Hello Phil

This is Terry Munson of CSL and we have studied and published dozens of case
studies and articles on the issue of Reliability Effects of Process Residues
(which include incoming material cleanliness causing electromigration
failures of built assemblies).

To Answer your questions

        Questions for Technet:

                Does lead actually electromigration along with the tin as an
alloy?  I
                thought the metals separate during electromigration and the
tin would be
                the culprit.

        All metals will migration given the voltage and corrosive residues.
We have even documented Gold migration under high voltage and very corrosive
residues.  The metals that migrate the easiest, such as silver, copper, tin
will migrate faster and farther, than the lead, but lead will migrate also.

                This occurs on a potted board but only in one location which
happens to be
                 a high frequency switching application.  Why wouldn't it
happen all over
                 the board uniformly?

Yes, and no, because all three parts of the corrosion cell need to be present
to cause the failure.  The problem will show up first in the areas of
sensitive circuitry.  The corrosion cell requires three things for
electromigration to occur.  First a voltage differential (with as little as
1.5 volts), second a liquid media (monolayers thick) for the metal to move
through and third a corrosive element to cause the deplate of the anode.
Without one of the three requirements, no metal migration will occur.  And
since the products are potted outside humidity is rarely a factor then where
is the liquid phase of the corrosion cell coming from?  The bare boards are a
good source of corrosive residue (chloride, bromide, MSA, sulfate, Nitrate),
but where would the liquid come from in a potted system.  It could be from a
liquid flux used during rework or touch up that wasn't completely complexed
or activated (all the carrier driven off) below a component.  The failure
will be due to the sources of contamination as well as the moisture or liquid
levels over the board surface or below entrapment areas of flux or internal
moisture.  The sensitive switch will be the first area to fail.

                 Is 2 volts enough to cause the electromigration to occur?

Yes!!!! over 100 mil spacing

                 Our solderpaste has antimony (0.4%) and silver (0.6%) in it
as well as 63%
                 tin ( and the rest is lead).  Shouldn't the spectrum be
showing these
                 components as well or is anything under 1% not discernible?

The elements in the failure area that are <1% in the EDX analysis will be
difficult to see.  The elements that migrate are the materials that are in
the largest quantities.  The amount or metal content is not as important as
the things that you can't see by EDX.  How much chloride, bromide, sulfate,
MSA and Weak Organic Acids (WOA) are present between the failure points and
how uniform is that across the board surface.

                Is it possible for the bare board to be the culprit as it is
a HASL board?

Yes!!! but this is only part of the picture.  I also have a JPEG movie of
dendrites growing with 10 volts between a 50 mil spacing.  This is a real
time movie of shorting in less than five seconds.  I am working on a video
tape of various cleanliness levels and different voltage.


We have established our cleanliness levels using ion chromatography and SIR
testing.

Terry Munson
[log in to unmask]
765-457-8095

################################################################
TechNet E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c
################################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body:
To subscribe:   SUBSCRIBE TechNet <your full name>
To unsubscribe:   SIGNOFF TechNet
################################################################
Please visit IPC's web site (http://www.ipc.org) "On-Line Services" section for additional information.
For technical support contact Hugo Scaramuzza at [log in to unmask] or 847-509-9700 ext.312
################################################################

################################################################
TechNet E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c
################################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body:
To subscribe:   SUBSCRIBE TechNet <your full name>
To unsubscribe:   SIGNOFF TechNet 
################################################################
Please visit IPC's web site (http://www.ipc.org) "On-Line Services" section for additional information.
For technical support contact Hugo Scaramuzza at [log in to unmask] or 847-509-9700 ext.312
################################################################