Hello everyone, I have been working on the layout ...
# magic
Hello everyone, I have been working on the layout of a ring oscillator whose frequency is controlled by variable capacitances, when performing LVS everything matches but when extracting parasitic I have this error message, does anyone know the reason?, I attach images of the cap_var, thanks.
I don't know the answer to this problem, but try to make your vss and vdd tracks nearer the transistors, or even above it. If needed, use other metal layers also. There will be a noticeable voltage drop, as the current is relatively large. You should also use a differential oscillator, because the varactor capacitance is very non-linear. Differential mode is a bit better.
@Antony Brayan Sanabria Calderón: There is a line in the nwell layer near the top of the varactor that makes me think that there might be cells overlapping in a way that is difficult to see. If you post or send me the layout file, I can figure out what it is. Or post a screenshot of the layout when all subcells are unexpanded.
Hi Tim, thanks for your help, I attach the layout files, the line you mention at the top of the varactor is a small addition of Nwell to the varactor, because when exporting to magic initially showed a DRC error related to lack of NWELL, I appreciate your help.
@Antony Brayan Sanabria Calderón: I believe that the problem is that you did not flatten the circuit before running extraction. "extresist" still does not work well with hierarchy. So while hierarchical extraction works great for LVS and basic simulation, if you want to do a full R-C extraction, use:
Copy code
load ring_oscillator
flatten ring_oscillator_flat
load ring_oscillator_flat
cellname delete ring_oscillator
cellname rename ring_oscillator_flat ring_oscillator
extract all
ext2sim labels on
extresist tolerance 10
ext2spice lvs
ext2spice cthresh 1
ext2spice extresist on
You don't want to save the flattened circuit, just use it for the full R-C extraction to get a netlist you can simulate with.
Hi Tim, I have tried your solution but I still get the same result in the extraction
@Antony Brayan Sanabria Calderón: What version of magic are you using?