On Tue, Sep 14, 2010 at 10:34:23PM -0500, John Griessen wrote: > On 09/13/2010 05:07 PM, John Doty wrote: > > > >On Sep 13, 2010, at 2:27 PM, DJ Delorie wrote: > So let me rephrase: Why have seven geometric holes, > >>one for each layer, when we can have one geometric hole applied to the > >>whole composite? > <snip> > >My notion is that you need a general mechanism to align objects between > >layers. > <snip> > Having such a mechanism, you should then use it universally. If > >you bypass general general mechanisms for special cases, you risk creating a > >mess > where general automation procedures fail in > >those special cases. > > John D has something here -- suppose we start using pcb to model the physical > in > a way that is used to analyze layer stack up alignment of weirdly processed > layers? > For instance layers that get folded and stretch some, and so need alignment > checking? > > Perhaps holes get made in one stroke still, but the via pads need a tolerance > to be able > to hit them close enough to their centers with the drill. Perhaps holes > don't even all > get made in one stroke! I heard about microvias being made with UV lasers > that would be on just a few layers. > Also there is a max feasible aspect ratio to a via hole width vs. its depth > to drill with carbide bits. > Some board makers said they handle this by drilling in from both sides. Take > it a little further > and you make two composites separately drilled, then bond together.... > > Sure, all of this is far out, but still worth planning for. >
I -think- that all these cases are accounted for with: 1. DJ's composite idea 2. Distinct canvas/layer structure with tagging 3. Custom vias At the very least, we're on the right track. For custom vias, we will probably want to allow drills in footprints. Andrew _______________________________________________ geda-user mailing list [email protected] http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
