Subject: | |
From: | "Jeff Seeger" <simon.ipc.org!bort.mv.net!rapidcad!jseeger> |
Date: | Tue, 6 Feb 96 13:48:11 EDT |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
A Design perspective on non-functional pads:
Submitting fab data to "any" fab house on behalf of a large
design customer base, we had a tug-of-war from fab house to
fab house regarding these pads, sometimes on the same product
being multiply-sourced.
Sometimes the customer would have an opinion, too.
Customers have told us to remove the pads and use the routing
space, usually on a .050x.000 pitch PTH connector. They don't
have to worry about primary drill @ DTP, compensation/registration,
or those other nasty little details. So we've been heading this
argument off for a while.
We've been asked to remove them on outer layers, also; this
caused cleaning problems in fab and assembly processes.
Experience showed that non-functional-pads were alot easier
to remove than to put back. An ad-hoc survey of CAM software
showed that most systems could remove them in a few keystrokes.
Our answer evolved to leaving the non-functional cu in place and
allowing addition/removal at the fabrication house's discretion.
They are the entities responsible for producing non-warped PWB's
that survive normal thermal cycling, without shorts or poor yield.
We can only try to help foster these goals by not boxing them in.
<applying good design practice helps *a little*> Only one fab-
rication house has insisted the pads be removed.
We also found use in keeping the non-functionals in place for drc,
both normal platform drc (some CAD systems don't look at details
like drill size) and our own post-gerber checks.
Good luck!
Jeff Seeger Applied CAD Knowledge Inc
Chief Technical Offficer Tyngsboro, MA 01879
[log in to unmask] 508 649 9800
|
|
|