The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> the .PZ analyses on cadence virtuoso
https://designers-guide.org/forum/YaBB.pl?num=1209149101

Message started by mowiehowie on Apr 25^th, 2008, 11:45am

Title: the .PZ analyses on cadence virtuoso
Post by mowiehowie on Apr 25^th, 2008, 11:45am

Hi there. Im new here, and new on cmos analog design. so it goes:
Im studying the frequency response of the common amplifiers such as common-source, source-follower, differential..
im using the cadence virtuoso with spectre. the question is: what is the right way to perform a .pz analyses? i mean, the results are too different from the expected by my manual math (using Razavi). its weird. im getting nervous. >:(
for example on a CS stage, what I do is to put my amplifier on a convenient dc bias point , choose the pz analyses between the out (drain of the transitor) and the ground. it returns 2 poles and 1 zeroe that DONT match with my manual math, and, worst, DONT MATCH with the AC response plot.. thats it... i hope i was clear. i can speak in portuguese or in french too.

Title: Re: the .PZ analyses on cadence virtuoso
Post by sheldon on Apr 25^th, 2008, 7:09pm

Mowie,

Don't think the language is an issue.

First question how are calculating the small signal equivalent circuit for
your transistors? From reading your post, it is not clear how these are
being calculated. As process scale, it is no longer to appropriate to hand
calculate the transistors equivalent circuit. You should use the values in
the dc operating point information.

Next question, how are you constructing your testbenches? Are you
using an ideal current source or transistor current source to bias the
follower? The more realistic the testbench the higher order the analysis
you will need to do.

From personal experience, hand calculations will not exactly match
simulated results and the understanding the difference in the results
can teach you a lot about the circuit (and how dangerous simplifying
approximations can be).

Best Regards,

Sheldon