End-to-end design in Protel Schematic and Autotrax

[budgie]

This falls under the ehading of "dumb questions" for some, I'm sure.

In Schematic, semiconductor elements appear with their pins labelled with their
function (A/K, E/B/C etc) whereas POST expects 1,2 (,3) etc to be able to
generate a valid netlist - otherwise none of the connections to these semis
appear in the netlist when imported into AutoTrax. I suppose one could
laboriously assign net labels to *each* connection, but I'm sure there is a
neater and simpler way. I'm curious to discover how others address this.

 

[Alan]

On Sat, 10 May 2008 12:10:05 +0800, budgie <me@privacy.net> wrote:

This falls under the ehading of "dumb questions" for some, I'm sure.

In Schematic, semiconductor elements appear with their pins labelled with their
function (A/K, E/B/C etc) whereas POST expects 1,2 (,3) etc to be able to
generate a valid netlist - otherwise none of the connections to these semis
appear in the netlist when imported into AutoTrax. I suppose one could
laboriously assign net labels to *each* connection, but I'm sure there is a
neater and simpler way. I'm curious to discover how others address this.
Use SLM.EXE to change the pin numbers from B C E to 1 2 3

Alan

--
Sell your surplus electronic components at
http://ozcomponents.com
Search or browse for that IC, capacitor,
crystal or other component you need.

 

[budgie]

On Sun, 11 May 2008 10:04:12 +0800, Alan <me@somewhere.com.au.invalid> wrote:

On Sat, 10 May 2008 12:10:05 +0800, budgie <me@privacy.net> wrote:

This falls under the ehading of "dumb questions" for some, I'm sure.

In Schematic, semiconductor elements appear with their pins labelled with their
function (A/K, E/B/C etc) whereas POST expects 1,2 (,3) etc to be able to
generate a valid netlist - otherwise none of the connections to these semis
appear in the netlist when imported into AutoTrax. I suppose one could
laboriously assign net labels to *each* connection, but I'm sure there is a
neater and simpler way. I'm curious to discover how others address this.

Use SLM.EXE to change the pin numbers from B C E to 1 2 3

yebbut ....

That infers that all (for instance) transistors using a TO-92D or whatever
footprint have the same pinout - which regrettably they don't. That was what I
was trying to avoid, as it fixes one problem and creates a new one - a trap
whenever a TO-92D footprint component with a different pinout was selected.

Certainly valid with diodes though - despite the lack of standardisation in
transistors there aren't too many diode pinout variations ;-)

 

[The Real Andy]

On Sun, 11 May 2008 13:59:48 +0800, budgie <me@privacy.net> wrote:

On Sun, 11 May 2008 10:04:12 +0800, Alan <me@somewhere.com.au.invalid> wrote:

On Sat, 10 May 2008 12:10:05 +0800, budgie <me@privacy.net> wrote:

This falls under the ehading of "dumb questions" for some, I'm sure.

In Schematic, semiconductor elements appear with their pins labelled with their
function (A/K, E/B/C etc) whereas POST expects 1,2 (,3) etc to be able to
generate a valid netlist - otherwise none of the connections to these semis
appear in the netlist when imported into AutoTrax. I suppose one could
laboriously assign net labels to *each* connection, but I'm sure there is a
neater and simpler way. I'm curious to discover how others address this.

Use SLM.EXE to change the pin numbers from B C E to 1 2 3

yebbut ....

That infers that all (for instance) transistors using a TO-92D or whatever
footprint have the same pinout - which regrettably they don't. That was what I
was trying to avoid, as it fixes one problem and creates a new one - a trap
whenever a TO-92D footprint component with a different pinout was selected.

Certainly valid with diodes though - despite the lack of standardisation in
transistors there aren't too many diode pinout variations ;-)

You can only go so far with generic footprints. What I used to do was
create a seperate schematic component, and a seperate footprint for
every component I used. This takes time, but eventually it pays off
when you have a good library. Fortunatly you can usually adapt
existing footprints to use for you components. I also used to maintain
a library of footprints from the IPC standards as well, so normally it
just required a cut and paste and your fottprint was completed.

 

[Alan]

On Sun, 11 May 2008 16:48:15 +1000, The Real Andy
<therealandy@nospam.com> wrote:

On Sun, 11 May 2008 13:59:48 +0800, budgie <me@privacy.net> wrote:

On Sun, 11 May 2008 10:04:12 +0800, Alan <me@somewhere.com.au.invalid> wrote:

On Sat, 10 May 2008 12:10:05 +0800, budgie <me@privacy.net> wrote:

This falls under the ehading of "dumb questions" for some, I'm sure.

In Schematic, semiconductor elements appear with their pins labelled with their
function (A/K, E/B/C etc) whereas POST expects 1,2 (,3) etc to be able to
generate a valid netlist - otherwise none of the connections to these semis
appear in the netlist when imported into AutoTrax. I suppose one could
laboriously assign net labels to *each* connection, but I'm sure there is a
neater and simpler way. I'm curious to discover how others address this.

Use SLM.EXE to change the pin numbers from B C E to 1 2 3

yebbut ....

That infers that all (for instance) transistors using a TO-92D or whatever
footprint have the same pinout - which regrettably they don't. That was what I
was trying to avoid, as it fixes one problem and creates a new one - a trap
whenever a TO-92D footprint component with a different pinout was selected.

Certainly valid with diodes though - despite the lack of standardisation in
transistors there aren't too many diode pinout variations ;-)

You can only go so far with generic footprints. What I used to do was
create a seperate schematic component, and a seperate footprint for
every component I used. This takes time, but eventually it pays off
when you have a good library. Fortunatly you can usually adapt
existing footprints to use for you components. I also used to maintain
a library of footprints from the IPC standards as well, so normally it
just required a cut and paste and your fottprint was completed.

Not easy to do a cut and paste in autotrax/schmatic or slm. However
if you name a new component you can then copy an existing component's
attributes and make suitable changes. So you could have a library of
transistors listed by part number or listed by footprint. If listed
by part number then it means creating a new part in slm each time you
use a new part but the advantage is that you just refer to say BC107
or 2N2222 without worrying about remembering the footprint number. If
using the footprint then generate say TO-92a, TO-92b, TO-92c, etc as
separate components (pretty quick to do with new, copy attributes,
etc) then make a layout with all the footprints on it and print it out
for easy reference!

You can do the same for SMD parts as well. By the way it is possible
to use multiple libraries in schmatic (just in case you didn't know)
although I always have to work out how to change libraries each time I
want to (old age I think!)

Alan

--
Sell your surplus electronic components at
http://ozcomponents.com
Search or browse for that IC, capacitor,
crystal or other component you need.