Mark Rages wrote: > On Tue, Nov 4, 2008 at 7:42 PM, Joerg <[EMAIL PROTECTED]> wrote: >> Sorry for asking a lot here lately but I believe others will hit this >> snag as well and I could not find much info about it, not even via a web >> search outside geda.seul.org: >> >> Often power pins must be made visible. For example if you have to filter >> the supply for digital parts or opamps separately for each instance. Or >> if you use a 74HC14 at 5V and another 74HC14 on the same schematic at 3.3V. >> >> Those power pins show up on every single slot. Looks quite ghastly on a >> schematic. So the LM358 dual-opamp was to be my first part made with >> gEDA. I tried this: >> >> Slotdef=2: 5,6,7 >> Slotdef=1: 3,2,8,4,1 >> >> That butchered it. Connected both outputs to pin 1 and assigned pin 7 to >> the supply (which should have been the 2nd output). Somehow I'd have to >> have power pins on slot 1 but none on all the others. >> >> Is there a somewhat orthodox way to do this? A method where renumbering >> at a later point doesn't break refdeses? >> > > I don't know about renumbering, but the standard way to handle this is > to use two different symbols with the same refdes. You could use one > symbol for each slot in a dual part, with one having power pins > included. Or as I do, have a two-pin symbol for the power pins, then > two instantiations of the LM358 op-amp symbol for the two op-amps. As > long as all three have the same reference designator, the netlister > will do the right thing. >
Yes, I thought about that, having generic opamp blocks and scooting a pair of "floating" power pins over the first instantiation of a two- or four-pack. That is how it's done in Eagle if you don't want to create a new symbol (make a part with implicit power explicit). But Eagle keeps all that together because it formally belongs to the same part. In gschem you'd have to design a separate power pair for pretty much every part that has more than one slot. Can be done but kind of messy. A renumbering script would probably butcher that up. -- Regards, Joerg http://www.analogconsultants.com/ _______________________________________________ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user