BGA Component Grounding Problem

We have an ongoing concern regarding reliability BGA components. Over the last year, we had boards returned due to a BGA issue. The boards use leaded balls, and leaded solder. All the boards show a resistance to ground at a data point of 3K ohms. A look at "good cards" shows a resistance of 3M ohms. X-rayed extensively and no issue found.

We remove the BGA from the card, and now both the BGA as well as the board show the proper 3M ohms. Each time, after the removal of the BGA, the data point path to ground problem goes away. Essentially we somehow lost the failure signature. Can you offer guidance?

M.S.

Expert Panel Responses

Have you cross sectioned the BGA looking for head in pillow HIP? There are a lot of issues that X-RAY will not show.

Terry Munson
President/Senior Technical Consultant
Foresite

Mr. Munson, President and Founder of Foresite, has extensive electronics industry experience applying Ion Chromatography analytical techniques to a wide spectrum of manufacturing applications.

his maybe useful especially if you are using 0.004" thick stencils. I have seen (last six months) many issues of failures and when investigated there is nothing to see under Xray analysis of a few manufacturers BGA's. We found another reflow again with a low solids Rosin flux liberally applied under neath the reflowed device or Rosin flux paste all around the BGA fixed the issue.

I'm pretty sure of the devices I have looked at the BGA balls on a number of pads are miss sized and when using a 0.005" Stencil it's enough to accommodate the miss size whereas the 0.004" is definitely not OK. If you have to use 0.004" Stencils then over print the Paste and allow it to pull onto the pads again making up the gap produced by the smaller balls. Hope it works

Greg York
Technical Sales Manager
BLT Circuit Services Ltd

Greg York has over thirty two years of service in Electronics industry. York has installed over 600 Lead Free Lines in Europe with Solder and flux systems as well as Technical Support on SMT lines and trouble shooting.

If the cause of your high resistance failures is not apparent using microscopy or X-Ray, you will probably need to perform dye & pry and/or cross-section analysis of the interconnects.

Rick Kompelien
Principal Product Engineer
Benchmark Electronics, Inc.

30+ years of experience working with electronic and electro-mechanical manufacturing and design (medical, automotive, military, computer, and industrial controls). Military veteran - served as a Combat Engineer with the United States Marine Corps.

Perform an Dye-n-Pry test on BGAs. It should identify any cracks or separations within the solder joints that could lead to false results. IPC-TM-650 Method 2.4.53 specifies a process for dye-n-pry.

Swaroop Pawar
PCBA Industrialization
Schneider Electric

Have 18 years of experience in electronic Industrialization. Specialties in PCB Design & manufacturing process, PCBA Process Development and Continuous Improvement.

It sounds like what is happening is dendritic growth. Over time, in the presence of moisture and ionic contaminants, a conductive path can form between conductors that are at different potentials. The 3k-ohm resistance is still relatively high for a dendrite, so you probably have very limited dendritic growth.

When you remove the part, you mechanically disturb the microscopic dendrites, and unless significant corrosion has happened, the failure site is then difficult or impossible to find. The "good" resistance of 3M-ohms actually seems very low as well. Normally, I'd expect a resistance above 1 G-ohm.

A resistance of 3M-ohm to ground will result in a small current always passing through that path, which can drive transport of metal ions, and eventually lead to a drop in resistance and failure of the circuit. Some of the most common contributors to this type of failure are:

Ionic contamination of the raw PWB
Incompletely reacted no-clean flux residues
No-clean flux residues that are inappropriate for the intended use environment
Too-small clearances between exposed conductors with continuous DC potential differences, e.g. 5V+ or 12V+ and GND

Fritz Byle
Process Engineer
Astronautics

Fritz's career in electronics manufacturing has included diverse engineering roles including PWB fabrication, thick film print & fire, SMT and wave/selective solder process engineering, and electronics materials development and marketing. Fritz's educational background is in mechanical engineering with an emphasis on materials science. Design of Experiments (DoE) techniques have been an area of independent study. Fritz has published over a dozen papers at various industry conferences.

If you are having issues at only a single data point and the issue is lower than expected resistance to ground, it may be that there are some flux residues allowing a secondary voltage/current path to ground, effectively reducing the resistance reading. If you are using an RMA or No-clean solder paste or flux, then I would even bet money on it. One way to check this is to simply remove the failing BGA and inspect the area for flux residues.

If none are seen, then perform site prep by applying flux and wicking the solder off of the BGA pads and clean the area thoroughly but do NOT replace and reflow the new BGA. Instead, simply place a new BGA onto the pads and press it down lightly and have someone measure the resistance at the data point. If you now have the required 3 Megohms, repeat the process with perhaps 10 more BGAs using that same assembly as a "test fixture", so to speak.

If all 11 BGAs display 3 Megohms, then lightly flux the pads with a good water-soluble tacky flux, solder one of the known-good BGAs onto the assembly, clean it thoroughly followed by an immediate blowoff to remove as much water as possible, bake it dry at 105 deg. C for 20 minutes, and test again.

If it tests good, do the same thing on 10 more assemblies with the 10 remaining known-good BGAs. If they all pass, you can safely assume the issue was flux residues allowing the voltage to leak to ground. If any should fail, then there is some other issue with the BGA caused by reflow.

Richard D. Stadem
Advanced Engineer/Scientist
General Dynamics

Richard D. Stadem is an advanced engineer/scientist for General Dynamics and is also a consulting engineer for other companies. He has 38 years of engineering experience having worked for Honeywell, ADC, Pemstar (now Benchmark), Analog Technologies, and General Dynamics.

Submit A Comment

Comments are reviewed prior to posting. You must include your full name to have your comments posted. We will not post your email address.

Free Newsletter Subscription
Circuitnet is built for professionals who bear the responsibility of looking ahead, imagining the future, and preparing for it.

Insert Your Email Address