CMRR simulation setup

Dear wgtkan,

A few items came to mind when I read your post and saw your comments on your simulation result. 

Shawn 

• What is it specifically about the CMRR response that "doesn't seem right"?
• What is the DC value of your voltage Vcm and does it keep your DC operating point of the op-amp in its high gain region of operation?
• Your circuit does not appear to have a load impedance. The load impedance can impact the frequency response of the amplifier and its stability. As a result, the CMMR you are observing may not be an accurate estimate of the CMRR when used in your application.

Hello Shawn,

1. The last plot seems correct but the earlier one was not right. 

2. The DC value of the voltage VCm is 0.9, yes it keeps the operating point of the Op Amp in a saturation region.

3. I did not place a load impedance. I will connect a load impedance and rerun the simulation.

I wanted to run the CMRR simulation using XF analysis as Andrew mentioned but It didn't work for me.

Thank you for your response. 

Dear wgtkan,

Thank you for your added comments - I just wanted to verify that the DC operating point was in the high gain region!

wgtkan said:

. The last plot seems correct but the earlier one was not right. 

My interpretation of your last plot is that it represents the differential gain of your op-amp versus frequency while the former curve represents the CMRR versus frequency. Am I correct?

A concern I have is that it appears you are simulating your op-amp in an open-loop configuration (i.e., without any feedback network). As a result, I am concerned that your CMRR and gain curves are not representative of what you are expecting. If you are intending to use this amplifier in a closed loop configuration, in addition to including the load, I might suggest including the feedback network and then simulating the differential gain and CMRR. I don't think the CMRR curve you generated is meaningful in an open loop configuration to easily extract the CMRR in any given closed loop configuration. If your interests are in the loop gain of your amplifier, I might suggest then adding the appropriate probes and use a stability analysis (differential) in its closed loop configuration rather than running an XF analysis in its open-loop configuration. Does this make sense wgtkan?

Shawn

Hello Shawn,

 My interpretation of your last plot is that it represents the differential gain of your op-amp versus frequency while the former curve represents the CMRR versus frequency. Am I correct?  

 No. It is representing the CMRR response against frequency. I understand your concern and I saw literature of the recommended means of simulating an opamp's CMRR and followed their method. I am attaching one such useful paper here.https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1375058In some popular textbooks also the load is not added.   

 I have done a stability analysis to determine the open-loop magnitude and phase response and checked my result by performing other open-loop gain simulation by adding high-impedance feedback components.   

Thank you always, Shawn. 

Dear wgtkan,

wgtkan said:

No. It is representing the CMRR response against frequency.

Thank you for the clarification. I was confused.

wgtkan said:

I understand your concern and I saw literature of the recommended means of simulating an opamp's CMRR and followed their method. I am attaching one such useful paper here

Thank you, wgtkan, for the reference! I took a look at it using my IEEE account (without an account, one cannot view the paper). I did notice that  all of the three amplifier configurations the authors propose for measuring CMRR do include the feedback network (albeit different feedback configurations).

wgtkan said:

and followed their method.

Great! Hence, I think you are on the right track to providing a good estimate of the CMRR.

wgtkan said:

I have done a stability analysis to determine the open-loop magnitude and phase response and checked my result by performing other open-loop gain simulation by adding high-impedance feedback components.   

I hope the two sets of results provided some assurance of your simulation methodologies for stability. Excellent! Thank you, very much, for letting us know your results and conclusions wgtkan!

Shawn