• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 62 subscribers
  • Views 913 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.

TPS659037: PMIC reset_in pin voltage issue

AM5728_User
Prodigy 250 points
Part Number: TPS659037
Other Parts Discussed in Thread: AM5728, TPS3808

Hello,

We have a design based on TMDXIDK5728 using PMIC TPS6590377 and AM5728. The board is intended to be always ON so we implement the "imminent shutdown" circuitry with TPS3808 voltage monitoring (based on schematic AM571_IDKEVM_Page5Rev1_3A).

During preliminary testing TPS3808 circuitry was not available and the boards have been turned on/off with a switch toggling imminent_shutdown signal between 0V and 5V (Pin is 5V tolerant).

Now, we are using the following design on the same boards:

On some of the boards (At least 3 on 5):

  • The imminent_shutdown signal is stucked to ground and the board does not start
  • If  imminent_shutdown signal is floating (pull up to VRTC_OUT), the board does not start
  • If I force imminent_shutdown signal to 5V, the boards starts

Some of the boards start correctly (At least 1 on 5).

We suspect the switch to have produced overshoot and/or undershoot on pmic_reset_in pin and damaged the component. We are going to replace a PMIC to validate this assumption.

Questions:

  1. Can undershoot/overshoot on PMIC_RESET explain damage on component?
  2. Because the same  problem occurred on several boards, do you think something else could explain the failure?

Thank you for your guidance.

  • 0
    TI__Genius 16410 points
    Hello,

    I think the more likely issue is that you have a 10k resistor from VRTC_OUT to 5V in the first design. LDOVRTC does not have any sink capability, so this will likely pull LDOVRTC_OUT higher than 1.8V, and could be up to 5V. You could verify this by measuring LDOVRTC on one of your boards. Increasing the VRTC voltage could cause damage. If you have a fresh board, you could try removing that resistor and see if it works correctly. Although if you are no longer using the first design, this may not be necessary, and a fresh board will probably work with the supervisor.

    The schematic of the second design looks correct to me, and I don't see any design issue.

    To answer your original question, any overshoot on RESET_IN above 5.25V could cause damage. This is the absolute maximum rating in the datasheet.

    Regards,
    Karl