• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 47 subscribers
  • Views 303 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: Inaccurate precision

Hui Cao22
Expert 1270 points

Part Number: ADS131M08

Tool/software:

When we were using the ADS131m08, the accuracy suddenly became inaccurate. For instance, when the source output a 5A AC signal, the meter could initially test the correct value of 5A normally. After running for a period of time, it showed 5.08A. When the source output was 4A, the meter tested a value of 4.03A. After troubleshooting, it was confirmed that there was no problem with the source output. At this point, the ADC data conversion rate is not an issue. It has always been the 8K rate we configured, and the SPI data frames captured are also normal. It can be restored after the watch is restarted (the board is powered off and restarted). This phenomenon occurs with a low probability in our area. Currently, we can't find the cause. I would like to ask how we can further locate and investigate it. By the way, the two values mentioned above, 5.08A and 4.03A, are not the results of the default gain value (8388608).

  • +1
    TI__Mastermind 49425 points

    Hi Hui Cao,

    Did you measure the ADC's input when the 5.08A or 4.03A were observed? Did this happen during a special testing(e.g. EMC)? Can you share the raw data (original data) from the ADC?

    BR,

    Dale 

  • 0 in reply to Dale Li
    Expert 1270 points

    1. The input at the ADC end was not tested when the problem occurred. 2. It was either tested in a special environment or in a regular test. 3. The original data was not captured. We have also been reproducing this issue recently. By the way, our code checks every few data frames to see if the frame header is valid. When the problem occurred, no abnormal frame header prints were seen.

    Next, we plan to review the ADC register value and the value of the original code converted into a physical signal when the problem occurs to see if it is normal.

    May I ask if there is anything else to do when the problem occurs, or what we need to do under normal circumstances

  • +1 in reply to Hui Cao22
    TI__Mastermind 49425 points

    Hi Hui Cao,

    1. I strongly suggest to monitor the ADC's input, this can tell you whether the interference or issue comes from the front-end circuit or not.

    2. This could happen when the circuit including the ADC or the front-end circuit is affected by an interference or coupling. More experiments including input monitoring and raw data can help to identify. The schematic and layout can also help.

    3. I do not know what the frame header is, you may clarify it.

    Except item 1 and 2, monitoring Vref on REFIN will help to identify as well.

    BR,

    Dale

  • 0 in reply to Dale Li
    Expert 1270 points

    The frame header is actually the value of the status register. When reading data from 8 channels each time, the first three bytes are the values of the status register. many thanks for your kindly supports, we will monitor the ADC's input.

  • 0 in reply to Hui Cao22
    TI__Mastermind 49425 points

    Hi Hui Cao,

    Thanks for your clarification. We do not call the STATUS word as the frame header.

    -Dale