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: Inconsistent RSTn Time Periods - Using As Watchdog - Recommendations?

Part Number: TPL5010

Hello, we are currently using the TPL5010 in two products as a watchdog. Our intended use was to have it reset the MCU if the MCU freezes and is no longer been sending DONE pulses. We have a product where we set the timer to 5 minutes (42.9K resistor), and then we have the MCU output low on the DONE pin. Every 1 minute the MCU outputs high on the DONE pin, for 1 second, then back to low. This has always appeared to work correctly, and if the MCU freezes and no longer is triggering the DONE pin, then the MCU is successfully reset after 5 minutes.

The problem is with our new product. We have it set up for 15 minutes now (69.532K resistor), and this time we are alternating the DONE pin high for 1 second, low for 1 second, indefinably. If the MCU freezes and stops doing this, we are getting resets that are very inconsistent.... e.g. 22 minutes, 30 minutes, etc.

I recently came across a thread that mentioned there is a phenomenon where if you send too many DONE pulses that it can append the time duration to the next time duration. How come we never noticed this on our first product?

In any case, can we please get some advise on how we should go about fixing this, or if we can get a better explanation on this issue? Thanks and any help is greatly appreciated!

  • Some additional info:

    We now set up the DONE pulse so that it stays low for 1 minute, goes high for 1 second, then back to low for another minute, then repeat. This is exactly how we have our first product setup.

    With that change, we then stop the DONE pulse coming out of the MCU and observe what the TPL5010 will do. We are getting similar results.

  • Hello, could you share the oscilloscope captures from your experiments? Seeing the waveforms for DONE, WAKE, RSTn and DELAY/M_RST will help me understand the issue better.

    Kind regards,
    Lane

  • Hi Lane, I'm not sure what you are expecting to see.... my oscilloscope cannot capture a full 30 minutes worth of waveform data to share. You just want to see a high changing to low, and a low changing to high on those pins? Because that's all they look like on the oscilliscope....

    For example the RSTn pin is high, and when it triggers it goes low for 320ms, then back to high. You want me to actually capture this on the oscilliscope? Its clearly documented in the TPS5010 datasheet. Same with WAKE. The DONE pin is exactly like I described... its low for 1 minute, high for 1 second, then low for 1 minute. You just want to see it going from low to high? What is there to gain from that? And like I mentioned earlier, oscilliscopes typically can't show a minute long waveform....

  • Hi Matt,

    Perhaps you could put together a timing diagram explaining the issue? I don't fully understand the problem based on the description and I also haven't encountered this issue before. 

    Could you also share the schematic? If you prefer not to post on the forum you can share it with me in a direct message.

    Based on my current understanding however, my guess is that explanation in the related thread still applies here IF you are measuring the time from the last DONE (MCU freeze) until the RSTn falling edge. In this case there would be no issue with the part and it would be operating as anticipated.

    Kind regards,
    Lane

  • Hello, I could not figure out how to send it to you through a message. I clicked your name and went to your profile, but not where did I see an option for sending a message. I uploaded the schematic to the case #, and asked that they forward it to you.

  • Hi Matt,

    Thanks, I will look forward to the email with the case #. Also, now that we are friends on E2E, you could send it to me in a private message.

    Kind regards,
    Lane

  • Hi Matt,

    I reviewed your schematic and it is OK.

    I think this same explanation in the related post applies: https://e2e.ti.com/support/clock-and-timing/f/48/t/823213

    Kind regards,
    Lane