DC noconvergence in Hspice

[wenty]

With the same netlist, Hsim is easy to converge than Hspice. So, can I include the '*.ic' got from Hsim to solve the DC noconvergence problem in Hspice?
thanks

 

[flushrat]

Yes, you can. And, you must setup a high accuracy for hsim. Actually, I sometimes do in this way for postsimulation.

Another way is to use uic to generate an IC. I think this should be the first choice.

 

[tmchen00]

You can only set .ic for few critical nets, or it will occur some convergency problem
at the middle of trancient simulation.

 

[alok_ky]

we can also use .nodeset command.

Again, iteration limit can also be set.

 

[jiangwp]

Maybe, the file is useful for you.

 

[nuiscet]

Another way may be used: change the step of the simulation to smaller.

Added after 2 minutes:

sorry for mistaking 'change the step of the simulation to smller ' for 'chang the step of the simulation to larger'

 

[zenisle]

the dc noconvergance maybe be solutioned by set init node voltage.