TINA/Spice/LMH6553: How to test CMRR of LMH6553?

Thank you very much for your revising. It is very helpful to me. However, (1) I still wonder that why its CMRR changes a little bit when I adjust the value of the resistors. (2) If I use Vcm/Vos, what is difference compared to your method? Since I use this way, I also can get around 180dB. The reason I want to find CMRR for this circuit because I want to find the precision mismatched resistors. It will affect a lot to CMRR with these discrete components.

Thanks again for your help.

Best regards,
Hieu

Hieu,

I would need to know the details of how you are changing the resistors before we can answer your question (1).

The method I am using estimates the common-mode voltage at the device input pins and divides it by the differential offset voltage at the device input pins.

That is because CMRR is often defined as an additional differential offset voltage that is caused by a common-mode voltage.

I hope this helps.

Please let me know if you have any more questions.

Regards,
John

Hi John,

Thank you very much for your reply.

"My question 1 is..."

The detail of changing the values of resistors is attached in this reply along with its CMRR is simulated. 

Another issue I would like to ask

I wonder that why the CMRR of the LMH6553 is 82 dB in its datasheet. However, we can get around 180 dB in this simulation. I do not know why?

Again, thank you very much for spending your time to reply to me.

Best regards,

Hieu

Hieu

The text header in the model netlist gives a list of specs the model is supposed to meet.
CMRR is one of the specs.

It has been through a few revisions since its was first released in 2008.
One of the revisions may have affected CMRR.

Regards,
John

Dear John,

Hieu,
Please attach the latest circuit that gives the results you mentioned and I will take a look at it.
Regards,
John

Dear John,

Please find circuit attached.

I changed resistor ratio to let them have a mismatch ratio. But CMRR is always higher than 100 dB. I don't know why.

I use ideal opamp to simulate CMRR in spectre, with resistor mismatched, CMRR only can get ~66 dB.

Could you please help to check this circuit revised by you in the previous?

Thank you very much for your help.

LMH6553 AC CMRR - rev TI.TSCBest regards,

Hieu

Hieu,

Apologies, but I did not at first notice what kind of device this is.
It may be that some of what we are seeing is because this is a current-feedback device.
In other words, it amplifies input pin current instead of the input pin voltage.

You can see this by looking at Figure 37 and 38.
Instead of open-loop voltage gain, they show the Open Loop Transimpedance, which is defined as d(Vout)/d(Iinput).
There are also some comments on page 19 of the data sheet.

I will continue to look at the CMRR and will reply soon to this thread with my findings.
Thank you for your patience.

Regards,
John

Dear John,

I want to find CMRR for this opamp LMH6553, then find CMRR for the whole system which is included many opamps.

At first, I try to find CMRR for LMH6553 without resistor mismatch then find CMRR with resistor mismatch. 

As the circuit attached to your mail, without resistor mismatch, CMRR is roughly 187 dB, while with resistor mismatch, CMRR is roughly 110 dB.

With resistor mismatch, as I understand CMRR cannot reach to 110 dB. 

Therefore, I need your help to check this circuit. What's wrong with it?

Thank you very much.

Best regards,

Hieu

Thank you very much for your help.

Did you use the same circuit architecture given to you to test or you now build a new test?

I would like to know that architecture to test other chips. Could you send me the circuit to test CMRR if you don't mind?

Thanks for your help again.

Best,

Hieu

Thanks for your revising circuit.

Please let me know when you have a new revision model.

Best,

Hieu