post-layout simulation using ocean, errors

[ansonyeap]

Hi all, recently i doing my post-layout simulation using ocean. the simulator that i used is spectre.

when the process is running, a notice from spectre pop up. the notice is as shown below.

Notice from spectre during topology check.
No DC path from node `I0.N_NET36_c_4352_p' to ground, Gmin installed to
provide path.
No DC path from node `I0.x_PM_NHP_REV1%NET30.259' to ground, Gmin installed
to provide path.
No DC path from node `I0.NET30' to ground, Gmin installed to provide path.
No DC path from node `I0.x_PM_NHP_REV1%NET36.50' to ground, Gmin installed
to provide path.
Warning from spectre during heuristic topology check - set topcheck=fixall to
fix all floating nodes.
No DC path from node `I0.x_PM_NHP_REV1%VDD.7159' to ground.


And when the simulation finished, all the result i get vary a lot from what i expected. what seem to be the problem and how to solve the above problem? furthermore, i check already my design, my design not suppose to be connected to ground for the corresponding nodes shown in the notice.

 

[varunmjman]

OCEAN simulation errors

In order to calculate the node voltages u need a ground reference. Like for example u have to capacitors connected in series then the node between them is floating in DC analysis. But to find the initial condition the tool must have a voltage there. So it assumes the presence of a high value resistor from that node to ground. Gmin = 1/R.
So check the value of Gmin in your script file, which if u dont specify will be a default value. Typically the value will be in the range of pS. U can check the value of Gmin the tool uses in your log file.
If the Gmin ur simulator using is high make it a low value and rerun the simulation

 

[tzg6sa]

Re: OCEAN simulation errors

As it was said above these are usual messages and the installed gm should not cause a problem. I am sure that the problem lays somewhere else. (for example the parasitics could modify the frequency response or even the bias point)

 

[JoannesPaulus]

OCEAN simulation errors

What you are showing is not an error but just a notice. As the previous two posters said, there are floating nets in your post-layout simulations. This is NOT the cause of your variations in the simulation results.

If I were you, I would check the effect of parasitic capacitance that you did not consider in the schematic simulations.