Dear May William:
We have seen similar failures and came to the conclusion that this is
CAF-growth (conductive anodic filament), we received many helpful comments
from technet. I would like to get more information about your materials, to
see if we have a problem in common and maybe can help each other; please
contact me offline under [log in to unmask], as we are not yet far enough to
go to public: We don't want to mess up the so far good contact between board
manufacturer and assembly house, who try together to solve this problem - it
has probably a lot to do with contamination issues.

If you want to test for derating, try and use a multilayer test board with
inner layer comb structures, like for SIR-testing. This will enable you to
test for threshold values - IF THEY EXIST (!). For more details, as above,
get in touch.

Schön' Gruss, best regards,

Thomas Ahrens, Memellandstr. 8, D-24598 Boostedt

-----Ursprüngliche Nachricht-----
Von: May William D (Dean) CNIN <[log in to unmask]>
An: [log in to unmask] <[log in to unmask]>
Datum: Montag, 15. Mai 2000 15:42
Betreff: [TN] (FAB) Shorts Through Thin Epoxy Dielectric at 150V


>Interlayer shorts developed in modules subjected to elevated humidity,
>temperature, voltage testing.  The shorts go through a 2 - 2.5 mils thick
>epoxy glass dielectric (2 plies 1080, expectations were that the dielectric
>would be closer to 4 mils thick).  Test conditions were 150V (DC), 150*F,
>50%RH.  Time to failure varied from 130 to 1011 hours.  Current was limited
>to 1mA.  Failure was defined as (leakage) current above 0.5mA.  The shorts
>go nearly straight through the epoxy (not tree-like dendrites and do not
>pass-through glass bundles), usually from a pad on one internal layer to
the
>solid plane on the adjacent internal layer.  Voids were observed in the
>dielectric layer, but none were noted at the failure sites.  SEM analysis
>did not identify any contamination.
>
>Questions
>1.  Are these "real" dielectric breakdown failures or have we induced some
>odd (irrelevant) failure mode?
>
>2.  Search of TechNet archives uncovered a 1998 (R Hersey) post stating
that
>dielectric strength should be derated to 50 V/mil for long-term reliability
>of "thin" dielectrics.  Our results seem to support this post.  Is there
>other data available to support (or refute) the 50 V/mil derating?
>
>Thanks,
>Wm. Dean May
>NSWC Crane
>Bldg. 38, Code 8086
>300 Highway 361
>Crane, IN 47522
>812-854-3073
>
>##############################################################
>TechNet Mail List 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://www.ipc.org/html/forum.htm) for
additional
>information.
>If you need assistance - contact Keach Sasamori at [log in to unmask] or
>847-509-9700 ext.5315
>##############################################################

##############################################################
TechNet Mail List 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://www.ipc.org/html/forum.htm) for additional
information.
If you need assistance - contact Keach Sasamori at [log in to unmask] or
847-509-9700 ext.5315
##############################################################