Mail Archives: geda-user/2016/01/09/07:50:14
--089e0158b5a09062400528e62217
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
On 9 January 2016 at 11:37, John Doty <jpd AT noqsi DOT com> wrote:
>
> One of the things I find confusing about pcb is that layers have types.
> That=E2=80=99s not physics. Layers are made of materials.
>
>
Whilst I feel you are probably being pedantic on this point, I actually
agree with you.
I think you would enjoy reading about the STEP AP210 board stack model.
(Actually, it is called a layered interconnect model). See this for the
basic concepts:
http://www.wikistep.org/index.php/AP210ed2_concept_of_operations
All entirely physics and materials based, _including_ explicitly defining
relevant properties such as electrical, thermal conductivity, permittivity
etc..
There is a mind-set shift between AP210 (which I think is along the lines
of your thinking), with traditional PCB design tools (not just gEDA). The
former, is the way 3D cad went a long time ago... you are modelling the
finished product. The later (old way), is using the layout tool as a 2.5D
drafting package, where drawing layers "might" model finished copper
materials in the PCB stackup - but really, are conceptually only modelling
instructions for the photo-plotter to make tooling for board manufacture.
In addition to layers which model the photo-tool for producing copper, you
have drawing layers that are used to produce silk-screens, solder mask
prints etc.. Layers can get used arbitrarily, and semantic meaning is only
conveyed by interaction with humans at the board fabricator. (Notes layer,
fab layer, "outline" layer etc..).
For the 3D stuff I've been working on, I need semantic data - where
"layers" model actual physical features with no requirement for human
interpretation. I need explicit properties like board/layer outlines,
thicknesses etc.. My prototype code infers some of this from "magic"
layers, such as "outline", but I will not upstream that, as it is not in
the slightest bit robust to do so.
FWIW, all of the later, non-gerber formats which the industry keeps trying
(and often failing) to replace gerber with, have better semantic notion of
the actual board construction. I'm skeptical any get as far as AP210 in
terms of completeness of their physical model, but I'd warrant they are
likely simpler to use!
Peter
--089e0158b5a09062400528e62217
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
<div dir=3D"ltr"><div class=3D"gmail_extra"><br><div class=3D"gmail_quote">=
On 9 January 2016 at 11:37, John Doty <span dir=3D"ltr"><<a href=3D"mail=
to:jpd AT noqsi DOT com" target=3D"_blank">jpd AT noqsi DOT com</a>></span> wrote:<br>=
<blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-=
left:1px solid rgb(204,204,204);padding-left:1ex"><span class=3D""><br>
</span>One of the things I find confusing about pcb is that layers have typ=
es. That=E2=80=99s not physics. Layers are made of materials.<br>
<br></blockquote><div><br></div><div>Whilst I feel you are probably being p=
edantic on this point, I actually agree with you.<br></div><div><br></div><=
div>I think you would enjoy reading about the STEP AP210 board stack model.=
(Actually, it is called a layered interconnect model). See this for the ba=
sic concepts:<br><a href=3D"http://www.wikistep.org/index.php/AP210ed2_conc=
ept_of_operations">http://www.wikistep.org/index.php/AP210ed2_concept_of_op=
erations</a><br><br></div><div>All entirely physics and materials based, _i=
ncluding_ explicitly defining relevant properties such as electrical, therm=
al conductivity, permittivity etc..<br><br><br></div><div>There is a mind-s=
et shift between AP210 (which I think is along the lines of your thinking),=
with traditional PCB design tools (not just gEDA). The former, is the way =
3D cad went a long time ago... you are modelling the finished product. The =
later (old way), is using the layout tool as a 2.5D drafting package, where=
drawing layers "might" model finished copper materials in the PC=
B stackup - but really, are conceptually only modelling instructions for th=
e photo-plotter to make tooling for board manufacture.<br><br></div><div>In=
addition to layers which model the photo-tool for producing copper, you ha=
ve drawing layers that are used to produce silk-screens, solder mask prints=
etc.. Layers can get used arbitrarily, and semantic meaning is only convey=
ed by interaction with humans at the board fabricator. (Notes layer, fab la=
yer, "outline" layer etc..).<br><br></div><div>For the 3D stuff I=
've been working on, I need semantic data - where "layers" mo=
del actual physical features with no requirement for human interpretation. =
I need explicit properties like board/layer outlines, thicknesses etc.. My =
prototype code infers some of this from "magic" layers, such as &=
quot;outline", but I will not upstream that, as it is not in the sligh=
test bit robust to do so.<br><br><br></div><div>FWIW, all of the later, non=
-gerber formats which the industry keeps trying (and often failing) to repl=
ace gerber with, have better semantic notion of the actual board constructi=
on. I'm skeptical any get as far as AP210 in terms of completeness of t=
heir physical model, but I'd warrant they are likely simpler to use!<br=
><br><br></div><div>Peter<br></div><div><br></div></div></div></div>
--089e0158b5a09062400528e62217--
- Raw text -