• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 79 subscribers
  • Views 336 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

MSP430FR2675: Temperature sensor TLV calibration accuracy

Ralph Braeumer
Prodigy 140 points
Part Number: MSP430FR2675

Hi,

we use the integrated temperature sensor with the calibration data from TLV. Unfortunately the results are not as expected.

cal30 (1.5Vref)  @ 0x1A1A  2114
cal105(1.5Vref) @ 0x1A1C  2683

We followed the sample code and got the following measurement results:

Temperature [degC] Measured ADC data Calculated temp [degC]
30 2115 30
105 2614 95,5



As you can see, ADC measurement result at room temperature (30 degC) is accurate and nearly matches the calibration data 2114 from TLV.
When measuring at 105 degC, the measurement result does not match the calibration data 2683 from TLV.
Why the measured value differs so much?
Is there someting wrong with the reference point at 105 degC or just inaccuracy of the sensor at high temperatures ?
We expected more accurate temperature data when using two point calibration, particularly at the two calibration points.

Regards,

Ralph

  • 0
    TI__Expert 4695 points

    Hi Ralph,

    So which expectation do you have to the accuracy of the calibrated data?

    Is there any specification which you could reference?

    Many thanks
    Lukas

  • 0 in reply to Lukas Badura
    Prodigy 140 points

    Hi Lukas,

    thanks for your immediate response!
    As far as I know there is no accuracy specified for the temperature sensor in the MSP430 datasheet.

    I'm just wondering why the ADC raw value exactly matches at 30 degC, but not at 105 degC. We measure the temperature directly on the cpu housing. I suppose the two calibration values have been determined the same way. I know there is always a difference between ext. temperature and die temperature, but we didn't expect a deviation of about 10%.

    Any information regarding sensor accuracy and how to improve it would be helpful for us.

    Best regards
    Ralph

  • 0 in reply to Ralph Braeumer
    TI__Expert 4695 points

    Hi Ralph,

    Thanks for clarification, I was just concerned I've overseen any specification for this device.

    It depends on your application, but expecting accuracy better then 10% I would recommend to perform a 2 point calibration on application level.

    Hope that helps.

    Best regards

    Lukas

  • 0 in reply to Lukas Badura
    Prodigy 140 points

    Hi Lukas,

    if I understand aright the calibration data in TLV seems to be correct and this is normal inaccuracy of the temperature sensor?
    With application level you mean that for getting higher accuracy we should not use the calibration points from TLV, but our own calibration data, right?

    Regards
    Ralph

  • 0 in reply to Ralph Braeumer
    TI__Expert 4695 points

    Hi Ralph,

    Yes, we recommend to generate own calibration data on application level.

    In addition, based on the expected accuracy you should maybe also consider to use an external temperature sensor.

    Best regards

    Lukas

**Attention** This is a public forum