• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 45 subscribers
  • Views 482 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.

ADS114S08: Calibration theory

Shinya Sawamoto
Expert 5285 points
Part Number: ADS114S08

Hi team,

I'd like to understand the theory how the self calibration is implemented in the device and the procedure how to operate.
Do the SYOCAL and SYGCAL for a manual calibration command with OFC and FSC registers?
How does SFOCAL perform for a self offset calibration internally on the device?

Thanks in advance.
S.Sawamoto

  • 0
    TI__Guru** 113765 points

    Hi Shinya-san,

    The calibration information is in the ADS114S08 datasheet in section 9.3.14.  This can be a rather confusing topic so I will try to give some further detail.

    There are two types of offset calibration.  There is the internal self-offset calibration (SFOCAL) and the system offset calibration (SYOCAL).  In most all cases you would choose to use the self-offset calibration.  The self-offset calibration uses an internal short within the ADS114S08 mux to set the ADC inputs to mid-analog supply.  The ADC takes the number of conversions as specified in the SYS register (CAL_SAMP bits default of 8 samples) and averages the result using the PGA, digital filter mode, reference and data rate settings configured for the device.  The expected result would be 0, so the conversion result will be placed automatically into the OFC registers to then be subtracted from every conversion result thereafter.

    The system-offset calibration (SYOCAL) would be similar using the same number of sample averages and placing the result in the OFC registers.  The difference is that the user must place an short externally within the correct input range for the PGA  settings.  This can be a very complicated procedure in some cases requiring shorting jumpers.  Where the SYOCAL is useful is if there is an external amplifier or sensor that has significant offset where the correction will simplify the process of removing the offset.  I think that it is often easier to use the SFOCAL and then store any sensor offset in microcontroller firmware/memory.  In this way periodic SFOCAL can be issued to remove any offset drift of the ADC over time, whereas the SYOCAL cannot easily do so.

    There is no internal self-gain calibration as gain is adjusted at final test to provide the gain accuracy specification shown in the electrical characteristics table for the internal reference.  SYGCAL can be valuable but you must apply a full-scale signal to the ADC inputs and the process is similar to the offset calibration but instead places the correction factor into the FSC registers.  Again this can be a complicated procedure with switches or jumpers applied for the calibration which can also add error.  Also as most external references are often used in ratiometric measurements, this calibration is unnecessary.  Again due to drift, the SYGCAL may not be the best method to use.

    So it really depends on the sensor and the desired response.  Usually 16-bits does not require the same level of offset and gain correction as would a 24-bit device.

    Best regards,

    Bob B