Re: gEDA-user: some problem about gnetlist and subsckt

2008-08-11 Thread Ludovic SMADJA 
thanks for the history .

Is there a new tutorial with gnucap use, somewhere ?

Ludovic

On Fri, Aug 8, 2008 at 9:04 AM, Ludovic SMADJA [EMAIL PROTECTED]wrote:

 Hi, everyone

 In order to create and simulate a quite complex schema, I've first tried to
 follow the tutorial (http://www.brorson.com/gEDA/SPICE/intro.html) to
 create a simple subckt with gschem and simulate it.

 The subschema is a simple filter with 1 resistor and 1 capacitor. I've
 integrated the schema in an upper one with one voltage source and a charge
 resistor.

 When I generate the netlist, gnetlist doesn't complains about anything but
 the generated netlist is incorrect, the subckt is created but with the full
 schema, not only the subschema :

 here's my files and the command I used. Could you indicate what's I've do
 wrong, I've tried all the past day to generate a correct netlist and I can't
 see what's the problem ?

 Regards,

 Ludovic SMADJA

 Commands :

 gnetlist -g spice-sdb -o filtre.cir filtre.sch

 gnetlist -g spice-sdb -o testFiltre.cir testFiltre.sch

 **
 [EMAIL PROTECTED] test]$ cat filtre.sch
 v 20071231 1
 C 4 4 0 0 0 title-B.sym
 C 46600 46300 1 0 0 resistor-2.sym
 {
 T 47000 46650 5 10 0 0 0 0 1
 device=RESISTOR
 T 46800 46600 5 10 1 1 0 0 1
 refdes=R1
 T 46600 46300 5 10 1 0 0 0 1
 value=100k
 }
 C 48100 45300 1 90 0 capacitor-1.sym
 {
 T 47400 45500 5 10 0 0 90 0 1
 device=CAPACITOR
 T 47600 45500 5 10 1 1 90 0 1
 refdes=C1
 T 47200 45500 5 10 0 0 90 0 1
 symversion=0.1
 T 48100 45300 5 10 1 0 0 0 1
 value=1Uf
 }
 C 44100 47600 1 0 0 spice-subcircuit-LL-1.sym
 {
 T 44200 47900 5 10 0 1 0 0 1
 device=spice-subcircuit-LL
 T 44200 48000 5 10 1 1 0 0 1
 refdes=A1
 T 44200 47700 5 10 1 1 0 0 1
 model-name=filtre_model
 T 44100 47600 5 10 1 0 0 0 1
 value=filtre_model
 }
 C 46100 46700 1 180 0 spice-subcircuit-IO-1.sym
 {
 T 45200 46300 5 10 0 1 180 0 1
 device=spice-IO
 T 45250 46450 5 10 1 1 180 0 1
 refdes=P1
 T 46100 46700 5 10 1 0 0 0 1
 pinseq=1
 T 46100 46700 5 10 1 0 0 0 1
 pinlabel=IN
 T 46100 46700 5 10 1 0 0 0 1
 value=IN
 }
 N 45900 46400 46600 46400 4
 N 47900 46400 47900 46200 4
 N 47500 46400 48600 46400 4
 N 47900 45300 47900 44900 4
 N 45900 44900 48700 44900 4
 C 46100 45200 1 180 0 spice-subcircuit-IO-1.sym
 {
 T 45200 44800 5 10 0 1 180 0 1
 device=spice-IO
 T 45250 44950 5 10 1 1 180 0 1
 refdes=P2
 T 46100 45200 5 10 1 0 0 0 1
 pinseq=2
 T 46100 45200 5 10 1 0 0 0 1
 pinlabel=GND1
 T 46100 45200 5 10 1 0 0 0 1
 value=GND1
 }
 C 48400 46100 1 0 0 spice-subcircuit-IO-1.sym
 {
 T 49300 46500 5 10 0 1 0 0 1
 device=spice-IO
 T 49250 46350 5 10 1 1 0 0 1
 refdes=P3
 T 48400 46100 5 10 1 0 0 0 1
 pinseq=3
 T 48400 46100 5 10 1 0 0 0 1
 pinlabel=OUT
 T 48400 46100 5 10 1 0 0 0 1
 value=OUT
 }
 C 48500 44600 1 0 0 spice-subcircuit-IO-1.sym
 {
 T 49400 45000 5 10 0 1 0 0 1
 device=spice-IO
 T 49350 44850 5 10 1 1 0 0 1
 refdes=P4
 T 48500 44600 5 10 1 0 0 0 1
 pinseq=4
 T 48500 44600 5 10 1 0 0 0 1
 pinlabel=GND2
 T 48500 44600 5 10 1 0 0 0 1
 value=GND2
 }

