UCD90320UEVM: Regarding the operation specifications for Time ON MAX WARN

Part Number: UCD90320UEVM
Other Parts Discussed in Thread: UCD90320

We are currently evaluating the UCD90320 and would like to ask a question regarding its power-down (re-sequence) behavior when a fault is detected.

When a fault is detected during the power-up sequence, the device transitions to the power-down sequence. However, we have observed that if, during this power-down process, the voltage of a certain AMONx rail does not sufficiently decrease and a Time ON MAX WARN is triggered, the EN pin that is supposed to be de-asserted last is instead de-asserted earlier, and the re-sequence operation starts while the power-down sequence is still in progress.

In this situation, is it correct to understand that when a Time ON MAX WARN occurs, the UCD90320 effectively interrupts the ongoing power-down sequence and initiates the re-sequence operation as part of its specified behavior?

We would appreciate your confirmation and guidance on this matter. Thank you in advance for your support.

  • Hi

    It is TON_MAX Fault, if you have a fault response set on TON_MAX, then UCD will follow the response to do. 

    TON MAX fault occurs during the power up instead of power off. 

    Regards

    Yihe 

  • Yihe-san,

    Thank you for your reply.
    I apologize.

    I made a mistake in the message.

    I will resend the message.

    We are currently evaluating the UCD90320 and would like to ask a question regarding its power-down (re-sequence) behavior when a fault is detected.

    When a fault is detected during the power-up sequence, the device transitions to the power-down sequence. However, we have observed that if, during this power-down process, the voltage of a certain AMONx rail does not sufficiently decrease and a Time OFF MAX WARN is triggered, the EN pin that is supposed to be de-asserted last is instead de-asserted earlier, and the re-sequence operation starts while the power-down sequence is still in progress.

    In this situation, is it correct to understand that when a Time OFF MAX WARN occurs, the UCD90320 effectively interrupts the ongoing power-down sequence and initiates the re-sequence operation as part of its specified behavior?

    We would appreciate your confirmation and guidance on this matter. Thank you in advance for your support.

  • Hi

    No TOFF MAX WARNIGN does not trigger shutdown since it is warning and it does not impact the system.

    you may need to increase the interval between the resequence if you don't want to start the resequence early

    Regards

    Yihe 

  • Yihe-san,


    Thank you for your reply.
    Based on your comment that “No TOFF MAX WARNING does not trigger a shutdown since it is only a warning and does not impact the system,” we will reconfirm this behavior on the actual unit.


    Please note that the issue this time does not occur at the resequence start timing, so we will reconfirm the behavior of “No TOFF MAX WARNING.”
    Best regards,

  • Yihe-san,

    I have reconfirmed the behavior.

    When a TOFF MAX WARNING occurs, Re-Sequence starts once the time configured for Re-Sequence has elapsed, even if not all rails have reached POWER_GOOD_OFF yet.
    Is it correct to understand that this behavior is by design?

    On the other hand, based on the explanation that “TOFF MAX WARNING does not trigger a shutdown since it is only a warning and does not impact the system,” my understanding was that the Re-Sequence timer would start counting only after all rails have reached POWER_GOOD_OFF, and that Re-Sequence would begin after the configured time has elapsed.

  • Hi

    Your observation is right. but you can set the TOFF_MAX delay to unlimits so that no rail will trigger TOFF_MAX warning if this is not what you expect. Obvious some kind of leakage or slow discharge causes the output voltage above 12.5% of the normal VOUT.

    Regards

    Yihe

  • Yihe-san,

    Thank you for your response.

    I now understand how TOFF_MAX delay works.

    I will now close this matter.