Abstract
The introduction and usage of smaller component packaging in electronic assemblies are driven by the need to
increase their I/O density and to reduce their size. Two of the more popular emerging packaging technologies are
Chip Scale Packages (CSPs) and 0201 discrete passive device packages. CSPs are currently in high volume
production with smaller pitch down to .5 mm on the horizon. 0201 packages are reportedly in full production now
by a few manufacturers with many others planing to be in full production later this year. These new packages are
initially expected to be used in hand held and portable products where size and weight are critical factors in the
success of the product. Even without the usage of the 0201 and CSP packaging, solder paste related defects
continue to top many manufacturers lists as the process step responsible for the greatest percentage of end of line
defects, typically in the 40-50% range1,2. The introduction of these new packages into the manufacturing process
will present new challenges for controlling the solder paste printing process.
Introduction
Reliability is also an important requirement for hand held types of products. Because of their portable nature, many
of these products will be subjected to harsh temperature extremes, mechanical shock, and vibration over the
course of their product life. For this reason, solder joint reliability is an important concern in the design and
manufacture of these products. For CSP packages, package standoff from the board has been determined to be
an important parameter for predicting long term solder joint reliability. Pad size and solder paste volume are the
primary factors that influence this3. A higher standoff seems to help the solder joint withstand more strain. Unlike
gull-wing QFP packages where the lead itself will absorb strain due to differences in the coefficient of thermal
expansion (CTE), for CSP packages the strain is concentrated primarily in the solder joint itself4.
The two most common problems with 0201 assemblies are reported to be tombstoning and solder beading. Both
defects are influenced by a variety of design and process parameters including pad size, placement accuracy, and
solder paste volume. Beading occurs when the solder paste brick splatters as a component is placed into it. After
reflow, the splatter becomes a spherical solder ball that can potentially cause a short circuit5.
Paste release from apertures is an important issue for both CSPs and 0201s with solder paste volumes decreasing
significantly to well below 1000 cu. mils. As the stencil aperture area shrinks to accommodate a smaller pad size,
the ratio of the aperture area to the surface area of the aperture sidewalls decreases, causing the paste to stick to
the walls rather than the intended pad. The general guidelines suggest that the aperture width to stencil thickness
ratio should be greater than 1.5. Square apertures are frequently used to deposit paste on round pads to facilitate
paste release. Paste formulation and the powder size in particular is also an important factor influencing paste
release from the stencil. The general rule here is the aperture opening should be greater than 3.5 particles across
to prevent clogging. Hence Type 3 and Type 4 pastes are required. However these paste formulations are less
viscous and they tend to slump on larger deposits intended for larger components5,6.
The introduction of these emerging package technologies significantly impacts the manufacturing process. Paste
formulations, pad sizes, and stencil designs must change. Several design and process guidelines have already
been established for the usage of CSP and 0201 packages. However much work remains to be done to
characterize the process before usage of these new packages becomes routine. Eventually “best practice” design
and process guidelines will evolve for these assemblies, but in the interim, many manufacturers are in the
experimentation phase to discover what works best.
The newer packages and the types of assemblies where they are used will also have a big impact on traditional
end of line test and re-work methods.Visual inspection of fine pitch assemblies, which requires 10X magnification,
is inconsistent and subjective7. In the case of CSP and other array packages, the solder joints are hidden beneath
the package so traditional test methods such as visual inspection are not possible. Camera based AOI systems
also cannot see these hidden solder joints. 0201 packages are about the size of a grain of salt so both the package
and solder joints are just too small to be visually inspected in a high volume production environment. CSP and 0201
packages will likely both be used in hand held products where package density will be high and this can cause
“line of sight” difficulties for some AOI systems. Also the types of products that use these packages will likely be
difficult to test with In Circuit Test (ICT) because of the small geometry and lack of test points. X-ray may be
successful and finding several types of end of line defects, but at this point the part must now be scrapped or reworked.
Rework is expected to be more difficult and expensive with 0201s and CSPs in particular because the
device and labor costs may be high8.With rework becoming more difficult and in some cases not feasible, the only
choices are to accept the increased costs, or to improve first pass yields. If the manufacturer chooses to improve
yield, it only makes sense to improve the process step responsible for the greatest number of defects and that is
most commonly the solder paste printing process1. When these problems are found before the next process step,
the cost to correct the problem can often be reduced by 10 times or more 6. A more effective test strategy would
be to improve first pass yields and prevent these defects from occurring in the first place by inspecting the paste immediately after print.
Why Is 3-D Better Than 2-D?
Solder paste inspection usually focuses on the variables that are the best predictors of solder joint quality: solder
paste volume, height, area, registration, and paste smearing. In order to accurately measure solder paste volume
and height, some type of 3-D sensor is required. The other requirement of a good solder paste inspection tool is
that it not only does a pass/fail inspection, but performs measurements. The measurements should meet
requirements for accuracy and repeatability so that the inspection system can be qualified for the process it is
trying to measure. A system without the required precision will not have the ability to resolve process variations that can affect product quality. Doing any type of inspection with a tool that is not stable, repeatable, and accurate
enough will not be effective and will most likely cause a lot of frustration.
A broad variety of inspection equipment are being marketed today for inspecting solder paste immediately after
print. These systems use a variety of methods and techniques, but can be generally speaking can be categorized
as either 2-D or 3-D systems.
Types of Inspection Systems
2-D systems are camera based systems that are capable of measuring area coverage and alignment. They are
sensitive to color and contrast changes, e.g. the color of the solder mask, pads and paste and often require
constant lighting or algorithm adjustments.They lack height and volume information and area alone is not a reliable
indicator of the correct amount of paste present. Since solder paste volume is the most important predictor of a
good quality solder joint, height information from an inspection system is essential.
3-D systems are capable of providing both height and area measurements. True solder volume information can be
acquired if a sufficient number of height data points can be measured. 3-D has the added benefit of being
insensitive to color and contrast changes.
3-D Data Acquisition
Almost all 3-D solder paste inspection systems utilize triangulation to compute the height of solder paste deposits.
Some systems use structured white light and a video camera to compute height while others use raster scanned
laser beams and proprietary detector technologies instead of video cameras. The 3-D laser scanner used here has
a solid-state scanning and detection system in conjunction with signal processing electronics to collect accurate
high-speed height data. A laser spot is projected down onto the surface being measured and receiving optics are
used to focus the reflected diffuse light onto a position sensitive photodetector. The height of the surface returning
the beam is determined from the position of the reflected laser energy on the photodetector. Analog electronics
then convert each detected spot into a digital word coded with a discrete height value. |
