Reading time ( words)
I’ll always remember the summer of 2004 as the “Summer of Lead-Free.” Finally, Pb-free circuit boards were going into standard production mode. Assemblers focused the majority of their efforts (often at my behest) on final finish and proper laminate selection. What none of us saw coming, however, was the rash of delamination that would burn the entire industry during that long, hot lead-free summer.
After a good amount of research, Isola came up with a Lead-Free PCB Fabrication and Assembly Guideline that outlined various steps in the PCB fabrication and assembly processes that are critical to successful lead-free PCB assemblies. The most prominent process step that was added in this guideline is baking—during fabrication and, most critically, just prior to assembly. The goal of baking is simply to drive moisture out of the PCB.
Moisture plays a critical role in lead-free PCB assembly. Therefore, it is important to discern the specifications of your base laminate. One knee-jerk reaction of North American PCB users was to specify what they believed to be the highest end laminates, those that qualify under IPC 4101/126 and /129 slash sheets. These are typically phenolic-cured materials, which have much higher thermal properties in terms of glass transition temperature (Tg) and decomposition temperature (Td) that are required to survive lead-free assembly without via failure.
Unfortunately, other properties of the laminate and assembly temperatures create a perfect storm for reliability failures. These include: moisture absorption, interlaminate adhesion strength, and water vapor pressure at lead-free assembly temperatures.
Moisture absorption of phenolic materials is more than 2x as compared to traditional FR-4 materials that qualify under IPC 4101/21. In the early days, the technical data sheets used a 0.028" thickness core material for moisture absorption measurement. Based on this test vehicle, phenolic laminates showed moisture absorption of 0.45%; comparatively standard FR-4 laminates scored a 0.20% rate of moisture absorption. Over time, I think someone got wise and changed the test vehicle to a 0.059" thickness core, thereby increasing the denominator of the weight loss calculation. This resulted in a moisture absorption rate of 0.20%, matching the rate of standard FR-4 laminate.
To read this entire article, which appeared in the November 2016 issue of SMT Magazine, click here.