• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 3 answers
  • Subscribers 63 subscribers
  • Views 808 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.

WEBENCH® Tools/TPS65400: It's SW1 is used for powering the DDR4 of a CNN compute board, there is a bad ripple on the output when the mcu run.

allan.han
Prodigy 30 points
Part Number: TPS65400

Tool/software: WEBENCH® Design Tools

 this is the ripple

this is the current waveform when position is 10ms

this is the current waveform when the position is 500us

this is the current waveform when the position is 200us

this is the current waveform when the position is 2us , and Fsw = 1.287MHz

and  this is my schematic

we think that the problem is the set of 'C238' and 'R245'，but we can't find the suitable set making the output stable,or how to caculate them.

thanks.

  • 0
    TI__Mastermind 19080 points

    Hi,

    Yes, seems the loop is not well compensated by C283 and R245. To confirm that, you can measure the waveform on COMP pin. It should be almost a DC signal when working normally.

    You can start with 15.4k+4.7nF, and change C284 to 22pF. If it doesn't work, how much loading do you have on this channel? This information is needed for the calculation.


    Regards,
    Sam

  • 0 in reply to Sam(MPAA) Zhang
    Prodigy 30 points

    Hi,

    I am very sorry for continuing the communication after a long time.  It doesn't work when set R245 to15k , C283 to 3.3nF(I haven't 4.7nF) and C284 to 22pF, and there is a large ripple on the COMP pin.

    The load on this channel  is changed all the time, and the load current waveform pictures has been put up.  On the other hand, can you teach me for calculating the  parameters of the loop, when the output load is changed on a large scale all the time. We will benefit from this skill on the other products.

    Thanks and Best regards

    Allan

  • 0 in reply to allan.han
    TI__Mastermind 19080 points
    Hi,

    Large ripple on COMP pin indicate the loop is not stable, or the layout is not good enough so that COMP pin had picked up some noise.
    For how to calculate the compensation parameters, please refer to the document below,
    www.ti.com/.../slva352a.pdf
    It's a good app note talking about the compensation for current mode device. Thanks!

    Regards,
    Sam
  • 0 in reply to Sam(MPAA) Zhang
    Prodigy 30 points

    Hi，

    What advice would you offer to us for the solution which the TPS65400's output load is is changed on a large scale all the time ?

    Thanks,

    Allan

  • 0 in reply to allan.han
    TI__Mastermind 19080 points
    Hi Allan,

    The device should have the capability to handle the always changing loading current. You can design the loop with the worst case, which is the highest loading current condition(because the load pole is in higher frequency with higher loading current).

    Regards,
    Sam