Dave N6NZ wrote: >DJ Delorie wrote: > > >>>Imagine, that you have a 431 pin BGA. Would you include 431 times >>>the same padstack in the footprint? I think one should bother with >>>the whatever shape, and size of the stencil, copper, mask, paste >>>layers. Those are just "pads". Then we could link pads to a >>>padstack, and the padstacks into the footprint. >>> >>> And footprints into the layout. It's a hierarchy with everything instantiated rather than embedded. This is the way many (most?) large commercial tools work. The layout editors I used 25 years ago were built this way. I realize it is a significant shift from the paradigm and the format of PCB's current data, but it is the direction we should be moving. And I would also second the earlier comment regarding the need for a more general "layer" definition to facilitate paste, clearing of solder resist ....
>>For sanity's sake, let's let a footprint define its pad stacks for its >>own pads, rather than trying to maintain a global cache of padstacks. >> >>Then, a BGA footprint would have one "here's what my pads look like" >>and 431 copies of "put one here". >> >> > >Yup. Right answer. > >A footprint needs to be entirely self-contained. > > I can't agree. There is no reason to have a pad stack (and drill and keepout and paste and...) defined in every footprint that uses a certain size pad or pin definition. How many places do we need to have copies of a pad suitable for a .5mm pitch QFP when every such QFP needs the same pad geometry? One reason to have the pad stacks separately defined and simply instantiated in the footprint is to account for a process or fab vendor change. So you want to convert your "library" to RoHS-compliant soldering process and the recommended paste patterns need to change. I'd much rather edit a pad stack definition once than go through every footprint that uses it to make the change. Joe T. >-dave > > > >_______________________________________________ >geda-user mailing list >[email protected] >http://www.seul.org/cgi-bin/mailman/listinfo/geda-user > > > _______________________________________________ geda-user mailing list [email protected] http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
