Hi Steve,
you`re describing interesting phenomena. Unfortunally I don`t have a
global answer but one (perhaps weird) idea relating to your first
question, how migration can take place in the silver epoxy. Let try me
to explain my thoughts:
At first, conduction in silver filled epoxies is in principal done by
random contacts between the randomly placed silver particles. Lets
assume there are two adjacent conducting paths formed by particles
contacting each another, but not particles of the other path (the two
pathes are isolated from each other). One path is at +28 V, the other
one (probably short and located at the edge of the conductive glue) is
isolated. Now lets assume migration starts and dendrites grow from the
-5.5 V side towards the +28V side. Where will they reach the +28V side?
Perhaps by chance at first at the isolated path. This path then drops
down to -5.5V and you`ve got two very close conducting paths in the
epoxy with a potential difference of 33.5 V. Now one has: silver, high
field strength, some ionic impurities (which are resulting from the
chemicals used in the manufacturing of the epoxy), obviously moisture -
migration starts.
As I understand from your mail, you got a chip land covered with silver
filled die attach epoxy and some bondpads very close (146 micrometers
seems to be unusual close to me). The epoxy is sometimes squeezed out
over the chip land, thus reducing the distance furthermore. The first
idea to reduce migration issues is to get bigger distances.
If, as you wrote, migration under identical build up conditions in some
cases took place and in others didn´t, I would look carefully on
moisture issues which will be probably different in these cases.
By the way, if you can post pictures to your ftp-side, you should also
be able to include them in emails. But to not upset some technetters by
getting megabyte big mails, if you like, you could send them directly to
me. I would be interested in receiving them.
Greetings,
Ulrich Korndörfer
##############################################################
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.
For the technical support contact Dmitriy Sklyar at [log in to unmask] or 847-509-9700 ext.311
##############################################################
|
