• State Resolved
  • Locked Locked
  • Replies 11 replies
  • Answers 3 answers
  • Subscribers 46 subscribers
  • Views 1999 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.

LMP90077: System calibration coefficients

Michael Hemetsberger94
Prodigy 100 points
Other Parts Discussed in Thread: LMP90077

Hi,


I am not sure if I understand the system calibration procedure correctly.

I want to use the LMP90077 together with a 2-wire RTD (at Pin 2 /VIN0) and Pin 10 (VIN6/REFP2)). A reference resistror is connected between Pin 10 and Pin 11.

When doiing the System Calibration Offset Coefficient Determination I have to short-circuit the RTD to get a zero voltage at the ADC. Or is it possible to do the Offset Coefficient Determination with an ADC input voltage greater than 0V (300mV).

Thanks and regards,

Michael

  • 0
    TI__Genius 16890 points
    Hi Michael,

    Yes, you have to have a zero scale condition at the input. See section 9.3.1.2.1, step 1 in the datasheet.

    Mike
  • 0 in reply to Michael Stout
    Prodigy 100 points
    Hi again,
    I am not sure if I do the system calibration in a correct manner.

    - First of all I select the channel I want to calibrate (CH0)
    - I apply a zero voltage at CH0, tie VIN1 and VIN0 to GND
    - initiate the system calibration offset coefficient determination mode (0x01 to register 0x17)
    - then I apply a full scale condition at CH0, tie VIN0 to VREF and VIN1 to GND
    - initiate the system calibration gain coefficient dtermination mode (0x02 to register 0x17)

    However I do not get an ADC value I expected after performing the system calibration.

    Do I have to preload the CH0_SCAL_GAIN register with 0x7FFF or 0xFFFF? Or is the default value of 0x8000 correct in this case?
    Do I have to do the calibration with the channel gain I want to use? Or should I use a gain of 1 during system calibration?

    Thanks for any hints,
    regards Michael
  • 0 in reply to Michael Hemetsberger94
    TI__Genius 16890 points
    Hi Michael,

    Yes, you do need to preload the CH0_SCAL_GAIN register with the value you expect. Since you have the reference voltage on the input, the value you should put in the register should be 0xFF_FFFF. I would do this calibration with the gain you are expecting to use.

    When you do the offset calibration just tie the inputs together without tying to ground. The inputs are not allowed to go to ground and this may be affecting your calibration.

    Mike
  • 0 in reply to Michael Stout
    Prodigy 100 points

    Hi Mike,

    I changed the setup for the system calibration.

    Now I tie the inputs VIN0 and VIN1 together when doing the offset calibration. Afterwards I apply VREF to the ADC CH_0.

    I get feasible values for the offset and gain correction (Offset: -2 and Gain: 1.0625)

    However after system calibration, the value in the ADC_DOUT register doesn't match the expected value.

    Do I have to subtract the calibrated system offset from the read ADC value and divide it by the calibrated gain coefficient manually?

    If I do that, I get an ADC value, that would fit the expected value.

    Regards Michael

  • 0 in reply to Michael Hemetsberger94
    TI__Genius 16890 points
    Hi Michael,

    The LMP90077 should take care of it automatically once you have the calibration values in the registers. Try changing the ScanMode. Does that help? Also try turning on background calibration mode 3.

    Mike
  • 0 in reply to Michael Stout
    Prodigy 100 points
    Hi again,

    i had to pause my LMP90077 project for a while but a few days ago I contiuned with this project.
    Now I am able to do a offset and gain calibration with feasible and reproducibly results.

    Therefore I have to preload the SCAL_GAIN register with 0x7FFF when applying the reference voltage to the selected measuremnt channel (not 0xFF_FFFF as suggested before). Furthermore I have to issue a stand-by command followed by an active command before gain calibration.

    Why do I have to restart the LMP90077 with a valid first sample before doing the calibartion? Or is there still an error in my application?

    Best regards,
    Michael
  • 0 in reply to Michael Hemetsberger94
    TI__Genius 16890 points
    Hi Michael,

    The reason that step 1 is included in section 9.3.1.2.2 is that you have to get rid of the offset error before calibrating out the gain error. It was put in as step 1 so that if someone went directly to that section they would know that they have to do the offset calibration first.

    Mike
  • 0 in reply to Michael Stout
    Prodigy 100 points

    Hello Mike,


    I already do the offset calibration before gain calibration. But I need to send a stand-by command followed by an active command before gain calibartion to get stable and plausible values for the gain coefficient.

    I am allowed to issue these commands (stand-by -> active) before gain calibration?


    Regards, Michael

  • 0 in reply to Michael Hemetsberger94
    TI__Genius 16890 points
    Hi Michael,

    Yes, that is ok to do. Going into standby does not erase the offset calibration value.

    Mike
  • 0 in reply to Michael Stout
    Prodigy 100 points

    Hi Mike,


    that's good news. Can you explain me, why I have to enter stand-by mode before gain calibration?

    I think this isn't standard procedure for gain calibration.


    Regards, Michael

  • 0 in reply to Michael Hemetsberger94
    TI__Genius 16890 points

    Hi Michael,

    As far as I know, you shouldn't have to enter stand-by mode before doing the gain calibration.  Is there something else going on with the part that going to stand-by mode causes to stop and then when you go back to normal mode it does not restart?

    Mike