kicad-developers team mailing list archive
-
kicad-developers team
-
Mailing list archive
-
Message #18455
Re: Schematic Symbol Philosophy?
On Fri, Jun 05, 2015 at 11:00:45AM -0700, Andy Peters wrote:
> At risk of reiterating myself again, what Chris describes above is how professional engineering groups do their libraries. Every company I’ve worked for has a vetted parts list and the PCB layout library includes only those parts. The symbols and footprints are married to a part number. Under no circumstance would someone choose, say, an NPN transistor from a library and then later match it to a footprint and to something that can be ordered. The chance of an expensive fuck-up happening is way too high.
I do this with eeschema aliases and pre-initializing the footprint
field. Since as I said packages are mostly standardized these days
I only need about 3-4 symbol 'duplicates' for each logical symbol (the
MO236-GDS fet, TO220-GDS fet, SC70-GDS fet and so on).
It's hugely important for some ICs where the part number itself changes
the package: while an SN74LVC00AD is a SOIC part, SN74LVC00ADBR is the
same part in SSOP (these are TI part number, I remember NXP uses another
coding...).
But, even more important, some parts actually *change pinout* when
switching packages; the awfully popular MSP430F5528 has 64 pins in QFN
(they call it RGC), 64 pin in the YFF BGA package (numbering changes
from numeric to coordinates) and even 80 pin in ZQE BGA package (a lot
of reserved balls:P). Other parts have some pin unbonded in the smaller
packages (something like you lose 2 GPIO but have a really smaller
body), and so on...
In these cases is essential to have different schematic parts, in
theory: MSP430F5528IRGC, MSP430F5528IYFF, MSP430F5528IZQE; these would
have the same pin names and drawing, but with different numbers *and*
different footprints associated (of course I only do them on an
as-needed basis, since BGAs are a PITA to handle anyway :D)
--
Lorenzo Marcantonio
Logos Srl
References