+ Post New Thread
Results 1 to 2 of 2
  1. #1
    Junior Member level 3
    Points: 512, Level: 4

    Join Date
    Jun 2017
    Posts
    29
    Helped
    0 / 0
    Points
    512
    Level
    4

    quadrature ring oscillator 5GHz

    Hello,
    I would like to design a quadrature ring oscillator at 5GHz.. Could someone suggest me which topology will be a good starting point? and some interesting papers that I could read for better understanding

    Thanks in advance!!

    •   AltAdvertisement

        
       

  2. #2
    Advanced Member level 5
    Points: 38,992, Level: 48

    Join Date
    Mar 2008
    Location
    USA
    Posts
    6,302
    Helped
    1836 / 1836
    Points
    38,992
    Level
    48

    Re: quadrature ring oscillator 5GHz

    Easiest way to a quadrature oscillator is 4 stages of fully
    differential gain, with three of them cross-connected (so
    as to be inverting) and one straight (noninverting).

    But you need to think about what you want besides
    5GHz small signal oscillation. Presumably these quadrature
    clocks have a purpose and that purpose wants "stuff" -
    like drive strength, amplitude / logic levels, phase accuracy....

    If you have to get to bang-bang square-looking CMOS
    logic clock field, that needs downstream gain and some
    efforts to make switching symmetric (5GHz out of a logic
    inverter, with symmetric rise / fall times which consume
    each 10% or less of the period, wants a pretty short gate
    length and it's all going to go to hell if you have to throw
    those signals across random routed logic for a millimeter,
    etc.).

    General purpose is probably not where this is headed, but
    whatever the purpose wants, you need to make clear if
    you want to get there by other than "random walk".


    1 members found this post helpful.

--[[ ]]--