slower for the top hurt our process. So, they are removed. Leg time) results in short molding or porosity
die. Faster flow from path and top reaches 4 is the best material. It shows least war- inside the package due to turbulent flow at
the end die. So, void is trapped under the page level, no MUF void, no external void, the filling end region. Inside the cavity, the
chips. There is a void forming flow. Never- no internal void, no delamination, and no dimension of resin path is quite different.
theless, Legs 4, 5, and 6 show no voids. It FM contamination. In the narrow pass area, the resin flow
means chip under resin flow is faster or the Confirming the process parameters is speed is fast even though it is set at high
same compared to both path and topside. done with additional tests. Table 5 shows pressure. In the wide pass, the resin flow
The same resin flow speed needs to be the test Legs and second test results. Trans- speed is reduced at lower pressures. During
determined in terms of material and flow fer pressure and speed are the main process resin transfer, the melted resin velocity fluc-
simulation. During test, foreign material parameters causing void formations. So, tuates. By the pass shape and register, long
contamination is determined. It is called five Legs with different transfer pressure transfer time makes flow more stable
8
. Flow
the other external visual defect. This defect and speed are determined. From the speed control is very important to prevent
happens for Legs 5 and 6 EMCs. Figure second test, only Leg 5 shows MUF void. void trap under the chip resin. Flow under
6 shows the FM (foreign material) defect Leg 5 has the least transfer pressure (800 the die is unavoidably slow by high register
image. This problem is material, and Legs 5 psi) and the fastest transfer speed (8 sec). In in comparison to resin flow around the die.
and 6 EMCs are eliminated from the list. Consequently, it ultimately traps air under-
!
neath the die, thereby negating the inten-
!
tion of the underfill. Hence, die dimension
is limited. Underfilling big die dimension
with transfer mold, material, process, and
vent hole designs is needed. However, in
this test, three 3 x 3 mm chips SiP molded
underfill is possible. Underfilling quality
!"#$ !%#$
depends not only on material property
Figure 3. MUF (molded underfill) good sample SAT Figure 4. MUF (molded underfill) void sample SAT
such as filler dimension spiral flow, but
(scanning acoustic transmitter) analysis image. (scanning acoustic transmitter) analysis image.
also on molding process parameters, PCB
(a) C-scan image (b) T-scan image. (a) C-scan image (b) T-scan image.
surface energy, chip dimension, and unit
Figure 5.Cross-section Legs analysis images. (a) Leg 1 (b) Leg 2 (c) Leg 3 (d) Leg 4 (e)Leg 5 (f) Leg 6 (g) Leg 7 (h) Leg 8.
14 – Global SMT & Packaging – December 2008
www.globalsmt.net
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60