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.

ADS1247: Follow up on Calibration response to un-used ADC channels

Part Number: ADS1247

Hello,

I have a follow up question pertaining to ADS1247. In my application I have 5 channels for temperature measurement and each channel has an ADS1247 chip. I am using the device for temperature measurements and using 4-wire configuration.

During testing, I have only 2 channels connected with an RTD (or a resistor). I perform calibration (SYSOCAL and/or SYSGCAL) and after calibration is complete, the channels (that don't have a resistor or RTD connected) behave differently. For e.g. the output value on those ADC channel keep fluctuating randomly and more often than not, they indicate a negative number (ECXXXX - 24 bit output from the channel). In one case, the channel gets stuck to a particular value. When I try to connect a resistor across any of the unused channels, the ADC does not respond with the expected value and behaves in a similar manner as if nothing is connected across it's terminals. 

After I perform the reset (clearing OFC and/or FSR register values), the "unconnected" ADC channels start working properly.

I wanted to understand the reason for the behavior and possible means to fix it.

Thanks

  • Hi Hemant,

    When the SYSOCAL is issued, there must be an external short applied to the inputs selected in the mux. SYSGCAL requires that a full-scale signal be applied to the inputs selected in the mux. Most likely you are calibrating something, but have pushed the calibration settings outside of a useful range.

    Here is what I would suggest. Use the SELFOCAL for removing ADS1247 device offset. This places an internal short at the mux and issues an offset calibration. The gain calibration is done at the final test of the ADS1247 for each PGA gain setting, so there really isn't a need to calibrate the gain. What you really want to do is calibrate the sensor, and you would do this in software for known conditions. One method is to use a 2 point calibration where you can determine a slope which is multiplied times the conversion result.

    Best regards,
    Bob B