+ Post New Thread
Results 1 to 4 of 4
  1. #1
    Advanced Member level 3
    Points: 3,812, Level: 14

    Join Date
    Jun 2011
    Posts
    839
    Helped
    17 / 17
    Points
    3,812
    Level
    14

    boost converter in light load

    Hello,

    Please can you confirm that if a boost converter is correctly compensated for max load and it is in ccm at max load.

    Then it will automatically be stable when it goes into DCM due to being lightly loaded?

    •   Alt8th February 2012, 10:14

      advertising

        
       

  2. #2
    Advanced Member level 5
    Points: 19,696, Level: 34
    Achievements:
    7 years registered

    Join Date
    Jan 2011
    Posts
    3,269
    Helped
    1160 / 1160
    Points
    19,696
    Level
    34

    Re: boost converter in light load

    In general yes, I'd expect it to remain stable. The RHP zero frequency should increase at light load, and as your controller starts operating in DCM, its gain should decrease. Those two things should only increase stability (unless the system was only conditionally stable). The only thing it could worsen is the damping of your LC, which might cause some extra phase lag, but it's unlikely to be enough to overcome the loss of gain and cause instability.

    Is this still the same issue as in your other thread? Based on what you explained over there, it sounded like it must be a relaxation oscillation, not a small signal instability. Did you notice that the amplitude or frequency of the ripple changes depending on the load impedance?



    •   Alt8th February 2012, 14:57

      advertising

        
       

  3. #3
    Advanced Member level 3
    Points: 3,812, Level: 14

    Join Date
    Jun 2011
    Posts
    839
    Helped
    17 / 17
    Points
    3,812
    Level
    14

    Re: boost converter in light load

    yes its the same issue as the TPS40210 problem.

    I thought in current mode the resonance in the LC never occurred because the inductor is being controlled, and its pole is thus removed from the transfer function......?

    .....or is it that when slope sompensation is involved, one is not fully in current mode, and so therfore the inductor's pole raises its ugly head, and resonates with the output capacitor to give instability?



    •   Alt9th February 2012, 10:26

      advertising

        
       

  4. #4
    Advanced Member level 5
    Points: 19,696, Level: 34
    Achievements:
    7 years registered

    Join Date
    Jan 2011
    Posts
    3,269
    Helped
    1160 / 1160
    Points
    19,696
    Level
    34

    Re: boost converter in light load

    Right, I forgot that it was a current mode controller. Yes, it theoretically eliminates the pole caused by the inductor. In practice, it just moves it to a much higher frequency (I'm not clear on the math). But that can't be your problem, since your oscillation frequency is ~5KHz, which is far below the poles in your converter. There should be negligible small signal phase shift at such low frequencies.



--[[ ]]--