Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

how to simualte CMRR for diffrential amp?

Status
Not open for further replies.
lm386 psrr and cmrr simulation

is there any material about how to simulate PSRR?

i dont know how to do it?even i know what PSRR means,thanks!
 
cmrr baker

see allen book, he give the testbench on how test psrr and cmrr.
 

cmrr vcvs

derekqiao said:
yaxazaa write:
you can simulate by tying the diff-pair to vcom and inject small ac signal, and measure the gain at output.

But use the manner which you have mentioned to simulate Acm, and then caculate CMRR. A very high CMRR date will be exported. Such as an example Ad=60dB, CMRR=129dB. In fact, CMRR=129dB is unreasonable.
Click to expand...

I think derekqiao is right!
 

obtain large cmrr

wow,no ideal......
 

measure cmrr

CMRR is depended from transistors matching, especially transistor matching of differential pair. That is why simple ac analisys cann't predict resonable results. It's necessary to perform monte-carlo analysis for get results close to reality. Or introduce missmatch manually in schematic (only for estimation).
 

Hi,
As far as the setup is required u can refer to the book by
Baker Li Boyce.

There they are using 2 opamps one is biased to find the closed loop gain and the other is fed with a common mode signal.

You can find CMRR = Adm/Acm
 

note that u must have different zero for examin it in labratoary
 

Hi,
I have a new understand. In the formula CMRR = Adm/Acm, the Adm is unit or 0dB. Actually, the opamp is simular with a unit gain follower when simulation CMRR, which can stabilitate the operation point of opamp.
Regards!
 

Hi,

derekqiao said:
Hi,
I have a new understand. In the formula CMRR = Adm/Acm, the Adm is unit or 0dB. Actually, the opamp is simular with a unit gain follower when simulation CMRR, which can stabilitate the operation point of opamp.
Regards!
Click to expand...

Can you clearly define your understanding.
As far as I can understand from this is that you are saying Adm is unity or 0dB.
This can only be in case of Unity voltage gain configuration. As far as for other cases the Adm of differential amplifier is quite high.

Thanks
Shaikh Sarfraz
 

Ok, I will explain it. Please refer P321~P313 of Allen's book. Because opamp has offset, we couldn't add same commom signal in inp and inn which may saturate the output voltage. One option is add another dc voltage to cancel offset, option2 is design a feedback loop and the loop ac gain is unit.
The formula CMRR = Adm/Acm is only for option2. You can use option1 to simulate CMRR directly, CMRR=delta(Vout)/delta(Vicm)
 

Hi,
You have to compensate for offset error, as the open loop gain of Op Amp is very high, and any slight difference in the input voltage can produce large voltage which is not desirable as the common mode gain of Op Amp sholud be low.
As far as the book is concened they have given 2 more techniques to find CMRR directly with the tool.

Thanks
Shaikh Sarfraz
 

[check this doc
 

Attachments

  • opampsim_good_1799.pdf
    117.2 KB · Views: 240

[the document by jonathan david was really nice, thx
 

CMRR is depended from transistors matching, especially transistor matching of differential pair. That is why simple ac analisys cann't predict resonable results. It's necessary to perform monte-carlo analysis for get results close to reality. Or introduce missmatch manually in schematic (only for estimation).
 

I can't download the tutorial.
:cry:
 

DenisMark said:
CMRR is depended from transistors matching, especially transistor matching of differential pair. That is why simple ac analisys cann't predict resonable results. It's necessary to perform monte-carlo analysis for get results close to reality. Or introduce missmatch manually in schematic (only for estimation).
Click to expand...

Does Cadence have the monte-carlo analysis function? Is the value of standard variation of mismatch strongly dependent on the porcess?
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top