Mail Archives: geda-user/2016/03/08/09:16:52
--089e0149d0747074d6052d8a39a5
Content-Type: text/plain; charset=UTF-8
>
> ... and workflow nightmares :D (no, really, I had to redo a schematic
> because I needed some footprints, found Erich's kicad converted stuff on
> gedasymbols, and then I thought "let's get all of gedasymbols and add it
> to the component library, what could possibly go wrong?")
>
> some curation will be needed somewhere, and things will still need
> manual tweaking to integrate nicely, because conventions may differ: for
> instance, those converted nixie footprints have alphanumeric pin names
> (A, LHDP, RHDP), while the convention I was aware of is pin name == pin
> number (so my symbol's pins didn't match up with footprint pins);
> however, changing the footprint pin names was easy, I very much prefer
> this to drawing the thing from scratch :)
>
> another example of clashing conventions is an Arduino nano symbol (I
> think also from Erich), which has pins very tightly packed and
> everything aligned at 25 grid units, not 100 like regular symbols.
>
>
I have been reluctant to put much more onto gedasymbols just yet because of
the need for curation, and the very arduino symbols you refer to prompted
my enforced pins spacing code which can enforce a specified spacing (when I
nearly lost my mind trying to wire up an arduino nano in gschem with
oddball pin spacing).
In doing this, though, other silk elements that prettify the symbol may be
ruined.
My impression is that the worst offenders are symbols that have been
converted to kicad with an eagle ulp script - they seem to have some very
odd pin naming conventions.
My eagle -> kicad code tries to be more sensible, and applies the
(multiple!) pin mappings provided in the eagle deviceset (symbol flavour ->
particular footprint) to create distinct symbol versions with an associated
footprint also converted from Eagle.
As you say though, the hard work is the geometry, and pin mappings often
need to be verified before sending a board off for production.
I will try to flag oddball spacings and prompt the user to enforce a
preferred pin spacing.
Thanks for the feedback.
Erich.
--089e0149d0747074d6052d8a39a5
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
<div dir=3D"ltr"><div class=3D"gmail_extra"><span class=3D""></span><span c=
lass=3D""></span><div class=3D"gmail_quote"><blockquote class=3D"gmail_quot=
e" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">=
<span class=3D"">
</span>... and workflow nightmares :D (no, really, I had to redo a schemati=
c<br>
because I needed some footprints, found Erich's kicad converted stuff o=
n<br>
gedasymbols, and then I thought "let's get all of gedasymbols and =
add it<br>
to the component library, what could possibly go wrong?")<br>
<br>
some curation will be needed somewhere, and things will still need<br>
manual tweaking to integrate nicely, because conventions may differ: for<br=
>
instance, those converted nixie footprints have alphanumeric pin names<br>
(A, LHDP, RHDP), while the convention I was aware of is pin name =3D=3D pin=
<br>
number (so my symbol's pins didn't match up with footprint pins);<b=
r>
however, changing the footprint pin names was easy, I very much prefer<br>
this to drawing the thing from scratch :)<br>
<br>
another example of clashing conventions is an Arduino nano symbol (I<br>
think also from Erich), which has pins very tightly packed and<br>
everything aligned at 25 grid units, not 100 like regular symbols.<br>
<br></blockquote><div><br></div><div>I have been reluctant to put much more=
onto gedasymbols just yet because of the need for curation, and the very a=
rduino symbols you refer to prompted my enforced pins spacing code which ca=
n enforce a specified spacing (when I nearly lost my mind trying to wire up=
an arduino nano in gschem with oddball pin spacing).<br></div></div><br></=
div><div class=3D"gmail_extra">In doing this, though, other silk elements t=
hat prettify the symbol may be ruined.<br></div><div class=3D"gmail_extra">=
<br></div><div class=3D"gmail_extra">My impression is that the worst offend=
ers are symbols that have been converted to kicad with an eagle ulp script =
- they seem to have some very odd pin naming conventions.<br><br></div><div=
class=3D"gmail_extra">My eagle -> kicad code tries to be more sensible,=
and applies the (multiple!) pin mappings provided in the eagle deviceset (=
symbol flavour -> particular footprint) to create distinct symbol versio=
ns with an associated footprint also converted from Eagle.<br><br></div><di=
v class=3D"gmail_extra">As you say though, the hard work is the geometry, a=
nd pin mappings often need to be verified before sending a board off for pr=
oduction.<br><br></div><div class=3D"gmail_extra">I will try to flag oddbal=
l spacings and prompt the user to enforce=C2=A0 a preferred pin spacing.<br=
><br></div><div class=3D"gmail_extra">Thanks for the feedback.<br><br></div=
><div class=3D"gmail_extra">Erich.<br></div></div>
--089e0149d0747074d6052d8a39a5--
- Raw text -