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.

TPS3813K33-EP: Watchdog window frame is shrinking at -55C out of limits in datasheet

Part Number: TPS3813K33-EP

PN TPS3813K33MDBVREP is used in an application where it is tested at -55C.  Some parts are causing failures, where the processor is resetting because the watchdog window has shrunk and the part is emitting a reset pulse earlier than it shoud.  Testing the parts with a test circuit shows that in some parts, the watchdog window grows to 320ms at -55C, when the upper limit of the spec is 300ms.  This has been confirmed on multiple parts.  Other parts tested have a window of 266ms at -55C. 

  • Nicholas,
    Could you provide some scope shots of your testing?
    What is your testing methodology? Measured from power up, or from last WDI rising edge?
    You also indicate that the window has shrunk, then later indicate that window grows. Can you confirm?

    Is your system sensitive to slightly larger window?

    Regards,
    Wade
  • Hello Wade,


    Apologies for the confusion, the issue is that the window is growing, not shrinking.  On one sample it was measured to be 284.9ms at 21C, and 320.5ms at -55C.  On a known good part, it was measured to be 250.6ms at 21C, and 266.0ms at -55C.

    Test methodology was measuring from last WDI rising edge.  A pulse was applied (5% duty cycle), and the period was increased until the part began to trigger a reset.

    The system is sensitive to a slightly larger window.  We perform a built in test that checks the functionality ofthe watchdog timer, and if the window is too large this test will fail. 

    I am having trouble getting the scope images to upload in this tool, can I share those via email instead?

    Thank you,

    Nick

  • Thanks Nick.
    To insert images or documents, select the "Insert Code, Attach Files and more..."
    Then use the paperclip icon to insert the image.

    If that does not work, I can send you an email.
    BTW, I am out tomorrow for long weekend. I should be able to reply back on Tuesday.
    Regards,
    Wade
  • Hi Wade,

    In the followng waveforms channel 2 (blue) is WDI and channel 3 (purple) is RESET*.

    Thanks,

    Nick

    Resetting with 285.7ms between WDI pulses (ambient):

    No reset with 284.9ms between WDI pulses (ambient)

    No reset with 318.5ms between WDI pulses (-55C)

    Resetting with 321.5ms between WDI pulses (-55C)

  • Nick, thank you for the plots.
    Would it be possible to have you send a representative unit of each passing and failing case?

    If so, I can send you and email with shipment information.
    Regards,
    Wade