PoP: An EMS perspective on assembly, rework and reliability
sample size
supplier Flux Paste
A 26 29
B 26 26
C 26 26
Table 1. Build matrix for fluxes and pastes.
sample size
alloy Flux Paste
Figure 11. Upper package balls after dipping in Type Figure 12: Upper package balls after dipping in Type
V solder paste. VI solder paste.
SAC105 26 29
SAC305 13 14
Once all packages had been placed, the shorting on the 0.5 mm pitch joints on the
SAC405 13 13
entire assembly was reflowed in a nitrogen lower package. As these joints were formed
environment. A single reflow profile with by the conventional screening and reflow
Table 2. Build matrix for ball alloy variations.
settings suitable for all ball alloys included process, these failures are not specifically
in the study was linked to the package on package assembly
Paste
used. The peak process but rather indicate the need to
supplier underfill Pattern sample size
temperature for adjust the screening process to improve
the reflow process the quality of the paste prints. An x-ray
A 1 Bottom 13
ranged from 239°C image of one of the shorted components is
A 2 Bottom 13
to 243°C, with time shown in Figure 13.
above 217°C rang- No electrical or x-ray failures were
B 1 Bottom 13
ing from 64 to 76 noted on any of the reworked components.
B 1 Both 13 seconds.
B 2 Bottom 13
Time zero cross sections
Forced reworks Five components were removed for cross
B 2 Both 13
Two cells of re- sectioning prior to thermal cycling. Three
C 1 Bottom 13
worked components primary attach package stacks were sec-
were included in tioned, one of each component ball alloy.
C 2 Bottom 13
the test matrix. The All of the primary attach samples for time
Table 3. Build matrix for underfill variations.
first cell consisted of zero cross sectioning were assembled using
11 SAC105 package dip paste from Supplier A. Two rework
stacks reworked with samples were also sectioned, one from each
Test vehicle assembly dip paste from Sup- rework cell. As the top and bottom rows of
All test vehicles were assembled on a con- plier B. The second set consisted of nine joints in the package do not align exactly,
ventional SMT line. The test vehicle PCBs SAC305 package stacks reworked with the two sections for each package were made,
were all screened with a no-clean SAC405 same dip paste. The sample size for the sec- one through the centre of a row of joints
solder paste using a 5 mil thick stencil. The ond group of reworks was limited to eight on the upper package and one through the
bottom packages were placed directly on due to the number of packages available. centre of a row of joints on the lower pack-
the test vehicle while the top packages were One sample from each of these two cells age. Images of typical upper joints formed
dipped in either flux or dippable paste was used for a time zero cross section, and from SAC105 balled components and
prior to being placed on top of the lower the remainder were subjected to acceler- lower joints formed from SAC125 balled
packages. A linear dip unit installed in the ated thermal cycling. components are shown in Figure 14 and
placement machine was used for all pack- The rework process consisted of bond- Figure 15, respectively.
age dipping. As the dippable solder paste ing the packages together using adhesive, Standoff measurements were made on
is a relatively new material, dipping trials then removing the stack of packages from
were carried out prior to the main test the card using a hot air rework station. Af-
vehicle build. These trials aimed to ensure ter site dress, the lower package was dipped
adequate transfer of paste to the compo- in paste then placed on the card. The up-
nent solder balls was taking place. Good per package was then also dipped in paste
results were obtained for both the two and placed on top of the lower package.
Type V materials and the Type VI material, The entire stack was then reflowed.
although differences were noted in the ap-
pearance and uniformity of the transferred assembly results
material. Generally, the Type VI material Assembly yield
yielded smoother, more uniform paste Following assembly, all packages were
deposits on the ball. The Type V deposits inspected using transmissive x-ray and
tended to be taller and more variable in electrically tested using a multimeter. Only
Figure 13: Short on first layer, primary attach
volume of material transferred. two components failed—both failed for component.
20 – Global SMT & Packaging – March 2009
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 |
Page 61 |
Page 62