Re: [Qucs-devel] Exporting netlists

[Thomas.lepoix <tho...@pr...>]

You're welcome.

It is a common thing for an electronic component to have multiple ports connected to the same wire, this is not a bug nor an error.

I don't know, I also spotter "_ref" in your netlist sample but I don't know what it is.

‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
On Thursday 9 April 2020 14:56, Chris <chr...@ym...> wrote:

> That confirms what I thought. I have written something robust enough to deal with this, but I wanted to be sure that my guesses were correct.
> Thanks Thomas!
>
> I have already posted the questions below, but since I now have your attention :)
>
>> 1) This component line in a sim file refers to the same net twice, that doesn't look right to me.
>>
>> BJT:T1 _net16 _net17 _net18 _net17 Type="npn"
>>
>> 2) In the component lines in a sim file, the net names appear to start with '_net', but there is also 'gnd', are there any other special net names like this?
>
> I'm hoping these are the last questions, as the script is working apart from problems with (1).
>
> Chris
>
> On 09/04/2020 1:45 pm, Thomas.lepoix wrote:
>
>> I don't remember having ever found a documentation about the netlist format.
>>
>> As far as I know, each line follow the same model :
>>
>> type:label net property="value"
>>
>> net fields are "wire" labels, for example, in the following sequence there is a wire that connects the second port of the diode D1, the first port of D2 and the first port of D3.
>>
>> Diode:D1 _net0 _net1 Is="1E-11" N="1.27" Rs="3.858 ...
>> Diode:D2 _net1 _net2 Is="1E-11" N="1.27" Rs="3.858 ...
>> Diode:D3 _net1 _net3 Is="1E-11" N="1.27" Rs="3.858 ...
>>
>> property fields are also present in the .sch file.
>>
>> It seems to me that def paragraphs wrap subcircuits. "Z_Diodes_1N756" is a particular model of diode. So this library component is a subcircuit wrapping a classic diode component with particular properties.
>>
>> "PVR100AZ_B12V_mod" is a more complex library component so it wraps another subcircuit "PVR100AZ_B12V" with its own def paragraph and then used through the "sub" component. that's why defs can be nested, a def contain a piece of schematic.
>>
>> however that is true that the file indentation is rather strange.. I guess your parser have to be robust enough to this kind of randomness.
>>
>> ‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
>> On Thursday 9 April 2020 00:13, Chris via Qucs-devel
>> [<quc...@li...>](mailto:quc...@li...)
>> wrote:
>>
>>> Files created using qucs -i filename1 -o filename2 -n
>>>
>>> Is there a definition of these files, and if so where?
>>>
>>> I think I can just use the component list at the end, so I need to skip
>>> the '.Def's, but they are confusing, since sometimes the .Def:End starts
>>> in the first column and sometimes not, and sometimes the .Def paragraphs
>>> are nested. Are they supposed to be that way?
>>>
>>> .Def*:*Z_Diodes_1N756 _net0 _net1
>>>       Diode:D1 _net0 _net1 Is="1E-11" N="1.27" Rs="3.858"
>>> Cj0="2.749E-10" Vj="0.75" M="0.33" Fc="0.5" Tt="5E-8" Bv="8.123"
>>> Ibv="0.01" Af="1" Kf="0"
>>> .Def:End
>>>
>>> .Def:PVR100AZ_B12V_mod _net4 _net1 _net2 _net3 _ref
>>> Sub:X1 _ref _net1 _net2 _net3 _net4 Type="PVR100AZ_B12V"
>>> .Def:PVR100AZ_B12V _ref _net1 _net2 _net3 _net4
>>> BJT:Q1 _net1 _net4 _net3 _ref Type="npn" Is="9.198e-014" Nf="1.003"
>>> Ise="4.468e-016" Ne="1.65" Bf="338.8" Ikf="0.4913" Vaf="107.9"
>>> Nr="1.002" Isc="5.109e-015" Nc="1.071" Br="29.48" Ikr="0.193" Var="25"
>>> Rb="1" Irb="1000" Rbm="1" Re="0.2126" Rc="0.143" Xtb="0" Eg="1.11"
>>> Xti="3" Cje="3.825e-011" Vje="0.7004" Mje="0.364" Tf="5.229e-010"
>>> Xtf="219.7" Vtf="3.502" Itf="7.257" Ptf="0" Cjc="1.27e-011" Vjc="0.4431"
>>> Mjc="0.3983" Xcjc="0.4555" Tr="7e-011" Cjs="0" Vjs="0.75" Mjs="0.333"
>>> Fc="0.905"
>>> Diode:D1 _net1 _net2 Is="9.794e-015" N="1.101" Bv="13" Ibv="0.005"
>>> Rs="0.3363" Cj0="4.377e-011" Vj="0.5656" M="0.3289" Fc="0.5"
>>> .Def:End
>>> .Def:End
>>>
>>> Qucs-devel mailing list
>>> Quc...@li...
>>>
>>> https://lists.sourceforge.net/lists/listinfo/qucs-devel