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.

CMFB, Continuously or Discrete?

Status
Not open for further replies.

nxing

Advanced Member level 1
Advanced Member level 1
Joined
May 10, 2004
Messages
421
Helped
25
Reputation
50
Reaction score
10
Trophy points
1,298
Location
China
Activity points
2,856
Hi Guys,

I am designing a differential Opamp for my Pipelined ADC. Can anybody tell me should I use continuously or discrete CMFB circuit if both power dissipation and layout area are considered? Why?

I also tried the continuously one (in Martin's book "analog integrated circuit design") with feedback to control the output stages bias current, but it didn't work (my opamp is a gain-boosting folded-cascode). Does anyone has any suggestion that I should pay attention to?

Thanks
 

If your application is switched capacitor, use a discrete CMFB because, it saves power and do not have stability problems in the CMFB network. In same cases you also reduce area, by using small capacitors. However discrete time or continuous time, both work in high gain amplifiers. I have used both for 110 dB DC gain.

Bastos
 

If you have SC design - use SC CMFB, it is simpler to design (no stability problems), besides it does not reduce the signal swing.

Some people are using continuous CMFB even for SC circuits because they say that the OTA output would bounce the output at clock phase changes which will result in some performance degradation but I did not meet with such problems.
 

I have used SC CMFB in high speed and high resolution pipelines with no problem.

bastos
 

i prefer to use sw CMFB in pipelined ADC. i think it is much simpler and more reasonable. :)
 

Discrete would be preferred in switcap implementation when we have phase1 and phase2 of the clocks which are available or generated in any ADC .Stability or Voltage swing not an issue here as we do not use any extra transistor's or there is any sort of feedback .But I think the discrete would load the opamp's output node and which might not be desirable in high speed ADC design where output caps are small.If common mode cap size is small ,we need large driving current to the common mode cap (charging cap to common mode ).Trade off between timing ,Area and Swing .
 

Can anybody suggest me how to simulate the characteristics of Opamp after the switch-capacitor CMFB is added to the circuit?

Thanks
 

Do a tran analysis to compute the delay or settling time of the op-amp .Putting a .IC stmt ,one can get the op-amp Gain & GBW.
 

Hi Mady79,

Can you tell me what's .IC stmt?

Thanks!
 

To get a stable operating point you need to add a voltage controlled voltage source that only gives you the correct CM voltage of the amplifier but does not count for the amplifier characteristics. This is done for AC. For transient add the same but remove the VCVS after 0.1ns this will give you the correct DC voltages to start the simulation whout having to wait the SC CMFB circuit to settle.

Bastos
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top