SMTA International Conference Proceedings

Thermal Cycling – It Doesn’t Have to be a Waste of Time and Money

Authors: P. Borgesen, J. Jiang, R. Sivasubramony, L. E. Alvarez, T. Alghoul, and C. Greene, L. Wentlent, M. Meilunas
Company: Department of Systems Science & Industrial Engineering at Binghamton University, Universal Instruments Corporation
Date Published: 9/17/2017   Conference: SMTA International

Abstract: Something can be learned from almost any kind of testing, but much too often it is not actually clear what it is and how that may be of practical use. Great amounts of time, effort and money are expended on accelerated thermal cycling testing across the microelectronics industry without realizing that the implicit assumptions underlying the interpretation of results may not be met. Thermal cycling is relatively simple in the sense that the ultimate life of SMT and area array assemblies is almost always limited by the failure of the second level interconnects due to the growth of cracks through the solder joints. Failure by pad cratering or cracks through intermetallic bonds or via structures can be ascribed to poor design or material selection, or to avoidable defects induced in manufacturing or test.

It is commonly recognized, albeit too often ignored anyway, that accelerated thermal cycling is only really meaningful to the extent that the results tell us something about the anticipated performance in service. Interpretations of results in terms of an actual life under the service conditions of concern rely on the assumption of an acceleration factor or the use of a quantitative model. An industry ‘consensus’ with respect to the sufficiency of the life for certain applications based on test results and field experience with previous products (‘1000 cycles are enough’) implicitly assumes that the acceleration factors are all the same. We present experimental results to show that common models and acceleration factors do not properly capture the variations in thermal cycling life times with individual parameters such as temperatures and dwell times. In fact, even relative comparisons between alternatives may not be as simple as expected (‘best in test’ does often not mean ‘best in service’). All of this may be explained and accounted for based on a comprehensive mechanistic understanding.

Key Words: 

Reliability; thermal cycling; model; solder

Members download articles for free:

Not a member yet?

What else do you get when you join SMTA? Read about all of the benefits that go along with membership.

Notice: Sharing of articles is restricted to just your immediate work group. Downloaded papers should not be stored on an external network or shared on the internet.


SMTA Headquarters
6600 City West Parkway, Suite 300
Eden Prairie, MN 55344 USA

Phone +1 952.920.7682
Fax +1 952.926.1819