 
 [EMAIL PROTECTED] test]$ cat filtre.sym
 v 20071231 1
 B 300 300 2800 1200 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
 T 3100 1600 9 10 0 0 0 0 1
 device=filtre_model
 T 1700 400 9 10 1 1 0 3 1
 refdes=X?
 T 1700 800 9 10 1 1 0 3 1
 Filtre
 P 0 1100 300 1100 1 0 0
 {
 T 350 1100 9 10 1 1 0 1 1
 pinlabel=P1
 T 200 1150 5 8 1 1 0 6 1
 pinnumber=1
 T 200 1150 5 8 0 1 0 6 1
 pinseq=1
 T 0 1100 5 10 1 0 0 0 1
 refdes=IN
 }
 P 0 700 300 700 1 0 0
 {
 T 350 700 9 10 1 1 0 1 1
 pinlabel=P2
 T 200 750 5 8 1 1 0 6 1
 pinnumber=2
 T 200 750 5 8 0 1 0 6 1
 pinseq=2
 T 100 700 5 10 1 0 0 0 1
 refdes=GND1
 }
 P 3400 1100 3100 1100 1 0 0
 {
 T 3050 1100 9 10 1 1 0 7 1
 pinlabel=P3
 T 3200 1150 5 8 1 1 0 0 1
 pinnumber=3
 T 3200 1150 5 8 0 1 0 0 1
 pinseq=3
 T 3400 1100 5 10 1 0 0 0 1
 refdes=OUT
 }
 P 3400 700 3100 700 1 0 0
 {
 T 3050 700 9 10 1 1 0 7 1
 pinlabel=P4
 T 3200 750 5 8 1 1 0 0 1
 pinnumber=4
 T 3200 750 5 8 0 1 0 0 1
 pinseq=4
 T 3400 700 5 10 1 0 0 0 1
 refdes=GND2
 }
 T 1700 400 8 10 0 1 0 0 1
 source=filtre.sch


 **
 [EMAIL PROTECTED] test]$ cat testFiltre.sch
 v 20071231 1
 C 4 4 0 0 0 title-B.sym
 C 47100 46700 1 0 0 filtre.sym
 {
 T 50200 48300 5 10 0 0 0 0 1
 device=filtre_model
 T 48800 47100 5 10 1 1 0 3 1
 refdes=X1
 T 48100 46500 5 10 1 0 0 0 1
 file=filtre.cir
 T 47100 46700 5 10 1 0 0 0 1
 model-name=filtre_model
 }
 C 51900 47100 1 90 0 resistor-2.sym
 {
 T 51550 47500 5 10 0 0 90 0 1
 device=RESISTOR
 T 51600 47300 5 10 1 1 90 0 1
 refdes=RTest
 }
 C 46200 47200 1 90 0 voltage-1.sym
 {
 T 45700 47300 5 10 0 0 90 0 1
 device=VOLTAGE_SOURCE
 T 45700 47500 5 10 1 1 90 0 1
 refdes=V1
 }
 N 46000 48100 47100 48100 4
 N 47100 48100 47100 47800 4
 N 46000 47200 47100 47200 4
 N 47100 47200 47100 47400 4
 N 50500 47800 50500 48400 4
 N 50500 48400 51800 48400 4
 {
 T 50500 48400 5 10 1 0 0 0 1

Re: gEDA-user: some problem about gnetlist and subsckt

2008-08-09 Thread al davis 
On Friday 08 August 2008, Ludovic SMADJA wrote:
 In order to create and simulate a quite complex schema, I've
 first tried to follow the tutorial
 (http://www.brorson.com/gEDA/SPICE/intro.html) to create a
 simple subckt with gschem and simulate it.

Yikes   That document is 5 years old, and doesn't reflect 
what we have now.  

For one, it only briefly mentions that gnucap exists, even 
though the main gnucap author is here, and has been since the 
beginning of geda, and it emphasizes other simulators, from 
people never involved with geda.  For one of the simulators 
emphasized, the person in charge hates open-source, and has 
said so openly.

 The subschema is a simple filter with 1 resistor and 1
 capacitor. I've integrated the schema in an upper one with
 one voltage source and a charge resistor.

 When I generate the netlist, gnetlist doesn't complains about
 anything but the generated netlist is incorrect, the subckt
 is created but with the full schema, not only the subschema :

I have found that if I just make a schematic in a way that seems 
to make sense, gnetlist will not generate a netlist that is 
directly useful.  It is almost correct and requires some 
editing.  Eventually, you can learn the tricks to make the 
generated netlist closer to being useful, but often still not 
100%.

I just accept that I will need to edit the netlist after 
generation.

This issue has been known for years and has come up in 
discussion many times, but nobody has stepped up to seriously 
address the problem.

A while back, someone (Kai?) did a survey about how people are 
using geda, and revealed that few are using it for simulation.  
This is why.


___
geda-user mailing list
geda-user@moria.seul.org
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user

Re: gEDA-user: some problem about gnetlist and subsckt

2008-08-08 Thread John Doty 

On Aug 8, 2008, at 3:04 AM, Ludovic SMADJA wrote:

 In order to create and simulate a quite complex schema, I've first  
 tried to follow the tutorial (http://www.brorson.com/gEDA/SPICE/ 
 intro.html) to create a simple subckt with gschem and simulate it.

 The subschema is a simple filter with 1 resistor and 1 capacitor.  
 I've integrated the schema in an upper one with one voltage source  
 and a charge resistor.

 When I generate the netlist, gnetlist doesn't complains about  
 anything but the generated netlist is incorrect, the subckt is  
 created but with the full schema, not only the subschema :

Here's your problem:

 source=filtre.sch


If your symbol has that attribute, gnetlist will expand the  
subcircuit. If you leave that attribute out you'll get the gnetlist  
behavior you want. But, unfortunately, then the Hd command in gschem  
won't work.

John Doty  Noqsi Aerospace, Ltd.
http://www.noqsi.com/
[EMAIL PROTECTED]




___
geda-user mailing list
geda-user@moria.seul.org
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user