PSpice Simulation

[annex666]

I'm trying to simulate a filter and would like to know if there's an easy way of drawing the ideal response along with the worst case response(s) for a range of component values within the specified tolerances.

I've tried applying "worst case" in the simulation settings, but it only plots two traces and changing the tolerances it would appear to be the response from the ideal values and then the reponse with ALL circuit values at their max tolerance. What I want is the responses from ALL permutations of component values at minimum and maximum.

Can anyone suggest an answer? Is this even possible?

It should be 2^n + 1 simulations, which isn't too intensive for a circuit with 12 components - 4097 simulations * < 1sec = just over an hour.

Thinking about it I don't even need all the simulations performed if PSpice can intelligently deduce which simulations will give the worst results.

Any ideas? Any alternatives?

[jcx]

I think pspice supports the "dumber" approach; monte carlo simulation - you need lots more sims but you can get distribution data

Linear Technology' SwitcherCAD free spice has a “.step param”
option that can form nested loops, I'm not sure how deep – probably similar cmd in other spice’ as well

Kharitonov polynomials let you bound the root location of a polynomial with given ranges for the coefficients – turns out to be a 4 sided rectangle with vertices defined by the 4 Kharitonov polynomials – if you have zeros too you would get 16 "extremal" transfer functions