How to Model a Low Pass Filter using VerilogA ?

Probably the simplest is to either model an R and a C, or to use the laplace functions to give a s-domain description of the filter. You could use the various filters in rfLib that is shipped with Virtuoso, although note that these are built on the fly as you choose the parameters (this is because of the variable order of the filter; I forget now why it was done that way - maybe it preceded generator support in the language. So you might need to use them and then follow the link to the model within the simulation netlist or spectre.out file when simulating with them to see how they are implemented.

Regards,

Andrew

Thank you Andrew. I just modelled the R and C as mentioned by you. It seem to work fine as of now. I have no clue about how to do modelling in s-domain so did not choose that. Would love to know any references for the same, i just read about the laplace_sd and laplace_zp functions, couldn't really figure out if this would give me a time domain output for a transfer function H(s). Also i was unable to find rfLib in my Lib Manager for some reason. Does it comes pre defined in Library Manager like the analoglib or i need to include it in my CDS.lib ? 

Thank you for your help again. Me and my colleagues Really appreciate your answers.

There are some examples of using these in  The Designer's Guide to Verilog-AMS  (on pages 182, 183 - there are some earlier examples of modelling skin effect). The simplest example is using the _nd form:

V(out) <+ laplace_nd(V(in),{0,3},{4,0,1});

which implements:

V_out(s) =  3s
------    ------
V_in(s)      s²+4

The coefficients are in the order s⁰, s¹, s², s³ etc (in increasing power of s) for both the numerator and denominator. The _zp is similar but there you are specifying the poles and zeros (as an s-domain transfer function).

For rfLib, add this to your cds.lib:

DEFINE rfLib $(inst_root_with:tools/dfII/bin/virtuoso)/tools/dfII/samples/artist/rfLib

Regards,

Andrew

As always it helped Andrew ! I was able to implement a 1st and 2nd order filter using the above function. It required some coefficient tuning as per my R and C and gain but once it was fixed it worked perfectly fine.

Thank you for making life easy again ! 

Attaching a link for the explanation of above function for future reference :

https://help.simetrix.co.uk/8.0/simetrix/mergedProjects/verilog_a_reference/topics/veriloga_verilog_areference_verilog_afunctions.htm#laplacezp  

Regards,

Kulmani