• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 32 subscribers
  • Views 289 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.

TMP119: TMP119 run into a wrong status

Lingnan Zeng
Prodigy 10 points
Part Number: TMP119

Tool/software:

There is a small probability that the TMP119 will get a wrong temperature which may 4-5 °C higer than normal. And it will keep the status always unless we  repower the IC, it will recover. Read out all the register and compare with normal one, no abnormal found. What is the reason will cause the IC run into a wrong status?

  • 0
    TI__Expert 8850 points

    Hello Zeng,

    In what condition are you seeing the wrong temperature measurement? What do you mean by wrong status? Are you using Alert or Threm Mode? Can you provide a schematic and layout of the TMP119 being used?

    Best Regards,

    Meredith McKean

  • 0 in reply to Meredith McKean
    Prodigy 10 points

    Hello Meredith McKean,

    Thanks for your reply. In my use case, we use TMP119 to monitor the temperature of a iron surface.And there exist no only one TMP119 sensor.All the sensor can get the right vaule of the temperature at the begining, but there is a small probability that one or two TMP119 will get the wrong temperature(4-5 °C higer) if TMP119 keep power on for more than 24 hours, and the vaule still keep stable.

    The Sch and layout are based on the datasheet, I don't think it can cause this phenomenon. Now we have two suspects:

    1.Static electricity has accumulated on the surface of the iron block under test. Because the iron block is isolated without GND connection.

    2.Power supply for TMP119 have jitter , and it may cause chip generate reset fail or the inside ADC work abnormal.

    Can you help me to analysis whether the suspect is resonable? If you have more suggestion or analysis base on the IC inside circuit , please tell me. Thanks.

  • 0 in reply to Lingnan Zeng
    TI__Expert 8850 points

    Hello Zeng,

    Yes a clean supply is needed for good accuracy. This seems like a latch up issue from ESD. ESD protection diodes | TI.com

    Best Regards,

    Meredith McKean

  • 0 in reply to Meredith McKean
    Prodigy 10 points

    Hello Meredith McKean,

    Thanks for your reply. can you provide more detail to help me understand " a latch up issue from ESD". As the IC datasheet showed, can I  understand that "a latch up issue" means ADC circuit inside the chip work abnormal?

    In addition, in my application, we use TMP119 to monitor the temperature of a iron block.How can I to avoid the ESD issue.Add ESD for which signals? VCC? I2C? or we must to add ESD between GND and iron block? 

  • +1 in reply to Lingnan Zeng
    TI__Mastermind 43425 points

    Hi Lingnan,

    Your problem statement matches that of a latch up event:

    •  The issue goes away on power cycle
    • The current path created in the latch up event could be responsible for 4-5C self-heating 

    All IC are susceptible to latch up, and there is a JEDEC standard related to it. The TMP119AIYBGR passed the JEDEC JESD78 testing as outlined in it's qualification summary. You can read more about latch up in this white paper: https://www.ti.com/lit/scaa124

    Latch up is a problem we are commonly seeing when the sensor is mounted close to metal. The metal object is a path for ESD to travel to the sensor and dissipate to ground through the sensor. All pins of the sensor may need additional ESD protection to prevent the discharge from passing through the IC itself. The metal should also be grounded, so that charge can pass to ground without approaching the sensor at all. 

    thanks,

    ren