This Is AuburnElectronic Theses and Dissertations

A Study of the Vibrational Reliability Performance of Different Doped Low-Creep Lead-Free Solder Paste and Solder BGA Packages

Date

2015-07-31

Author

Stone, Derrick

Type of Degree

Master's Thesis

Department

Mechanical Engineering

Abstract

The growing use of smaller and more powerful electronic components in more varied and adverse environments, especially in regards to mobile consumer electronics, has created the need to better understand how recent lead-free solder replacements for the toxic eutectic tin-lead alloy are impacted by manufacturing variations and solder doping, especially in terms of their general reliability performance under vibration and thermal shock environments. This thesis focuses on the fatigue life characteristics of 15 mm CABGA packages with 208 perimeter solder balls on a 0.8 mm pitch which have been mounted on a vertically-oriented printed circuit board using various formulations and manufacturing parameters of lead-free solder balls and solder paste before being subjected to 4.6 Grms of random vibration using an LDS LV217 electro-dynamic shaker machine over the course of about 20-hour testing spans and with electrical tests being performed once every hour. The test vehicles were built to withstand JEDEC JESD22-B103B standards of high stress while the test alloy formulations involved 12 different lead-free pastes, including a SAC305 equivalent paste, and seven solder ball alloys that were manufactured using three reflow profiles (Low, Best, & High) and two stencil sizes (4 mil & 6 mil). Half of the formulations were tested immediately after they were manufactured, and the other half were placed inside a thermal aging chamber that maintained an internal temperature of 125oC for 6 months before being removed and subjected to vibration testing. A 2-parameter Weibull analysis was primarily used to characterize the ATTIF and MTTF of the samples in an attempt to discover any general trends affecting the vibrational reliability performance of the samples inherent to the variation of certain experimental parameters. In-depth analysis of the data revealed that two of the specific formulations, J6BN and J6L, achieved improvements in ATTIF of 1900% and 900%, respectively, over the performance of the equivalent test samples from the control group which used SAC105 and SAC305 solder balls with SAC305 paste. Even though the long-term performance of these two formulations was disappointing, their outstanding initial performance still earns them the greatest overall recommendation for short-term reliability. In terms of long-term reliability, however, the K6H test group emerged as the overall frontrunner. This test group was calculated to have an average 20% greater initial ATTIF than the equivalent control test group and was predicted to have a general failure rate that was either the same or better than the control group over the course of its life. This formulation could provide manufacturers with the ability to only employ a single SAC305 paste substitute and a slightly higher reflow temperature in order to dramatically increase the long-term reliability performance of the products constructed in their single or mixed-SAC BGA production facility as well as boosting the initial reliability performance for their SAC305 BGA components. If used as such, the K6H formulation would be most appropriate for use in less reliability-critical applications, namely consumer mobile electronics.