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.

TPS6594-Q1: Watchdog window exception on pmic

Part Number: TPS6594-Q1

This project uses PMIC TPS6594-Q1, which uses the watchdog inside. The QA mode and window parameters are set to a maximum value of about 70ms. It was found that the system did not restart after feeding the dog for 50ms. According to the manual, the calculation window should be 66.88ms~73.92ms. Why does it not restart even after feeding the dog for 50ms?

#define WIN1DURATION    70400U//us
#define WIN2DURATION    70400U//us
  • Hello,

    It was found that the system did not restart after feeding the dog for 50ms.

    Was an answer given every 50ms? The PMIC is expecting 4 "answers" from the MCU, 3 during WINDOW1 and 1 during WINDOW2. If the last answer comes too early this will cause a fault/error. 

    Based on the above figure from the data sheet and your settings. Answer 3, 2, and 1 must be received in first 70ms (Window-1) and Answer 0 would have to wait until Window-2. 

    I suspect that you are sending WD answers too early. Is it possible to query the PMIC with another I2C device after the rails shut down? If so, please read WD_ERR_STATUS register.

  • Thank you.

    I have a question about watchdog window time,

    as I defined :

    #define WIN1DURATION    70400U//us
    #define WIN2DURATION    70400U//us
    Which understanding is correct below? Figue 1 or figure 2?
    figure 1:

    figure 2:

  • Figure 1 is correct.

    The 0.95 vs 1.05 in the data sheet equation is to account for the 5% variance in the clock.