AnsweredAssumed Answered

Long Time Simulation in Ptolemy/Envelope Cosimulation

Question asked by tp78 on Jan 30, 2013

Im trying to run a two channel system with Ptolemy/Envelope, where I have already made the analog circuits with HB simulation. It contains filters (S2P data), behavioral amplifiers/mixers, and attenuators. In each channel takes the signal down converts to IF, runs through some ptolemy elements, then goes back to the analog circuit for upconversion and combines with the other channel. Each channel has its own operating BW and thus each upconversion/downconversion analog subcircuit branch has its own Envelope controller. Each controller has BOTH LOs and the RF signal as analysis frequencies. Its a little complicated to explain without pictures, but thats the best I can do.

The Problem:
I need to run a pulse-modulated signal with a significantly long pulse period (several milliseconds), with a very small time-step (must be 100ps). I can run a simulation of about 25 microseconds, but it takes almost 30 hours! Is there any reason why it takes so long to complete a seemingly simple simulation? Is there something simple I may be missing to simplify the simulation? 

Also I tried to simplify the system by removing the ptolemy elements and only running in a pulsed source. Furthermore the timestep was .5*modulation bandwidth (significantly larger). The simulation couldn't even get past 10 microseconds. 

Ultimately is there any way that I can run a several millisecond (at least) simulation, with the required timestep for the modulation bandwidth????