• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 20 subscribers
  • Views 1364 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.

TPL5010: Reset signals completely disabled after a certain condition

Mohamad Nizar Kezzo
Prodigy 175 points
Part Number: TPL5010

Hi, 

We're developing a product that uses tpl5010 as a watchdog timer. I handed bare PCBs to the software team to develop the code. 

After a while they complained that sometimes the reset signal is not being issued by TPL5010 at all. 

I further checked, and noticed that this specifically happens when a connector that goes to the DONE pin is touched by hand. The problem happens more consistently if both the WAKE and DONE pin connectors are touched by hand at the same time. Touching the WAKE pin connector only does not cause the issue. 

It appeared that the human body couples the 60Hz from the wal to the DONE pin connector. Taken from the oscilloscope this is the trace on the DONE connector when touched by a finger (notice the considerable amplitude of >2V):

My questions are the following:

- Do you agree with the above explanation?

- What is disturbing me is that the Watch Dog does not resume operation after this, although the finger is removed (i.e. DONE signals no longer flow to the DONE pin). The reset signal does not get triggered at all after this case. Is there a fix to this (our product is operated close to the human body)? 

Best regards, 

Nizar

  • 0
    TI__Genius 16845 points

    Hi Nizar,

    I wasn't able to replicate your issue on my bench using TPL5010EVM. In my setup:

    • VDD = 5V
    • R_ext = 500ohm => T_ip = 100ms
    • DONE pin is pulled down to ground

    With this setup, RSTn will be periodic; RSTn will remain high for 100ms, then go low for 320ms.

    When I touch the DONE pin with my finger, it does not affect the RSTn signal. Touching DONE and WAKE simultaneously also does not affect RSTn.

    Kind regards,
    Lane

  • 0 in reply to Lane Boyd
    Prodigy 175 points

    Hi Lane,

    Thank you for your reply.

    I resolved the issue. But not sure if my explanation of the solution is valid.

    When oscilloscope probe is NOT connected to DONE:

    The DONE pin was floating in my test setup. When I touched it with my finger, it apparently accumulated some charge and continued to pick the noise. This caused the TPL5010 to keep thinking it is receiving DONE signals for some reason, and it never recovered. It kept sending a trail of WAKE signals. 

    When oscilloscope probe is connected to DONE:

    When I connected the oscilloscope probe to the DONE connector, and touched it with my finger, the TPL5010 acknowledged the noise as a DONE signal but was able to recover. Apparently the 1MOhm resistance to ground on the oscilloscope input drained any accumulated charge and pulled the line down. You can see the noise caused by a finger touch on the DONE line. 

    Is there a requirement on the DONE and WAKE lines to be pulled down by the microcontroller's GPIO? Do you think the above explanation is valid?

    Thank you!