TECHNET Archives

December 2005

TechNet@IPC.ORG

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
"Wenger, George M." <[log in to unmask]>
Reply To:
TechNet E-Mail Forum <[log in to unmask]>, Wenger, George M.
Date:
Fri, 16 Dec 2005 00:14:46 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (233 lines)
David,

The short answer to your question "Is it possible to get an acceptable
solder joint if a Pb-Free BGA is put onto a board using SnPb solder
paste and normal SnPb profiles?"  The longer answer depends on ones
definition of "acceptable".  If you had asked if mixed alloy solder
joints are reliable my answer would have been that nobody right now can
answer that question with any degree of certainty.  I know of a few
companies who have shipped product (either intentionally because they
didn't have an option or unintentionally because they didn't find out
until after they shipped it, that have area array components with SAC
alloy solder balls that were soldered using SnPb solder paste and a
typical SnPb reflow process and they are asking the same questions.
There are probably many other companies who have also shipped products
with mixed alloy solder joints and don't even know it yet.

There are many consortia activities currently planning or doing mixed
alloy reliability evaluations but my concern is that the data won't be
available early enough to avoid having to ship mixed alloy solder joint
product.  

I realize that there are many right now who have indicated that the
reliability of mixed alloy solder joints isn't "good".  However, I
believe that the published information I've seen so far doesn't support
their believes'.  The data appears to suggest that reflow solder joint
assembly quality and PCB surface finish have a big effect on the solder
joint integrity/reliability.  There are also others that are suggesting
that mixed alloy solder joints are or can be reliable and the only data
I've seen that tends to support that is from NPL.

At the present time I've been asked to give my risk assessment about
models that we've built with mixed alloy solder joints.   The problem
I'm having is the similar to the problems other companies are having.
Companies want answers now not later and they don't have the resources
to obtain the answers and they usually don't have the samples (i.e.,
products) to devote to a reliability evaluation.  So those who have to
do the risk assessment have to do it on limited data and the amount and
quality of the data usually defines the "comfort" level in the risk
assessment.

In the next week or two I'm hoping to finish up finishing up an
evaluation on a mixed alloy product.  I don't want to call it a
reliability evaluation because is doesn't follow IPC-9701 or for that
matter any recognized reliability methodology.  However, it is the only
data I can obtain in the required time period to do a risk assessment
and my attitude is any "Good" data is better than none.  I must
re-emphasize the "Good" data part because I've seen many cases where
some "Questionable" data has actually been worst than none.

Several months ago I was aware of a several models being built using a
16-Layer 0.100" thick FR370 PCB.  Four out of nine area array components
on these models had SAC alloy solder balls and all of the model boards
were built using our standard SnPb solder paste reflow assembly
soldering process.  After the assembly process I was asked will the
models be reliable.  I could hardly stop laughing until I realized this
question is being and will continue to be asked by many companies.  As
hard as I tried to obtain several model for temperature cycling the only
assembled PCB board I received was one defective sample.  This sample
wasn't defective because of a failed solder joint.  It was defective
because a design engineer manually modified circuitry on the outer
layers for design modification trails.  Since this wasn't a daisy
chained test vehicle and it was only a sample size of line as well as
the fact that there was no time, doing a controlled ATC made no sense so
we decided to do destructive physical analysis (DPA),i.e., cross
sections, to determine the degree of alloy mixing and visual quality of
the solder joints.  The DPA revealed alloy mixing of 40 +/- 10 percent,
minimum solder joint voiding, and "good" looking solder joints on the
mixed alloys area array solder joints as well as the SnPb solder joints.

Although I wasn't able to obtain an assembled PCB for temperature
cycling one board was subjected to HALT testing and when it was finished
with HALT testing it was used to develop a PEST (i.e., burn in)
procedure.  After HALT and PEST I was told I could have the board for
DPA so that I could give a more informed answer to the reliability
question.  Unfortunately, this time period came when my work load was at
it peak.  Fortunately, this was just before Thanksgiving and our
reliability organization had a temperature cycling chamber available so
I was able to add an addition 240 temperature cycles to the assembly
during the Holiday.  So far during the DPA what I see is extremely
encouraging.  If I can find time I hope to put the result together into
a paper and present the finding.  Most certainly it won't be entitled
"Everything You Ever Wanted or Needed to Know About Mixed Alloy Solder
Joint Reliability" but it might be something like "Mixed Alloy Solder
Joint Integrity/Reliability Sanity Check"


My apology for the long answer but I'm on a six hour flight and the
movie wasn't very interesting.





