Gary,
     Just to clarify things.  What you are saying is to connect the
components to the internal ground plane only.  Then to have your external
plane (call it a shield) connected to the internal ground plane periodically
around the outside.
     I agree that 99.9% of the time this would work great.  I also think
that it is possible that a component close to the edge could find a
preferential return path in the external plane.  If you had a separate
guard ring in the internal plane layer it would work 100% of the time
because there is no choice for that return path.  It HAS to be in the
internal plane.
     I have had many boards with mixed analog and digital circuits which
have required separated (read moated) ground planes.  In these cases simply
connecting the external shield to the planes would have violated the
isolation we were trying to achieve.  It would also have created a prefered
return path in the shield.
     I know that designs like this are in the minority but I think people
should understand that there are designs which may require a separated
guard ring.

Don Walker
[log in to unmask]

>    Don,
>
>    As long as the return paths are already taken care of, (not
>    unnecessarily breaking up the internal ground plane) having the
>    internal plane contiguous to the guard ring won't put the shield in
>    the return path as long as they are only tied together at the board
>    edge. (At least that's what the instructor said) It's all BLACK MAGIC
>    anyway.
>
>    Gary Fuller
>    [log in to unmask]

***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************