DC operating point Vs Transient analysis - SPECTRE

[yagi]

dc operating point analysis

Hello,
I am designing a Fully differential two stage opamp with CMFB. It is being used as inverting amplifier Configuration (gain = -R2/R1). The design is for resistor based programmable gain amplfier with gain programmable from -11dB to 1dB. I am using SPECTRE for simulations.

When I do a DC operating point analysis on the PGA with the opamp, for gain setting above -3dB, the dc operating voltages look good and match my hand calculations. But for the gain settings below -3dB the Dc operating analysis shows that output common-mode of the PGA is going to ~0V. BUT the transient analysis with same DC voltage bias with "skipdc=yes" option, looks good and voltages at all nodes match my hand calculations.

Can I believe transient analysis results with "skipdc=yes" option, eventhough my DC operating point analysis is failing.

Is there anyway to reproduce transient analysis results in DC operating analysis for cases where DC operating point analysis is failing and Transient analysis are passing.

Please help me.

Thanks,
Yagi

 

[liuyonggen_1]

spectre transient operating point

maybe, the feedback resistance influence the DC operating point. what's the resistance for the feedback resistors?

 

[DenisMark]

what is transient operating point

It's not enough information to undestand ur problem.
I can make 3 guesses:
1) gmin can affect on results of DC simulation. It can have value 1e-6 as default. Check it and set more biger (e.g. 1e-9).
2) Can u show ur test bench, may be the problem in this.
3) Check ur design and operation points of transistor.