netlist2ss 0.2.1

A package to convert spice netlist into space state representation

Netlist to space state

This program converts a spice like netlist into a state-space representation of the system. You can choose the input variables, linearize unlinear systems, calculate the dc operating point, and choose the output variables. The states are column vector taken as the voltage in the capacitors and the current in the inductos, respecting the same order in which they are included in the netlist from top to bottom.

More details and examples can be found below:

Files

test: folder containing all the tests for the code. Type python3 test/test.py to run them.

examples: sample netlists

/netlist2ss/netlist2ss.py: calculate the space space-state representation

/netlist2ss/sisotf.py: instantiate the netlist2ss package in order to compute the transfer function for a single input and single output system (the netlist2ss-sisotf command is added when you install this package using the setup-tools)

/netlist2ss/__init__.py: init file

Simple Example

1. Open python3 in a terminal

2. Calculate the space state matrices of 'cap_filt.sp' circuit by typing the following lines in the python interpreter

    from netlist2ss import netlist2ss 
    handle  = open('./examples/cap_filt.sp', 'r')
    netlist = handle.read()
    handle.close()
    A,B,C,D,OP = netlist2ss(netlist,['IN'],['VnOUT'])

* ['IN'] is the list of the input variables. Input variables can be any variable used as value of a device in the netlist.
* ['VnOUT'] list of output measurements. Use Vn<node> to mesaure the voltage in a specific node, Vd<dev> to measure de voltage across a device, and Id<dev> for the current across a device. Vc<dev> and Ic<dev> returns the voltage and the current between the control pins of controlled sources, respectively.

3. The transfer function can be calculated by running:

    import sympy as si
    s = si.symbols('s')
    H = si.simplify(C*((s*(si.eye(A.shape[0]))-A).inv())*B + D)[0,0]

Second Example - Non linear system

netlist2ss will always calculate the A, B, C, and D matrices as the jacobian matrix of the state and output equations a function of the states and inputs. As a results, netlist2ss will return the small signal variation about the operating point when the equations are non-linear.

You can try the following:

    from netlist2ss import netlist2ss 
    import sympy as si
    netlist = ("V1 N1 GND ln(IN)\n"
               "R1 N1 N2  R1\n"
               "C1 N2 GND C1\n")
    A,B,C,D,OP = netlist2ss(netlist,['IN'],['VnN2']) 
    s = si.symbols('s')
    H = si.simplify(C*((s*(si.eye(A.shape[0]))-A).inv())*B + D)[0,0]
    print(OP)
    print(H)

The operating point will be ln(IN) and the transfer function will be 1/( IN (C1R1s + 1) ), which means that the transfer funciton depends on the DC value of the input.

Netlist

As shown in the example, the netlist is composed of a list of devices. The interconections between the devices (the nets) can written as any valid spice net name. The keywords gnd (in any combination of up and lower case letters) and 0 means the reference potential node and they must be present at least once.

The value of the device can be any variable name or constant. Don't use any reserved word of sympy!

The device type is identified by its first letter. You can find a list of the supported devices below:

type	Description
V	Independent voltage source
I	Independent current source
E	Voltage controlled voltage source
F	Current controlled current source
G	Voltage controlled current source
H	Current controlled voltage source
L	Inductor
C	Capacitor
R	Resistor
T	Ideal transformer

Limitations

* capacitors can't be connected in parallel with voltage sources or in parallel with other capacitors.
* inductors can't be connected in series with current sources or in series with other inductors.

Install

If you with to install the package in your computer, type:

    pip3 install netlist2ss