• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Subscribers 47 subscribers
  • Views 748 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.

ADS131M08: ADC Offset

Adla
Prodigy 80 points
Part Number: ADS131M08

Hi,

I am measuring offset error for the first four-channels of ADS131M08. Global chop mode is always active and PGA  is set to 32. Firstly using calibration registers I shorted ADC inputs to AGND and measured around 2µV to 15µV. Which is I think is acceptable according to the datasheet (Section 6.5 Electrical Characteristics). Now I applied 0V externally to ADC and measured around 0.005mV at the input of ADC. When I look at the output code of the ADC it is giving me 1.4mV for channel-0, 731µV for channel-1, 639µV for channel-2, 274µV for channel-3. I expected the offset error will differ from channel to channel but will remain under +-32µV because global chop mode is enabled. From my measurements that is not the case and would like to know what might be causing this behavior.

Thank You

Best Regards

Adla

  • 0
    TI__Guru 55152 points

    Hi Adla,

    Offset error is input referred which means that when measuring from the output, the result should be divided by the gain of 32. 

    This explains the difference in expected vs measured values for channels 1-3, but doesn't explain why you are reading 1.4mV for channel 0. 

    There are a few possibilities to explain this difference such as the voltage on the inputs was not settled, the instrument used to measure 0.005mV has some error associated with the measurement, or the offset calibration was performed/loaded incorrectly. 

    Are these measurements constant and re-producible? If you re-do the calibration does the issue persist? 

     

  • 0 in reply to Alexander Smith
    Prodigy 80 points

    Hi Smith,

    Yes, the measurements are reproducible and constant. I am able to calibrate out offset but the question is why I am seeing different offsets on different channels in the first place especially the first channel which is double the value of channels 2 and 3. Is it something related to the input impedance of the channels?

  • 0 in reply to Adla
    TI__Guru 55152 points

    Hi Adla,

    The ADS131M08 is a simultaneously sampling converter, meaning that each input channel has a different analog signal chain. Small differences in the manufacturing of each signal chain accounts for the differences in offset for each channel. 

  • 0 in reply to Alexander Smith
    Prodigy 80 points

    Hi smith,

    I attached the image of my schematic. From the image, you can see there is only an RC Filter before ADC. For the first four channels(0-3) I used the same configuration. I know that the RC filter also effects offset error. Now can I say that the difference between internal offset correction and external offset correction is only RC filter? When I measured offset by shorting ADC inputs to AGND using calibration registers I measured 2µV to 15µV. When I measure externally I measured 1.4mV for channel-0, 731µV for channel-1, 639µV for channel-2, 274µV for channel-3 as stated in my original question. This high offset is caused by an external RC filter?

  • 0 in reply to Alexander Smith
    Prodigy 80 points

    Sorry smith I could not upload the image due to some issue. I created a case for that too with customer service. Is there any other way I can send you the image?

  • 0 in reply to Adla
    TI__Genius 9880 points

    Hi Team,

    I'm uploading the image in behalf of Adla. 

     

    Regards,

    Maynard

  • 0 in reply to Adla
    TI__Guru 55152 points

    Thanks Maynard.

    Hi Adla,

    Yes, the RC filter (and the tolerance of the resistors) will have an impact. It also depends on what your input RC filter nets are connected to, which are not shown in the schematic. The bias current of typ +/-1uA will flow in/out of the ADC inputs to AGND. These bias currents flowing through resistive paths on their way to AGND will create voltage drops across the resistances in their path and any difference in resistance between the positive and negative bias current paths will create a differential offset voltage seen at the input of the ADC. 

    Internal offset calibration takes care of the inherent offset that the ADC has, however it does not account for the offset of the external components since it is performed by shorting the inputs of the ADC to AGND. So the external circuitry is disconnected when the internal offset calibration is performed. 

    If better offset performance is desired, instead of performing the internal ADC offset calibration I would recommend performing the offset calibration in your uC. To do this, short the inputs of the whole signal chain, take and record measurements for each channel, then apply the inverse to your measurements during operation. 

    Sorry about the issue with uploading images. I typically use either .png or .jpg file extensions and either drag and drop into the message window or use the "Insert" -> "Image/video/file" button to upload.