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.

Why we need to use CMFB in a differential amplifier?

Status
Not open for further replies.

JiL0

Member level 1
Member level 1
Joined
Jan 29, 2004
Messages
36
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,286
Activity points
444
CMFB

Hope someone can help. Designing an differential amplifier.

I dont really understand the problem that necessitate the use of CMFB circuit and what problem will arrise without the CMFB circuit. Does it mean that if we dont have CMFB, the output voltage will have a DC error? Does this problem happen only when we are using the amplifier in feedback?

I understand from reading razavi's book that the problem arrises from the error mismatch between the p-type current source and the n-type current source (does this error arrise due to the differenct input voltage or issit due to fabrication prob?). THis causes voltage error at the output. By using CMFB, we sense this error and minus this error from the input. And this is done by re-adjusting one of the current source so as this will not be any mismatch. Is my understanding correct?
 

Re: CMFB

Since in OTA design the output resistance of MOSFET is very high, the output common-mode voltage is extremely sensitive to mismatches between n- and p-MOS current source. Only very little mismatches can caused the output common mode voltage go all the way up or down to the rail, forceing one of the output transistors in troide region and thus the gain drop significantly. In single-ended circuits, the output voltage can be stablize by external feedback loops (such as resistors and caps), however in fully differential circuits, external feedback can only stablize differential mode voltage but not common mode voltage, thus required extra common mode feedback circuit to stablize the output common mode. Without this circuit the fully differential OTA bascially cannot work properly.
 

Re: CMFB

Can this error be tested using simulation software like cadence? Or is this problem only occuring on the actual circuit implementation?
 

Re: CMFB

This type of error can also simulated. If you do not included CMFB circuit, you will found that the output node of OTA is always neared one of the supply rail, and one of the output transistors always in troide region. Also, the output node sensitivity is very high to mosfet ratio errors since no feedback is applied.
 

CMFB

Just do a .tran analysis, add the two output and divided it by 2, you will see the variation of common-mode voltage.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top