• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 64 subscribers
  • Views 1307 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TPS51220A: High no load power consumption related to Rx

Nate Enos
Genius 9430 points
Part Number: TPS51220A

Hi Team,

Posting for a colleague. We are seeing excess power consumption into the TPS51220A. Currently on some of the boards we see 11mA into the TPS51220A (18V input), while others are at 5.8mA. If Rx (R14 in the typical application circuit) is reduced from 8.2kohm to 6.2kohm, then this fixes the issue.

Can you explain why the filter for the current sense amplifier could have this much of an effect on the power consumption at no load? There is no current going into the sense amplifier.

Thanks,

Nate

  • 0
    TI__Mastermind 18510 points
    Hi Nate,
    We havent seen the issue like this before, Amplifier is high-impendence, should be almost no power consumption, and it's weird if the amplifier sink current, the power cusumption should get bigger after RX changed from 8.2K to 6.2K, because 6.2K is samller which can make the leacakge bigger, right?
    By the way, you said some boards are 11mA, Did you exchange the chip to do borad to board compare to check the issue follow IC or board?
    and also i think you can check the SW/Vout waveform to compare before/after change Rx.

    Regards,
    Yuchang
  • 0 in reply to Yuchang Zhang
    Prodigy 50 points
    Hi Nate,Yuchang,
    Updates on the issue.
    We observed efficiency dropping in light load operation when we use RX=8.2K, RC=3.74K. We are using these values in our current design.
    We observed up to 11mA from 18V input (198mW) when our system is powered off (ie TI controller is running on light load). This is more than we expected. Later we found the values on our current design is not optimum to meet the equations on Page 32 of datasheet.
    The equations are
    Equation(1) : Cx * (Rx||Rc) = Lx/Rs.
    Equation (2) :Iocl = Vocl x (1/RS) x (Rx+Rc)/Rc
    Where Lx= Output Inductance, Cx=Current Sensing Filter cap, Vocl = 60mV or 30mV (We are using 60mV Selection),Iocl = Peak Current Limit .
    In our circuit we are using Lx=3.3uH,Rs of Lx is 0.027Ohm(TYP),Cx=0.1uF.
    The Iocl for our design is ~ 7A.(Isat of Inductor =7.3A).

    We solved the equation we found RX=3.9K,RC=1.78K gives better results on light-load operation efficiency. It was taking only 2.5mA from 18V input.
    Please let us know whether we can choose RX=3.9K,RC=1.78K for our application.
    Are you foreseeing any issues with these values?
    Is there any impact expected during normal/high-load operations? Is there anything else we need to consider before choosing this values? Let us now your expert opinion on these
  • 0 in reply to BIJU ERAYAMKOTPANAYAM
    TI__Mastermind 18510 points

    Hi,

    For this calculation setting, I think RC match and OC setting are OK.

    But as we know, there is always some offset between calcution and real system, so I suggest double confirm the OC point and Vout (transient/ripple...) by testing.

    Regards

    Yuchang

  • 0 in reply to Yuchang Zhang
    Prodigy 50 points
    Thanks ..I tested the new values on board and found the power-off mode current reduced a lot.
    Another thing noticed is the maximum output current is bit reduced. With new values,the system loses the voltage regulation at load current 5.2A. It was 5.4A with original Resistor network values
  • 0 in reply to BIJU ERAYAMKOTPANAYAM
    TI__Mastermind 18510 points

    Hi,

    What do you mean about "loses the voltage regulation at load current 5.2A"? what's your system max loading requirement?

    The oc point is set around 7A, Have you checked if this is right?

    and by the way, check load transient and ripple is more important.