+ Post New Thread
Results 1 to 8 of 8
  1. #1
    Full Member level 3
    Points: 1,063, Level: 7

    Join Date
    Jul 2016
    Posts
    161
    Helped
    2 / 2
    Points
    1,063
    Level
    7

    Simples current mirror input impedance

    Hi everyone,

    How can we decrease further the input impedance of a simple current mirror? Has anyone tried to do this?

    BR.

    •   AltAdvertisment

        
       

  2. #2
    Super Moderator
    Points: 245,634, Level: 100
    Awards:
    1st Helpful Member

    Join Date
    Jan 2008
    Location
    Bochum, Germany
    Posts
    42,678
    Helped
    13006 / 13006
    Points
    245,634
    Level
    100

    Re: Simples current mirror input impedance

    Are you talking about BJT or MOS current mirror? The input impedance of a "simple" (two transistor) BJT current mirror is the diode impedance Vt/id, you can't change it. The MOS input impedance depends on Vth and Idss and can be adjusted by varying the transistor geometry.



    •   AltAdvertisment

        
       

  3. #3
    Advanced Member level 2
    Points: 4,992, Level: 16
    Dominik Przyborowski's Avatar
    Join Date
    Jun 2013
    Location
    Norway
    Posts
    667
    Helped
    281 / 281
    Points
    4,992
    Level
    16

    Re: Simples current mirror input impedance

    In general simple current mirror has always input impedance equal to 1/gm at dc. If you want to decrease it by some serious factor you will need to apply negative feedback loop to cover input node. However, it would not be a simple current mirror anymore



    •   AltAdvertisment

        
       

  4. #4
    Full Member level 3
    Points: 1,063, Level: 7

    Join Date
    Jul 2016
    Posts
    161
    Helped
    2 / 2
    Points
    1,063
    Level
    7

    Re: Simples current mirror input impedance

    Thank you all for the reply.

    I am talking about a simple current mirror using mosfets.

    Dominik, you are right and it was the first thing that came in to my mind. However no idea how this can be done.

    Do you have tip?



  5. #5
    Advanced Member level 2
    Points: 4,992, Level: 16
    Dominik Przyborowski's Avatar
    Join Date
    Jun 2013
    Location
    Norway
    Posts
    667
    Helped
    281 / 281
    Points
    4,992
    Level
    16

    Re: Simples current mirror input impedance

    The comprehensive solution is described here.

    However, the simplest solution is to provide a current buffer (CG configuration) on the input. Check how high swing cascode CM and/or flipped source follower looks like.



    •   AltAdvertisment

        
       

  6. #6
    Full Member level 3
    Points: 1,063, Level: 7

    Join Date
    Jul 2016
    Posts
    161
    Helped
    2 / 2
    Points
    1,063
    Level
    7

    Re: Simples current mirror input impedance

    Thanks for the link.

    In regards to your suggestion, are you suggesting to check these on the paper or in another place?

    - - - Updated - - -

    I am just asking because on that link I can't see anything related to decreasing the input impedance using feedback. Just normal cascodes, etc.



  7. #7
    Advanced Member level 2
    Points: 4,992, Level: 16
    Dominik Przyborowski's Avatar
    Join Date
    Jun 2013
    Location
    Norway
    Posts
    667
    Helped
    281 / 281
    Points
    4,992
    Level
    16

    Re: Simples current mirror input impedance

    The high swing cascode biasing is the same as flipped voltage follower which is nothing more than applying negative feedback loop with current buffer - as in attached scheme:
    Click image for larger version. 

Name:	bitmap.png 
Views:	1 
Size:	6.4 KB 
ID:	147816


    In the linked article are almost all known current mirror configurations and in one section the configuration with decreased input impedance are discussed.



  8. #8
    Full Member level 3
    Points: 1,063, Level: 7

    Join Date
    Jul 2016
    Posts
    161
    Helped
    2 / 2
    Points
    1,063
    Level
    7

    Re: Simples current mirror input impedance

    Thank you Dominik, very helpful. The thing is that I don't have access to that paper through Elsevier website. I have tried to find it on ieeexplore and they don't have it.
    Any idea how to get it?



--[[ ]]--