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: Digital Temperature Sensor Questions and Device Selection Help

Part Number: TMP117

Hello,

I have a few questions regarding the TMP117 device.  This device is going to be used in cold temperatures (specifically a refrigerator).

  • In regards to  Package Temperature Error vs Temperature graphs,  I'm confused on why the two packages (DRV vs. YBG) have different results.  Can someone explain to me why this is the case?  Also, since this devices is going to be used in cold environments, could you confirm that the TMP117 is the best device for this application?  My customer is open to switch to another device it would better fit their application.  By the way, accuracy is very important.

  • In regards to the non-volatile memory on the device, what are some ways the EEPROM is utilized?  Personally, I figured it could be used to store temp data over time to track if the product has been stored properly.  I'm curious to hear how other people have used it.

I'd greatly appreciate any assistance.

Best,

Alex

  • Hi Alex,

    The charts show actual data from different sets of units. There is room for variation within the spec, and the data shown does not violate spec. Additionally, we have seen that mechanical stresses affect the output of high precision analog ICsy. These packages represent different stresses. TMP117 is our highest accuracy temperature sensor.

    There are only three 16bit EEPROM registers. It's not enough to retain measurement history. To log temperature, you will need an MCU to fetch the temperature measurements from TMP117 and store them in nonvolatile memory. 

    thanks,

    ren

  • Thanks for the response Ren.  I'm still a little confused.  I greatly appreciate your help.

    This product will be stored in a temperature range from 0 to 8 degrees Celsius. Can you give an explanation for why the DRV package has a negative temp. error while the YBG has a positive temp. error? It also appears that the YBG is more stable over the entire temp. spectrum.  Would you say the YBG package is a better option in terms of accuracy?  Also, is there anything that can be done to minimize this error?

    Thank you for the clarification on the EEPROM.  How do you usually see the customers utilizing the EEPROM?

  • Alex,

    It is normal variation. 

    I would not say one package is better than the other in terms of accuracy.

    There isn't anything that can be done to improve over-temperature behavior. You could buy a better sensor, but this is the best one we make. Since you only care about a very narrow range of temperature, it is possible for customer to calibrate the offset at that temperature. However, calibration at this level of accuracy is not realistic in production for any customer that I've encountered. 

    We store identification numbers in the EEPROM for NIST traceability. Most customers do not overwrite our NIST IDs. 

    thanks,

    ren

  • Ren, 

    What is the purpose of NIST traceabilty?   In what applications is NIST traceability used?  Since it is pre-programmed into the register, I'm confused on how that would be utilized.  Would like an MCU read that register and make sure that the temp. sensor its reading has NIST standard?

  • Some applications have government regulation and approval. You can see a list of applications on the front page of the TMP117 datasheet. Our statements and documents regarding NIST traceability establish the credentials for our products in these situations. You can access these documents by clicking the request link on the TMP117 product page.

    The actual data in the EEPROM is not publicly documented and cannot be used directly by the customer. 

    Further info on NIST:

    e2e.ti.com/.../tmp117-nist-calibration-certificates

    https://e2e.ti.com/support/sensors-group/sensors/f/sensors-forum/1000579/tmp117-tmp117-nist-byte-order-and-eeprom4-address

    thanks,

    ren