Vladimir Igoshev posted an inquiry about the "so called 'Brittle Fracture'
phenomenon, which sometime can be observed in solder joints between BGA
components and ENIG finished boards." Here's a paragraph from a recent
Circuitree article by Don Cullen (November 2002,
http://www.circuitree.com/CDA/ArticleInformation/coverstory/BNPCoverStoryIte
m/0,2135,86472,00.html):
"The failure modes associated with ENIG are numerous and frequent. Solder
mask lifting was the initial showstopper. Overplating can result in shorts,
impedance problems, and pieces of metal drifting around the assembly. Hey,
just stabilize the nickel and catalysis to decrease overplating, right? Skip
plating. The interaction complexity of catalysis, rinsing, nickel stability
and gold adhesion could fill volumes. The only answer is process control.
True, fabricators can ensure a good functional ENIG coating with painstaking
process control. What the industry has not been able to ensure is the
eradication of the mother of all PCB failure modes: black-line nickel. This
insidious corrosion of weak nickel by aggressive gold results in BGAs
popping off high-density, expensive boards-sometimes after they are in
service. One final ENIG problem worthy of mention: studies have demonstrated
that solder joints formed with nickel are not as tolerant of physical shock.
For this reason, manufacturers of devices that are frequently dropped, such
as mobile phones and PDAs, will prefer to solder to OSP, HASL, silver or
tin."
I have just discovered that ENIG is also used as the pad finish for PBGAs,
so the risks are there for both connections - component to solder and solder
to board. I spoke to a number of people at the recent IPC meeting in New
Orleans who are recommending not using ENIG as a finish. They are
recommending immersion silver. See in the Proceedings S-02-2, "What the EMS
Provider wants in a Board Finish", by Bruce Houghton, and S-02-3, "SURFACE
FINISHES - An OEM Perspective" by Mike Barbetta.
You may also want to look at the article by Robert Rowland, "My BGA Learning
Curve", SMT, October 2002. The article may be seen (after free registration)
at
http://smt.pennnet.com/Articles/Article_Display.cfm?Section=Archives&Subsect
ion=Display&ARTICLE_ID=157960.
Specifying the finish you want for a board is fairly straightforward, but
how do you specify the finish for a BGA? I doubt that most component
manufacturers even tell you what finish they are using.
If you push on a BGA solder sphere, it takes a good deal of force to push it
off, but when it does go, it is by interfacial fracture through the
nickel-tin intermetallic compound. In some areas, it leaves bare nickel. I
found this out while trying to determine the cause of missing solder spheres
on some BGAs we received. In this case, the problem seems to be not the
finish but rather the manufacturer dislodging of the spheres with test
tooling. This was also found by Mr. Rowland (private communication). Don't
count on automated optical inspection to discover missing solder spheres
unless you have verified that your equipment can do it with high
repeatability.
Gordon Davy
Baltimore, MD
[log in to unmask]
410-993-7399
-------------------------------------------------------------------------------Leadfee Mail List provided as a free service by IPC using LISTSERV 1.8d
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Leadfree
To temporarily stop delivery of Leadree for vacation breaks send: SET Leadfree NOMAIL
Search previous postings at: http://listserv.ipc.org/archives
Please visit IPC web site http://www.ipc.org/html/forum.htm for additional
information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700 ext.5315
-------------------------------------------------------------------------------
|
