Tutorial

Basic Usage

The core of nport are the classes NPort and TwoNPort. An NPort object stores frequency-dependent n-port network parameters, such as are typically stored in Touchstone (.snp where n is the number of ports) or CITI files. These files are generated by network analyzers and field solvers. We can load a Touchstone file into an NPort object like this:

>>> import nport
>>> from nport import touchstone

>>> filter = touchstone.read("filter.s2p")

Once network parameters have been read into an NPort object, they can be manipulated easily by using mathematical operations on it. Suppose we want to obtain the network parameters of two of the filters connected in cascade. The cascade connection of two 2-ports is given by the matrix multiplication of the transfer or ABCD 2-port matrices. An ABCD matrix is a 2n-port representation, so we first convert the filter NPort to a TwoNPort:

>>> filter2n = filter.twonport().convert(nport.ABCD)
>>> cascade2n = nport.dot(filter2n, filter2n)

We can now write the result back to a Touchstone file:

>>> cascade = cascade2n.nport()
>>> touchstone.write(cascade, "cascade")

If you are working with frequency-independent data, you can use the NPortMatrix and TwoNPortMatrix classes. These classes are used internally by NPort and TwoNPort to store the parameter matrices at each frequency sample. They support most of the same functions as their frequency-dependent versions:

>>> z = nport.NPortMatrix([[2.4+6.9j, 5.3+8.1j],
                           [8.4+6.4j, 6.7+2.7j]], nport.Z)
>>> z.convert(nport.H)
TwoPortMatrix(type=H)
 [[-1.93591414-6.57060176j,  1.09965504+0.76581066j],
  [-1.40973553-0.3871215j ,  0.12840169-0.05174396j]]

Extra Functionality

TODO:

filter.ispassive()

filter.ports

filter.get_parameter

transmission line conversion, change length (power), ...

plotting with mpl/chaco

smith plotting

stability functions (+ plotting of circles)

combining of NPorts

...