Steve:

Since you're referring to internal shorts, I'm assuming that the short is occurring
between two features on the same layer.  I recommend that you perform an examination of
the board artwork.  FR-4 has the ability to withstand (I think) about 500 volts per mil
(0.001 inch) perpendicular to laminations, i.e. between layers.  It has significantly
lower withstanding capability parallel to laminations (on the same layer) of 45 volts
per mil.

I have seen perfectly good video power supply boards fail after relatively short field
service because of carbon filament growth creating an internal short.  The short grew
from the barrel of a plated hole to a ground plane.  We modified the artwork by
increasing the ground plane clearance around the barrel of that hole and the field
failures ceased.   Note that the 45 volt per mil figure is a laminate specification and
an appropriate safety factor should be added to allow for manufacturing tolerances,
etc.

Hope you found this useful,

Don Vischulis

"Stephen R. Gregory" wrote:

> Hi ya'll!
>
> Hoo-boy! We be getting some weather today! They said that remnants of
> hurricane "Greg" would be going through our area this weekend, and it's
> comin' down like cats and dawgs now!
>
> Anyway, we just an assembly back from one of our customers that has a short
> internally in the fab, from a power connection on a terminal block that's
> installed into it, to the ground plane of the fab. It's a board that handles
> some high power. They said that the board powered-up fine, but over time the
> short developed. They also said that they suspect that another place will
> develop a similar short from some readings that they were getting when they
> began to trouble-shoot the first short. There's 4, 6-pin terminal blocks on
> this board.
>
> They sent us back the board and suggested that we give the assembly back to
> the fab vendor to look at it and find out what the problem is. There are
> those here of the opinion here that we should find a third-party somewhere
> and let them take a look at the problem, because they will be more objective
> in the diagnosis...if you catch my drift. Problem is, I don't know where I
> should start to look for somebody like that, do any of you have any
> recommendations?
>
> I'll admit that I don't know much about what can cause something like that.
> First thing that came to my mind, was some sort of electromigration happening
> internally to the fab...since it happened over time (I don't know the actual
> time frame), could that be it? Could it be caused somehow by the way we
> assembled it?
>
> Thanks for any suggestions!
>
> -Steve Gregory-

##############################################################
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 Gayatri Sardeshpande at [log in to unmask] or
847-509-9700 ext.5365
##############################################################