• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Subscribers 82 subscribers
  • Views 2760 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 of the MSP

Bastian Mieck
Prodigy 730 points
Part Number: MSP430F67471

Hi everybody,

one of my customers has problem with the MSP430F67471 temperature sensor.

  • It seems that the sensor is calibrated through TI at 30° and 85°. Thats correct right?
  • The customer measured these temperatures and saw that these temp measurements are very good (+- 1k)
  • But at negative temps they see that the error is about +- 5k. This error is way to high for them. They allow max +- 2k. Do we have a graph or something that shows the error at negative temps? Or can you explain why this high error occurs?

Now they want to understand why this error occurs and need more information about the sensor

  • What is the principle of the temp sensor? PN with a diode? Transistor? or something else?
  • Can the internal sensor measure negative temps? If yes what is the accuracy etc? Do we have app notes or TI-Designs reagrding the temp measurement with the MSP430F67471?

thank you

best regards

Bastian

  • 0
    Guru 244230 points

    The User's Guide says that

    the REFGEN subsystem also includes the temperature sensor circuitry, because it is derived from the bandgap.

    Wikipedia says about the Silicon bandgap temperature sensor:

    The principle of the sensor is that the forward voltage of a silicon diode, which may be the base-emitter junction of a bipolar junction transistor (BJT), is temperature-dependent
    […]
    If high precision is not required it is enough to bias a diode with any constant low current and use its −2 mV/˚C thermal coefficient for temperature calculation

    That is indeed the coefficient that is documented for the MSP temperature sensor voltage.

    In theory, the coefficient is constant, i.e., there should not be a very high deviation at very low temperatures. I don't know about the constant current source, but it appears that it does not work as well at low temperatures.

  • 0 in reply to Clemens Ladisch
    TI__Expert 4695 points

    This is what the UG says:

    At the both measure points the deviation can be up to +/-3C

    This means in the cold area <30C the deviation can be much higher

    Lukas

  • 0 in reply to Lukas Badura
    TI__Genius 14165 points
    Hi Bastian,

    maybe a look to following document might help:

    www.ti.com/.../slay025.pdf

    If customer requires more accuracy between points different than the calibrated ones a 2 point calibration at the corners of the customer required temp range might be required.
  • 0 in reply to Dietmar Walther
    TI__Prodigy 730 points
    Hi everybody,

    many thanks for the fast response! I talked to the customer and he understands now. But he istn happy about the part of the constant current source. They need to know the error of the "Temp Sensor" between -40°C and +85°C. Do we have numbers for the current source in this range? Or for the whole Temp sensor in this range?

    They use the correction coefficients from TI right now. And they know they have an error. But is it possible to get correction coefficients for values from -30 or -20 to +85?
  • 0 in reply to Bastian Mieck
    TI__Genius 14165 points
    Hi Bastian,

    I checked our database we do not have recorded the data you requested. On the other site even if we would have it would not help because device to device variation might very large. That's why we calibrate 2 points for each device.
    Is customer able to do it's own device specific calibration?
  • 0 in reply to Dietmar Walther
    TI__Genius 14165 points
    Hi Bastian,
    were you able to resolve this problem at your customer? If not please let us know otherwise we will close this thread.
  • 0 in reply to Dietmar Walther
    TI__Prodigy 730 points

    Hi Dietmar,

    i didnt get an answer yet. So i think you can clos it for now.. If not I come back to you guys.

    thank you for the help

    Bastian

**Attention** This is a public forum