• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 30 subscribers
  • Views 610 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.

TMP006 calibration / application question

Jim D.
Prodigy 65 points
Other Parts Discussed in Thread: TMP006

We've read many of the posts here related to the TMP006, and downloaded useful calibration spreadsheets and sample code.

In our application the TMP006's field of view will be the underside of a heatsink we wish to monitor.  An accuracy of +/-3 C is acceptable over Tdie and heatsink range of -20C to +70C.  In our application, Tdie and heatsink temp are not related and don't trend together.  We can gather two calibration points with a constant Tdie to obtain our calibration constant - at the lowest and highest expected heatsink temperatures.  We intend to use the 'transient temp correction' discussed in this forum.

My questions are:

1) Is the calibration data taken at Tdie=20C still valid for calculating the remote temp in actual use when Tdie varies between -20C and +50C?

2) Is the calculated remote temp accurate if the remote temp is significantly colder (say -10C) than Tdie, and Tdie is significantly higher than when the cal data was taken (say Tdie=50C)?

3) Can the cal data be gathered at elevated Tdie (say 50C) if we can hold Tdie constant?  (remote temp would be -20C to +70C)

Thanks,

  Jim D.

  • 0
    TI__Mastermind 27715 points

    Hello Jim,

    Allow me to first ask - what material is your heatsink made of? Per the TMP006 User's Guide, most metals are poor radiators of IR energy and will not be suitable for use with the TMP006 unless you paint or coat the area of interest on the heatsink.

    1. I understand that the target object (heatsink) temperature does not track TDIE. If this is the case, taking calibration data at the min/max temperature points of the heatsink while keeping TDIE at room temperature is acceptable. However, your accuracy will be best when TDIE is at the calibration temperature.
    2. In general, the TMP006 accuracy is best when TDIE is closed to the target object temperature. When they drift apart (in either direction), accuracy begins to degrade. In order to stay within datasheet performance of ±3°C, we specify that (TOBJECT - TDIE) = -10°C to 30°C.
    3. If you have a way to regulate TDIE, the best case for calibration would be to create a test matrix where you can sweep both TDIE and the heatsink temperature. Min/mid/max for each is preferred, but min/max is acceptable. I would recommend sending me your calibration data once it's collected as I can do a more thorough calibration which adjusts all the coefficients in the TMP006 equations - this algorithm will notice that TDIE has little to no effect on the object temperature and will de-emphasize it from the calculation. It will also result in greater accuracy than an S0-only calibration. To do this, I need three values - the TMP006 sensor voltage, the value of TDIE, and the accurate heatsink temperature measurement (taken with a thermocouple, RTD, or other known accurate temperature sensor).

    Best regards,

    Ian Williams
    Linear Applications Engineer
    Precision Analog 

  • 0 in reply to Ian Williams
    Prodigy 65 points

    Hi Ian,

    Thanks for your insight.  The area of the heatsink "seen" by the TMP006 with either be painted or powder coated with a dark color.  Will this yield a decent result?

    Also, thanks for the offer to supply a more extensive set of calibration data.  We'll take you up on this when we have the final mechanical parts and finishes, and take a final set of measurements.

    Regarding the (TOBJECT-TDIE) best accuracy range of -10C to +30C, We'll perform testing outside this range.  I think we can live within these limitations or at least compensate in our code when we're outside this range.  

    Best Regards,

         Jim D.

  • 0 in reply to Jim D.
    TI__Mastermind 27715 points

    Hi Jim,

    If you paint or powder coat the heatsink area then you will be in good shape! That should correct any emissivity issues the metal heatsink may have.

    Regarding the (TOBJ - TDIE) range - that's a general-case condition. With calibration, we've been able to get accurate results even when the temperature delta is much  higher.

    Let me know once you have some final data gathered and I'll be happy to help you with a full calibration.

    Best regards,

    Ian Williams