November- December 1998
eMail the Editor
Quality Assurance and Reliability: A System Approach
-By Dr. Reza Ghaffarian, Guest EditorReliability, irrespective of its definition, is no longer an "after-the-fact" concept. Today it must be an integral part of product development.
The product development process must provide a method for measuring customer expectations and preferences, competitive development and existing product performance. It must also respond to those results with continuous improvement.
This is specifically true in the microelectronics area, with its demands for miniaturization and system integration in a faster, better and less costly environment.
The rapid development and introduction of CSPs into the market is a good example of the trend toward miniaturization. The use of new materials, processes and new applications obscures the traditional definition of quality and reliability assurance.
Thermal cycling is the most commonly used method of characterizing devices and interconnections. Among the many predefined thermal cycling profiles, the military and commercial aspects represent the two extremes. Although the military standard (Mil-STD-883) was recently obsoleted, it is still used for benchmark testing.
Within Mil-STD-883, there are three levels of accelerated cycling temperatures:
Condition A -55/85°C
Condition B -55/125°C
Condition C -65/150°C
For benchmark conditions, devices are generally subjected to Condition C and assemblies most often tested to condition B. Assemblies were traditionally considered qualified when they survived 1,000 cycles. A commercial cycling profile, the J-12 IPC specification, recommends a thermal cycle in the range of 0° to 100° C. Within a temperature range, the dwell, heat and cool down rates are critical parameters and also affect cycles to failure.
NASA thermal cycling requirements are stringent and are specified in various revisions of NASA handbooks.
For example, in a previous revision, NHB 5300.4 (3A-1), there was a well defined requirement for the number of cycles and solder condition after exposure. No cracking of any solder joint was allowed after 200 NASA cycles (-55 to 100°C) with 245 minutes duration).
In a subsequent revision, the requirements were based on meeting the specific mission condition. The build and test methodology is expected to yield confidence in reliability to satisfy the mission conditions. Mission requirements are emphasized rather than a universal cycle and a value for all missions.
Testing to establish the confidence in reliability adopted by NASA, a long-time ago, is now the reliability theme for the commercial sector. Discussions on breaking traditional paradigms and rethinking of environmental reliability testing by authors from the commercial sector are becoming hot topics with the introduction of new, miniaturized CSPs.
These packages have their own unique form factor not seen in SMT. Unable to meet the stringent requirements established by the previous military standards, a new paradigm shift is considered to be the solution.
Additional unique tests have been adopted for specific consumer products. For portable electronics, bend test, drop test and possible "washing machine tests" are used or suggested.
IPC 9701, Qualification and Performance Test Methods for Surface Mount Solder Attachments, is aimed at some of these requirements.
We must recognize that no accelerated testing can be truly universal; field reliability is the ultimate test and either substantiates or invalidates the experimental tests. For space missions, gathering information on the root cause of field failure is almost impossible.
For commercial applications, rapid changes in technology render field information almost useless for new product development.
The only solution is to understand key reliability parameters and to design for reliability. Subsequent process controls, as well as efficient qualification and inspection, also help assure sufficient field reliability.
In other words, risk control and risk management must be practical.
Chip Scale Review o 7291 Coronado Drive, Suite 8 o San Jose, CA 95129 o Email: firstname.lastname@example.org
|© 1998 ChipScale REVIEW|