-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf of Guenter Grossmann
Sent: Thursday, December 15, 2005 10:38 AM
To: [log in to unmask]
Subject: [TN] Antw: [TN] Mixing SnPb and Pb free BGA's on an assembly

David

Lead free BGAs can be soldered with SnPb solder in a SnPb soldering
process. Even if you stay below the liquidus of SnAgCu the liquid solder
will solve the ball forming a lower melting alloy at the interface. The
process of the solution of the ball into the solder ends if
- The alloy that formes due to the continuous solution of Cu and Ag into
the SnPb has ist liquidus at the solder temperature
- The temperature is lowered below the liquidus of the SnAgCuPb alloy

This means that depending on temperature and time different fractions of
the balls are are solver in the solder. we made some trials and used the
temperature- time integral approach of James Vincent to characterise the
influence of the production parameters on the solder structure. 
Originally I was worried that the interface where the original ball
meets the solder will be the week point where degradation will take
place. However, looking at the slides of Dave Hillman I see that the
rupture takes place in the SnPb solder along the pad. This makes sense,
since SnAgCu it mor creep resistant than SnPb and thus the strain
occurring due to thermal cycling will be dominantly in the SnPb solder.
I have no quantitative data in hand to say how much crack growth is
accelerated due to this material combination.

A problem which is evident is the fact that you have no collapsing balls
(or not the entire balls) to compensate for coplanarity problems of the
balls. 

Best regards

Guenter





EMPA
Swiss Federal Laboratories  for Materials Testing and Research
Centre for Reliability
Guenter Grossmann,  Senior Engineer

8600 Duebendorf
Switzerland

Phone: xx41 1 823 4279
Fax :     xx41 1 823 4054
mail:     [log in to unmask]

>>> [log in to unmask] 13.12.2005 15:30:24 >>>
Is it possible to get an acceptable solder joint if a Pb free BGA is put
onto a board using SnPb solder paste and normal SnPb
profiles?
The lead free solder balls are SnAgCu 96.5/3/0.5.

Seems nigh impossible to get DDR2 packages with SnPb balls.

I'm designing a high end DSP into an industrial instrument. The DSP is
not available lead free, nor does the application call
for lead free (thankfully).
There is no option to use a different memory type with the DSP. Nor is
there any possibility of using a different DSP or FPGA
due to time/cost/sanity.
Don't want to explore going lead free on this one (for obvious reasons),
and certainly do not welcome the expense of reballing
0.8mm DDR2 chips.


Best Regards

David Greig
______________________________
GigaDyne Ltd
Buchan House
Carnegie Campus
Dunfermline KY11 8PL
United Kingdom
t: +44 (0)1383 624 975
www.gigadyne.co.uk <http://www.gigadyne.co.uk/>
______________________________


---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e
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 or (re-start) delivery of Technet send e-mail to
[log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to
[log in to unmask]: SET Technet Digest
Search the archives of previous posts at:
http://listserv.ipc.org/archives 
Please visit IPC web site
http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional
information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100
ext.2815
-----------------------------------------------------

---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e
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 or (re-start) delivery of Technet send e-mail to
[log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to
[log in to unmask]: SET Technet Digest
Search the archives of previous posts at:
http://listserv.ipc.org/archives
Please visit IPC web site
http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional
information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100
ext.2815
-----------------------------------------------------


------------------------------------------------------------------------------------------------
This message is for the designated recipient only and may
contain privileged, proprietary, or otherwise private information.  
If you have received it in error, please notify the sender
immediately and delete the original.  Any unauthorized use of
this email is prohibited.
------------------------------------------------------------------------------------------------
[mf2]

---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e
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 or (re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/archives
Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100 ext.2815
-----------------------------------------------------

ATOM RSS1 RSS2