• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 589 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.

TMP117: Pressure dependence of temperature reading

Adam Steele
Prodigy 160 points
Part Number: TMP117

Does the TMP117 have a known dependence of its reading on pressure? I appear to be measuring something of order 0.01 C/mbar

In my case, the TMP117 sensor together with some other components on the following board:

https://www.adafruit.com/product/4821

So I can't be sure its the TMP117 but thought I'd check to see if this is a known source of uncertainty.

  • 0
    TI__Mastermind 43785 points

    Hi Adam,

    There is a known sensitivity to mechanical stress. I would assume this is what you are observing. There is a mention of mechanical stress in the datasheet. We have observed milliCelsius errors due to mechanical stress. Please note that TMP117 has 7.8mC resolution, and you will need a suitable instrument (calibration bath or similar) to ensure you are not simply observing actual temperature variation. 

    thanks,

    ren

  • 0 in reply to Ren Schackmann
    Prodigy 160 points

    Hi Ren,

    I have a very sensitive system under test that can test temperature changes at or below the measurement threshold

    I can't rule out stress entirely, but I can say there is a measurement dependence on ambient pressure even with the board not clamped to anything. I put the board (not attached to anything but the PCB) in a sealed vessel and pumped to down from ~1000 mBar to about 900.  With this I can obtain a P vs T curve indicating the sensitivity, and a dT/dP calibration factor.

    There is some hysteresis because I think the chip may be partially sealed and takes a it of time to equilibrate.

    Since ambient air pressure changes by +-1% with regularity it would imply an error over some days of ~0.1 C, which is pretty significant.

  • +1 in reply to Adam Steele
    TI__Mastermind 43785 points

    Hi Adam,

    We don't have any testing under vacuum to share, and subsequently there is no specification for performance under vacuum. Any performance under vacuum characterization would also be dependent on board design. While your work does interest us, I'm not sure we can offer any further insight into your particular application. I can only confirm that all manner of mechanical stress could have a small impact on the output of TMP117. This effect is not unique to this device; any IC with bandgap voltage reference could have a measurable output shift due to applied stress. Mechanical stresses include both solder mounting stresses (as mentioned in the datasheet) and other forces applied after mounting such as encapsulation or PCB twisting/flexing.

    thanks,

    ren 

  • 0 in reply to Ren Schackmann
    Prodigy 160 points

    Ok thanks for your feedback.

    I still think this is worth investigation and mentioning in the datasheet specifically. After all, It means you could not for example calibrate a device at one location if it was going to be moved to significantly different altitude.

    Also with such a correction (which may well be independent of the board) the device's long-term accuracy could in fact be made much better than the marketed spec!

    These results may go into a Rev.Sci.Inst. btw.

  • 0 in reply to Adam Steele
    TI__Mastermind 43785 points

    All I can say is we are excited about upcoming products. Released devices rarely see changes to their specification.

    Please let us know if you publish. We are happy to see it.

    thanks,

    ren