If given a step signal, how to set "maximum step size" in PSPICE Trasient Analysis?

[Julia2013]

If given a step signal, how to set "maximum step size" in PSPICE Trasient Analysis?

Hi All:

I am reading an example, where

a IPWL from source.olb is placed and properties set as follows:

T1=0s I1=0A
T2=10ms I2=0A
T3=10.1ms I2=1A

It is approximately a step signal with 0.1ms for the rise time.


Later in the example , it uses "maximum step size" of 100ms for transient analysis,

I am doubting if this is accurate enough??? Because this is one thousand times of the rise time!



I know how to set the "maximum step size" for sine waves,

but which rules should I follow for such kind of signals?

Does it have something to do with the rise time?


Any help would be greatly appreciated. Thank you a lot.

 

[rahdirs]

Re: If given a step signal, how to set "maximum step size" in PSPICE Trasient Analysi

It depends on the your waveform.Are you losing any critical data-points during the period when step size is 1000 times rise time ?????

 

[keith1200rs]

Re: If given a step signal, how to set "maximum step size" in PSPICE Trasient Analysi

The simulator will vary the time step to suit the changes it sees in each iteration step so it is unlikely to jump over the transition. However, unless it is a complex circuit causing slow simulations I would normally set the maximum time step at least an order of magnitude smaller than the shortest rise time.

Keith

 

[abhajn]

Re: If given a step signal, how to set "maximum step size" in PSPICE Trasient Analysi

Keith is right - simulator checks for transitions and intelligently reduces the time-step. Maximum time step does not mean that the simulator will always use that time step.
Mostly I prefer to leave the Max time step undefined, so that the simulator calculates the time step on its own. If there is loss of resolution in results, only then would I set the max time step to a smaller value.

Also - PSpice supports SCHEDULE statement for the max time step. That means you can have a small TMax for a small part of the simulation (which has switching), and have it run fast otherwise.
e.g. SCHEDULE(0s, 100m, 9.9ms, .01m, 10.2ms, 100m) => TMAX of .01ms between TIME=9.9m and 10.2m, and 100ms otherwise.

 

[keith1200rs]

Re: If given a step signal, how to set "maximum step size" in PSPICE Trasient Analysi

SIMetrix has a similar "fast start" which can be useful for getting oscillators stable before slowing it down for more detailed results.

Keith