Ah Dave, so glad you asked that question.

Started doing work with plastic BGA's in the beautiful summer of 95. Did
some of the first IBM ceramic types about a year earlier. Had much fun with
both.

At Amdahl, and another not mentionable company, had the best of everything
except tools to place BGA's. This did not stop us from doing some serious,
at the time, soldering experiments using both home made table top diffused
IR cookers and the newest model, at the time, convection IR ovens.

Solder paste was a bit different then and we developed some for internal use
as required. My first thoughts were using micro vias, of a type I developed
in the early 80,s, in Kapton as a compliant layer on top of the rigid MLB
substrate. Used paste first but didn't like the voiding found using
transmissive x-ray equipment. Attributed most of it to outgassing from the
vias. This, as most know, wasn't the only contributing factor. Reliability
issues arose so I did some HALT/HAST testing to determine cycles to failure.
Mostly turned out to be no big deal but created an internal specification
limiting acceptable voiding to 20% maximum.

Did other work without solder paste. Instead, used flux paste and
determined, that with eutectic balls, no apparent voiding. Liked the idea
for no particularly good reason. Re visited the HALT/HAST testing and found
no significant difference between soldering with or without solder paste.
This meant finding the difference in solder volume was not a factor either.
However, coplanarity raised its ugly head occassionaly so I went back to
solder paste as a preferred process to overcome the issue. Later, with
ceramic device types becoming available, I went back to flux only and most
in the unmentionable company liked the results.

The big problem, of course, you can't use flux only in production for
obvious reasons. I believe Phil Zarrow had some comments and similar
thoughts on this issue as well. However, during rework operations, I never
used solder paste when the option is available. One advantage here is seldom
ever do shorting problems occur with no added solder paste and, again, even
with less solder volume in the joints, no negative reliability issues are
apparent.

Some other companies I've worked with (HP, Celestica, and others) also have
found this to be true. One example at HP concerned their inability to effect
good solder joints, on some perimeter BGA's (super BGA's) on a particularl
high density MLB. I introduced my idea, though they had been kicking it
around for a time, both in production and rework operations. Problems were
solved and no reliability issues or field return problems were encountered.


There's so much more, but as Phil Z said they ain't no big deal. Having said
that all, I'm now getting involved in some aerospace applications, again,
requiring COTS parts as plastic BGA's. The wheel just keeps on turning as
what goes around - yada yada. Who's willing to share their high rel plastic
stories? Anything new and exciting but for coplanarity, moisture
sensitivity, or?

Earl Moon

---------------------------------------------------------------------------------
Technet 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 Technet
To temporarily halt delivery of Technet send the following message: SET Technet NOMAIL
Search previous postings at: www.ipc.org > On-Line Resources & Databases > E-mail 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
---------------------------------------------------------------------------------