• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 81 subscribers
  • Views 767 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.

TMS320F28035: ADC Calibration

Stuart Baker
Genius 10790 points

Part Number: TMS320F28035

I'm working with the F28035 device family, and have a few questions about calibration.

  1. How often should calibration be run?  I assume that it should be performed whenever a change in temperature is observed, since it is factory calibrated at 25C.
  2. The algorithm says to "Perform multiple conversions on B5..."  but how many is multiple?
  3. While executing the self-cal, can the other channels still be used normally?
    1. I wonder if the ADCOFFTRIM would corrupt the results of other channels.

Thanks,

Stuart

  • 0
    TI__Guru 62975 points

    Stuart,

    Here is a discussion on periodic calibration for reference:  https://e2e.ti.com/support/microcontrollers/c2000/f/171/t/488749

    How often should calibration be run?  I assume that it should be performed whenever a change in temperature is observed, since it is factory calibrated at 25C.

    The amount of offset shift varies by device, but I think a good arbitrary starting point is to assume about 1 LSB shift per 4 degrees C.  With that anchor point, the frequency for calibration then depends on the error tolerance of the system and the expected change in temperature for the device.

    The algorithm says to "Perform multiple conversions on B5..."  but how many is multiple?

    The number of conversions should be large enough to minimize ADC errors from sources like quantization and noise in the system.  It will require some trial and error to determine -- the goal is to achieve repeatable results.

    While executing the self-cal, can the other channels still be used normally? I wonder if the ADCOFFTRIM would corrupt the results of other channels.

    The ADCOFFTRIM register affects all ADC channels so there will be some known, bounded offset error while the self-cal procedure is executing.  If you follow the recommendation of targeting +1 LSB offset in the thread referenced above, the disruption can be minimized.

    -Tommy