CMI-2426 warning inDC analysis

dcOpInfo: writing operating point information to rawfile.

********************************
DC Analysis `dc': Vcm = (0 -> 1)
********************************

Warning from spectre during DC analysis `dc'.
WARNING (CMI-2426): I5.M3.m1: `Cdsc' = -416.981 mF/m^2 is negative.
WARNING (CMI-2424): I5.M3.m1: `Acde' = 46.6148 mm/V is unusually small.
WARNING (CMI-2424): I5.M7.m1: `Acde' = 46.6148 mm/V is unusually small.
WARNING (CMI-2426): I5.M11.m1: `Cdsc' = -198.973 mF/m^2 is negative.
WARNING (CMI-2424): I5.M11.m1: `Acde' = 41.1951 mm/V is unusually small.
WARNING (CMI-2426): I5.M9.m1: `Cdsc' = -198.973 mF/m^2 is negative.
WARNING (CMI-2424): I5.M9.m1: `Acde' = 41.1951 mm/V is unusually small.

The properties of these transistors (I5.M3, etc) are fixed or some variables are used? Maybe some model parameters are not calculated property from dimensions?

Dominik Przyborowski said:

The properties of these transistors (I5.M3, etc) are fixed or some variables are used? Maybe some model parameters are not calculated property from dimensions?

Click to expand...

The device properties are fixed. Any other hints?

Dominik Przyborowski said:

The properties of these transistors (I5.M3, etc) are fixed or some variables are used? Maybe some model parameters are not calculated property from dimensions?

Click to expand...

The dimension is OK. It is a 65nm process, anyone met the same warning before?

Maybe You exceess the maximum dimensions? If your process has halo implants, the maximum length of RF fets is limited (240nm for example).
On cadence forum is thread about this warning for mimcap, exceeding maximum width.

Dominik Przyborowski said:

Maybe You exceess the maximum dimensions? If your process has halo implants, the maximum length of RF fets is limited (240nm for example).
On cadence forum is thread about this warning for mimcap, exceeding maximum width.

Click to expand...

It is not a RF mosfet. I use a standard general purpose MOSFET and it is not a large transistor, (0.2/0.065)x4.
I also doubt the initial conditions of some MOSFET might cause such warning, and I run the trans before DC, the warning is still there.
Any other hints or ideal that I could try to solve such warning?