By Jasbir Bath, IPC principal engineer
Over the last few years, there have been many discussions on the affect of voiding on BGA/CSP solder joint reliability. IPC-A-610 and J-STD-001 standards state that greater than 25% of voiding in any ball in the X-ray image area is classified as a defect according to Electronic Product Classifications 1, 2 and 3.
The challenge that has faced the industry is what data is available to support the greater than 25% voiding defect classification and how early will a failure occur when a BGA or CSP component solder joint has a void greater than 25% versus a solder joint which does not.
There have been many studies in the industry to try and understand how voids can affect solder joint reliability. IPC SPVC (Solder Products Value Council) conducted a study which found that there did not appear to be a correlation between void size and number of cycles to failure when comparing voiding levels below and above 25% of the X-ray image area. In a separate study (Ref. 1) solder joint reliability was reduced only when the voids were located in the crack path which reduced the overall path length to cause solder joint failure. The above studies were done with relatively mild thermal cycling temperature ranges from 0C to 100C. When investigating with wider temperature cycling ranges from -55C to +125C with larger stresses on the solder joint (Ref. 2), voids greater than 25% failed earlier than solder joints with less than 25% primarily due to the location of the voids in the crack path. It was also found that the failure cycle of void containing solder joints was only slightly earlier than the non-void containing solder joints.
A more recent study comparing voiding % at less than 25% and at 42% indicated the increased voiding % did give a reduced solder joint reliability when thermal cycling from -40C to +100C (Ref.3). It also indicated a group of smaller voids may be more of a concern than a single large void. The study highlighted the need for more work to determine acceptable levels of voiding from both a thermal and mechanical reliability point of view. This information would be used in developing standards going forward for areas of the BGA/CSP components where there are increased stresses such as the outer row and corners of the component as well as die shadow areas.
In many of the investigations conducted to date, it has been a challenge to produce large voids to understand the affect on solder joint reliability based on the improved solder paste materials and optimized reflow profiles used in the industry, which indicates that the voiding criteria could be used to establish soldering process controls rather than verifying component solder joint integrity. Changing the current IPC criteria for BGA voiding to be a process indicator rather than a reject criteria is an area which will be discussed at the IPC J-STD-001 and IPC-A-610 standard meetings and technical conference at IPC APEX next year in February. It should be a lively discussion.
Ref. 1: S. Sethuraman et al., The Effect of Process Voiding on BGA Solder Joint Fatigue Life Measured in Accelerated Thermal Cycling, SMTAI conf. (2007).
Ref. 2: D.Hillman et al., The Last Will and Testament of the BGA Void, SMTAI conf. (2011).
Ref. 3: G. Qin et al. Assessing the Impact of Temperature Cycling Reliability of High Levels of Voiding in BGA Solder Joints, SMTAI conf. (2012).