Ken,

     I've run across a discussion about a somewhat similar problem on another
INTERNET site (www.smtnet.com/smt_forum/smt_mail.html).

     The discussion centered around failed assemblies that was pointing to a
problem with the epoxy absorbing moisture from the air. The assemblies were
discovered failing at ICT, functional test, and out in the field. I don't know
what type of epoxy you're using, (and I'm just relaying what I had read in the
discussion) they were using Loctite 3609.

     They did some failure analysis on the caps that were failing (yes, they
had caps failing too...) and found when they removed the failed cap, there was
a liquid beneath the cap in the epoxy. They later discovered that it was
mostly flux that was trapped in voids in the epoxy.

      The way that it was explained is that Loctite 3609 is a hygroscopic
material. You can see this by doing what was called a glass slide test. Place
a dot of fresh Loctite 3609 that was just opened on a glass slide and put
another glass slide over top of it. Do the same thing with another set of
glass slides but with 3609 that's been sitting out of it's container on your
stencil for an hour or so. Run the two slides through your reflow oven to cure
at the same time. After they come out of the oven, look at the slides under
the microscope. You will see more voids on the slide with the adhesive exposed
to the air. The void are being caused by the outgassing of moisture that the
material has picked up when sitting out. This same voiding can be happening on
the adhesive that you're stenciling. The voids will allow flux to reach the
underside of the component.

     An aqueous clean will not remove this junk and you have flux entrapment.
Most water soluble fluxes are very active, and the cap will have an
electrolytic failure in no time.

     Supposedly this very thing really happened to a large contract
manufacturer who built a product for a major networking company. They built
the assemblies with the typical double sided process using epoxy. They also
had voids in the epoxy filled with water soluble flux. The end product went to
a MAJOR shipping company. This particular product ran the MIS system for
scheduling and shipping. It crashed three days before christmas a few years
ago. All their backup systems were useless and they were dead in the water.

      I don't know for sure if this is what is happening with you, but this
discussion really caught my attention...this is the first time I've ever heard
anything like that. The problems that you seem to be having could be related,
you never know.

      The credit for this information goes to Justin Medernach who shared this
in the forum I listed earlier. I'm just passing this along...

                                                                   I hope this
helps,

                                                                     -Steve
Gregory-

################################################################
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 web site (http://jefry.ipc.org/forum.htm) for additional information.
For the technical support contact Dmitriy Sklyar at [log in to unmask] or 847-509-9700 ext.311
################################################################