Ideal OpAmp using VCVS, output common-mode issue

[Javad1991]

Hi,

I need an ideal differential OpAmp/OTA to simulate Active-RC integrators in cadence.

I have read the previous threads in the forum about this topic. I know that I can use a very high-gain VCVS and emulate the poles using RC.

The input to the system is a sine wave with an amplitude of 1 mV and a DC level of 0.5 v. The problem is, the DC level at the output is also multiplied by gain.

Can anybody give me a suggestion on how to fix the output common-mode in VDD/2? I probably need a CMFB, but I don’t know how or where to connect it.

 

[barry]

I don’t know Cadence, but any opamp, ideal or real, will multiply the DC input by the gain. That’s the way they’re SUPPOSED to work. If you don’t want that, then AC couple the input, i.e., use a capacitor.

 

[crutschow]

Add a series capacitor with a resistor to ground to establish a ground (0V) reference at the input.
The RC low frequency roll-off must be below your lowest frequency of interest, of course.

 

[Dominik Przyborowski]

Show us your schematic.

 

[FvM]

I probably need a CMFB, but I don’t know how or where to connect it.

Yes, you need it. Review an analog design textbook to understand how CFMB works and how it is usually implemented.

 

[barry]

Yes, you need it. Review an analog design textbook to understand how CFMB works and how it is usually implemented.

it’s not clear that the DC offset is common-mode. CMFB will do nothing to compensate if it’s not.

 

[FvM]

Question title says "CM issue", we should believe in lack of complete information.

 

[dick_freebird]

Any DC pedestal will and should be amplified, and CMFB
is for differential-output amplifiers (not to "correct" an
output that is actually correct to begin with).

An answer you don't like, is not a wrong answer necessarily.

 

[Javad1991]

Thanks to you guys for all your suggestions.

Actually, I wanted to have the same DC level (Vdd/2) for input and output. Based on @barry suggestion, I removed the DC component of the input by a series capacitor before amplification using VCVS, and then after amplification I added a DC component to the output using a DC source. The result is exactly what I wanted


P.S: I tried CMFB but I didn't get the proper result.

 

[arthorios]

I don't know what the actual implementation is, but blocking the input CM with a capacitor doesn't seem practical since you need to provide the input CM to the OTA (e.g. for a gm-C filter). In any case (active RC or gm-C), you will end with a high-pass behavior with such a DC blocking cap.

There are many ways of doing what (I assumed) you need. The output CM problem can be set by forcing the output with an ideal voltage source and a pair of resistors. For such an ideal model, you don't need a CMFB.

You can see the examples in Cadence below for both OTA and OPAMP (assuming fully differential is needed). These are just examples, don't take them as a rule. I hope this helps. Also, as @Dominik Przyborowski pointed out, you should show the schematics of what you need/have so that people can understand your problem.