The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> Circuit Simulators >> gmin in hspice and spectre
https://designers-guide.org/forum/YaBB.pl?num=1265665676

Message started by Jin Tang on Feb 8^th, 2010, 1:47pm

Title: gmin in hspice and spectre
Post by Jin Tang on Feb 8^th, 2010, 1:47pm

Hi

I have a diode model in both spectre and hspice format. I try to compare the characteristics like dc, cv, et. al. I simulate both models in ICCAP. They are identical models (except the format is different) and i see almost identical behavior in all areas except the low current region where gmin is supposed to be in effect. Spectre shows about 1 order of magnitude lower current. Note that i set dskip=0 for spectre.

I set gmin=1e-15 (for spectre) and gmindc=1e-15 (for hspice). Any one has an idea why this is the case? Thanks in advance!

Jin

Title: Re: gmin in hspice and spectre
Post by Andrew Beckett on Feb 9^th, 2010, 3:34am

Maybe it's due to differences in iabstol (I forget the equivalent parameter in hspice). What is the magnitude of the currents that you're seeing? That might allow a more educated guess...

Regards,

Andrew.

Title: Re: gmin in hspice and spectre
Post by Jin Tang on Feb 9^th, 2010, 9:25am

The current level i am seeing is about 5e-16 for spectre and 5e-15 for hspice.

Title: Re: gmin in hspice and spectre
Post by Geoffrey_Coram on Feb 9^th, 2010, 1:08pm

Does the voltage source current equal the diode current?

I recall that some simulators don't include the GMIN current as part of the diode current.

Title: Re: gmin in hspice and spectre
Post by Jin Tang on Feb 9^th, 2010, 2:22pm

I think in my case GMIN current is indeed included, because the current level changes with GMIN.

Title: Re: gmin in hspice and spectre
Post by Andrew Beckett on Feb 10^th, 2010, 1:52pm

Bear in mind that this is a very small current - and it's almost certainly below the iabstol (unless you've tightened it) in both simulators, and so it's not going to be accurately resolved.

Why do you care about the accuracy of a current which is in femtoamps?

Regards,

Andrew.

Title: Re: gmin in hspice and spectre
Post by Jin Tang on Feb 10^th, 2010, 2:03pm

My project is to convert a spectre format diode to hspice format. and try to understand any difference. This indeed is a very small current. i think it is pretty much determined by the implementation of Gmin.