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.

TPS3823: WDI function of TPS3823

Part Number: TPS3823

There is an abnormal behavior on our Reset circuit, and we are using TPS3823 on here.

We found our system will reboot unexpectedly, it looks the RESET(PIN1) is asserted.

Currently we set WDI(PIN4) of TPS3823 to high-impedance , it means we purpose to disable the WDI timer from our system.

We found there is still a period pulse on WDI pin, it should be TPS3823’s own WDI pulse to ensure RESET does not assert, however the waveform is like triangle and peak voltage is lower(about 1.5V only), moreover the period is about 1.7sec(exceed the minimum watchdog time-out period: 0.9sec).

1. with 0.1uF cap(C260) on WDI, the pulse width is about 10ms,  can duplicate system reboot. 

2. Remove 0.1uF cap, the pulse width is about 2us, there is no reboot so far. 

We are not sure that 0.1uF decoupling cap is root cause, however even with no 0.1uF cap, we still concern this WDI waveform would trigger the RESET unexpectedly.

  • Sam,

    The period of 1.7 seconds is within the min (0.9s) and max (2.5s) so this appears to be the watchdog timeout that you are seeing.

    The WDI pin charges up to an internal reference voltage which sets the timeout. There should not be any decoupling capacitors on the WDI pin as this will affect the watchdog timing (because the pin which ramp up to the internal reference voltage with a different rate). Please remove the decoupling capacitor and check if the device works as expected.

    -Michael
  • Thanks Michael for your support, this resolved my issue.