I'm trying to run a Monte Carlo analysis, where the positive and negative deviations of some resistors and capacitors are not the same (ex: the positive deviation might be 1%, and the negative deviation might be 5%). I've tried a few things so far:
1.) I've tried using the resistors and capacitors from the ANALOG library, but they only have a single TOLERANCE value. While I am able to run a Monte Carlo analysis when I give the components a tolerance, I can't determine a way to add both a positive and negative tolerance, meaning both of them have to be the same.
2.) I've tried using the resistors and capacitors from the BREAKOUT library, and editing the PSpice model to include "DEV = 1%/5%". However, upon running the Monte Carlo simulation, it appears that the intended positive tolerance is used for both directions. So, if I put "DEV = 1%/5%", it will give me a +/-1% deviation. If I put "DEV = 5%/1%", it will give me a +/-5% deviation. Once again, it seems like I can't add a different positive and negative tolerance.
3.) I've tried using the resistors and capacitors from the PSPICE_ELEM library, which have both POSTOL and NEGTOL values. However, even when I put a value in for each of these, running the simulation results in the Monte Carlo analysis being skipped, because "no models had tolerances." It doesn't appear to recognize POSTOL and NEGTOL, treating them both as tolerances of 0%, even if I add in another component with a tolerance just to get the simulation to run.
So, now I'm starting to wonder if it's possible to run a Monte Carlo analysis, with different positive and negative deviations. Is there a way to do this?