+ Post New Thread
Results 1 to 6 of 6
  1. #1
    Advanced Member level 4
    Points: 6,013, Level: 18
    Achievements:
    7 years registered

    Join Date
    Jan 2012
    Location
    Italy
    Posts
    1,059
    Helped
    41 / 41
    Points
    6,013
    Level
    18

    Fully differential opamp is inverting or non inverting amplifier ?

    Hello,

    attached the picture of the fully differential closed loop feedback, which one of the configuration should be the standard one ? many authors refer to config 1 and mostly to config 2, while the first circuit provides inverting differerntial output to the input and the latter is non-inverting,

    Click image for larger version. 

Name:	New Doc 13.jpg 
Views:	8 
Size:	209.7 KB 
ID:	157117

    •   AltAdvertisement

        
       

  2. #2
    Advanced Member level 3
    Points: 6,893, Level: 19

    Join Date
    Feb 2014
    Posts
    987
    Helped
    338 / 338
    Points
    6,893
    Level
    19

    Re: Fully differential opamp is inverting or non inverting amplifier ?

    Its inverting. 2 is correct.

    Get familiar with a simulation tool to look at this stuff. LTSpice is free and has many differential amplifiers in its library (LTC6362).


    1 members found this post helpful.

    •   AltAdvertisement

        
       

  3. #3
    Advanced Member level 3
    Points: 10,320, Level: 24

    Join Date
    Nov 2001
    Location
    US
    Posts
    897
    Helped
    312 / 312
    Points
    10,320
    Level
    24

    Re: Fully differential opamp is inverting or non inverting amplifier ?

    This question is actually totally irrelevant. You have differential outputs and the signal is the difference between the voltages of the two single ended outputs. You simply designate one of them as Out+ and the other as Out-. Then the output differential signal is (Out+)-(Out-). If you want the differential output to be 180deg out of phase with respect to input and the output difference is like that, then good, you have it. If the output difference is in phase with input, then you just designate the output you called previously Out- as Out+ now and vice versa.
    What is important is not which one is + and - but what is the polarity of the two feedback loops - they have to be negative feedbacks. And they are in both figures you posted.


    1 members found this post helpful.

    •   AltAdvertisement

        
       

  4. #4
    Advanced Member level 4
    Points: 6,013, Level: 18
    Achievements:
    7 years registered

    Join Date
    Jan 2012
    Location
    Italy
    Posts
    1,059
    Helped
    41 / 41
    Points
    6,013
    Level
    18

    Re: Fully differential opamp is inverting or non inverting amplifier ?

    Thank you friends

    I just wanted to follow the standard, Gray and Jakob they refer to it as inverting ampliifer, other authors deal with it as a non inverting amplifier,
    any way as Suta expalined the important thing is to assure the negative feedback connection.

    Here I have an importnat question, can I consider the differential ampliifer as two inverting amplifier (the first inverter on the vin+ and the other on vin-),
    why this should be important ?

    because we know that inverting ampliifer is not suffering fron input common mode voltage variation, thus generally the inverting amplifier are preferable in case to obtain high CMRR.
    if the fully differential amplifier is a composition of equivelant two inverting amplifier then it should also has the same charasterictics of the inverting amplifier regarding the CMRR

    Thanks



  5. #5
    Super Moderator
    Points: 264,377, Level: 100
    Awards:
    1st Helpful Member

    Join Date
    Jan 2008
    Location
    Bochum, Germany
    Posts
    46,193
    Helped
    14049 / 14049
    Points
    264,377
    Level
    100

    Re: Fully differential opamp is inverting or non inverting amplifier ?

    we know that inverting amplifier is not suffering fron input common mode voltage variation
    Due to the trivial fact that the non-inverting input is tied to ground respectively fixed voltage.

    thus generally the inverting amplifier are preferable in case to obtain high CMRR.
    CMRR is a property of the OP or generally a differential amplifier, it can't be determined for a circuit that isn't exposed to common mode voltage. In so far the statement is wrong.

    if the fully differential amplifier is a composition of equivelant two inverting amplifier then it should also has the same charasterictics of the inverting amplifier regarding the CMRR
    No it isn't.


    1 members found this post helpful.

    •   AltAdvertisement

        
       

  6. #6
    Advanced Member level 3
    Points: 10,320, Level: 24

    Join Date
    Nov 2001
    Location
    US
    Posts
    897
    Helped
    312 / 312
    Points
    10,320
    Level
    24

    Re: Fully differential opamp is inverting or non inverting amplifier ?

    Quote Originally Posted by Junus2012 View Post


    Here I have an importnat question, can I consider the differential ampliifer as two inverting amplifier (the first inverter on the vin+ and the other on vin-),
    why this should be important ?

    because we know that inverting ampliifer is not suffering fron input common mode voltage variation, thus generally the inverting amplifier are preferable in case to obtain high CMRR.
    if the fully differential amplifier is a composition of equivelant two inverting amplifier then it should also has the same charasterictics of the inverting amplifier regarding the CMRR

    Thanks
    I think you can simplistically think of it as two symmetric inverting amplifiers, like a mental model. Yes, as usual, even for fully differential amplifiers you will have a virtual short at the input of the OTA (or opamp). It is a differential short, but the common mode is whatever you design for, not like in the single-ended case where usually the + input is connected to gnd. It might happen that in the fully differential case you have virtual differential short for the inputs but still varying input common-mode voltage and in this case you need a good CMRR. CMRR is better by definition for fully differential circuits but any asymmetries, being systematic or random will spoil it and reduce it.


    1 members found this post helpful.

--[[ ]]--