• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 62 subscribers
  • Views 216 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LM5163: Output Ripple at very light loads

Alex L
Prodigy 100 points
Part Number: LM5163
Other Parts Discussed in Thread: LMR38010

Hello

My understanding is that LM5163 (COT regulator) with a well calibrated type 3 feedback network does work well (with very little Output Voltage ripple, i.e. <2mV) when it regulates the output in CCM (Continuous Conduction Mode).

With very light loads (e.g. 10-20mA), LM5163 operates in DCM (Discontinuous Conduction Mode),
and indeed calculation sheet for LM5163 specifies the minimum Full Load Current at 0,1A.

Simulation with TI's model for LM5163 shows a high output voltage ripple (i.e. >5mV) at I_load < 30mA

My questions are these:

1) Is there a way to get low voltage ripple also for light loads (I_load<30mA) using LM5163 ?

2) What does dictate the threshold between CCM  and DCM ?

3) Is there any COT regulator with fixed frequency also for very light loads (or down to I_load=0mA) ?

4) Have I got to transition to more 'advanced' architectures, such as Peak Current Mode (e.g. LMR38010),
    and have I got to choose the fixed frequency version (LMR38010F)  to achieve small output ripple voltage also at very light loads?

Thank you very much in advance
Best regards

  • 0
    TI__Mastermind 23380 points

    Hi Alex

    1) Unfortunately there is no easy way to get the LM5163 to operate in CCM at super light-load, unless you use a larger inductor value. In that, you can lower the load current a bit more until it is less than half the inductor peak-peak ripple current. If you want CCM at light load, you will want something like the LMR38010 which has FPWM operation.

    2) The threshold between CCM and DCM is when the load current dips below half the inductor peak-peak ripple current. The part is then operating in DCM. This is because the inductor current actually hits 0A, so the inductor current is discontinuous.

    3) Unfortunately, there is no COT regulator with fixed frequency as that goes against the primary control of COT, since the COT device always delivers a fixed on-time pulse, but the off-time varies. 

    4) Yes that seems like the best option for FPWM operation. It is not as complicated as it appears because the compensation is internal, so really the component selection is simpler with less variables.

    Thanks,

    Richard

  • 0 in reply to Richard Fu22
    Prodigy 100 points

    Hi Richard

    Unfortunately, I can not choose a larger inductor value (it's already >100uH) without sacrificing 1)space or 2)full load current

    Additionally, switches overcurrent threshold would be higher than inductor saturation, and that is not a good thing to have, from what I understand. Is this right?

    Here are my questions:

    5) Is DCM (as compared to CCM) always detrimental to Output Voltage ripple ?
         I suppose this is true.

    6) Is LM5163 operating in DEM (diode emulation mode) for light loads ?
        This is intrinsic to COT, from what I understand, right ? (low side switch remains ON until the output capacitor begins to discharge)

    7) Can I fine tune the type 3 network to have lower Output Voltage ripple ?
        (e.g. increasing generated ripple voltage - cons and pros ?)

    8) Is FPWM (that by definition would be CCM) characterized by the lowest Output Voltage ripple ever possible ?
        Therefore, is LMR38050F the best way to go in terms of Output Voltage ripple ?
        What FPWM regulators does TI offer for voltages up to 70V-80V ?

    Thank you very much in advance

    Alex

  • +1 in reply to Alex L
    TI__Mastermind 23380 points

    Hi Alex,

    Yes that's correct. You would want the inductor saturation current to be higher than the switch current limit.

    1) Yes, DCM operation will generate higher output voltage ripple than CCM. 

    2) Yes, that's correct- the part will operate in DEM for light loads.

    3) yes, you can tune the type 3 network to have lower output voltage ripple. The main con is increased part count. 

    4) FPWM would more than likely be your best choice for getting the lowest output ripple with minimal additional design effort. I would recommend the LMR38010/20F for your application.

    Cheers,

    Richard