Hi, we have some odd post-layout simulation differ...
# ieee-sscs-dc-23a
aquiles viza
05/15/2024, 6:41 AMHi, we have some odd post-layout simulation differences.
The first image comes from a simulation with the schematic, next is the magic clean and klayout extraction and finally the magic pex extraction, which as different behaviour.
I'm sharing the files necessary to run this simulation, including spices and layout file.
j
Junbeom Park
05/15/2024, 7:32 AMHmm... very different from what you designed
a
aquiles viza
05/15/2024, 2:22 PM@Tim Edwards @Amro Tork @Mitch Bailey Do you have a clue of what might be the cause of this differences? I think the problem is on the extraction process of magic. There is a lot of FLOATING caps, I'm not sure what does it mean
t
Tim Edwards
05/15/2024, 3:10 PM@aquiles viza: The floating caps all come from the cell "resistor" which has a number of contacts above and below the ends of the resistor segments. A number of these contacts are floating, which produces the floating cap entries in the extracted netlist.
Tim Edwards
05/15/2024, 3:13 PMFloating caps will not cause an issue with simulation, although I have seen issues with the simulation just getting started, which requires setting the RSHUNT option in the simulator. I do not know if ngspice-42 still has that same problem. If you got a result at all from simulation, though, then you either didn't encounter that problem or else worked around it. Either way, it will not affect your simulation results.
3 Views