• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 80 subscribers
  • Views 3212 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.

MSP430F67471: temperature sensor calibration

DJ-NG
Genius 9355 points
Part Number: MSP430F67471
Other Parts Discussed in Thread: MSP430F147, MSP430F148, MSP430FG438

Hi champs,

cust is using the temperature sensor embedded in the MSP430F67471IPZR.

After a single-point calibration, as recommended in Note (4) of the datasheet, p. 91, we still see big discrepancies at other temperatures (slope problem). In our case, the single-point calibration was performed at +20 °C. Device brought to the temperature -40 °C, reads back only -30 °C. Theoretically one could perform a double-point calibration as described in chapter 1.13.5.3 of the User’s Guide (SLAU208O), however it is unpractical and extremely time consuming:

 

 

 

 

The User’s Guide presents a typical sensor transfer function as follows:

 

 

 

According to this function, the slope is TCsensor = 2.52 mV per °C.

Question:

how reproducible is this slope for different MSPs?

I find no specification of the TCsensor neither in the Datasheet nor in the User’s Guide (SLAU208O).

Can you provide me with the specification of the temperature sensor, please?

  • 0
    Guru 305970 points
    Aren't you using the TLV calibration values?
  • 0 in reply to Clemens Ladisch
    TI__Expert 4695 points

    You can find the TLV data in the Datasheet on page 148

    Here you see the TLV calibration date for 30 and 85C by internal VREF (1.5 / 2.0 / 2.5V)

  • 0 in reply to Lukas Badura
    TI__Genius 9355 points
    Looking at TLV calibration dates I can check if the MSP was already calibrated at TI.
    However, we still do not know, what is allowed discrepancy between measured and expected temperature.
    In our case, we performed a calibration at a single temperature +20 °C.
    At -40, some devices read -32 back.

    We would like to understand, if this is expected or rather a non-expected result.

    Therefore we would like to know more about the accuracy after single-point offset calibration.

    Can you help us, please?
  • 0 in reply to DJ-NG
    Guru 305970 points
    The datasheet does not specify a maximum tolerance for single-point calibration. So nothing is guaranteed.

    Apparently, it is expected that you apply the two-point calibration. (The calibration values are constants for any specific chip, so you can compute a multiplication factor at startup, and use fixed-point arithmetic.)
  • 0 in reply to Clemens Ladisch
    TI__Genius 9355 points
    Thank you for the information.
    We understand, the tolerance for single-point calibration is not specified. But what about the tolerance of the two reference values stored by TI in the TLV block?
    If we use them, what is the accuracy of the slope of the transfer function then?
    In the past, we designed lots of MSPs into our devices (e.g. MSP430F147, MSP430F148, MSP430FG438).
    All of them had specified slope accuracy at 3.55 +/-3% mV/°C.
    We really miss this information for MSP430F67471IPZR.

**Attention** This is a public forum