• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 46 subscribers
  • Views 1202 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

ADS1192 ADC conversion factor

Bharath G
Expert 1330 points
Other Parts Discussed in Thread: ADS1192, ADS1198

I am reading the ADC value from ADS1192 through SPI and trying to display on phone. What formula should I use to convert ADC read values to Voltages?

  • 0
    TI__Guru**** 168605 points

    Hi Bharath,

    Please review the Data Format section of the ADS1192 datasheet beginning at the top of page 25.  Your LSB size is dependent on the reference voltage you are using.

  • 0 in reply to Tom Hendrick
    Expert 1330 points

    Hi Tom,

    Thanks for the suggestion.

    In our app, If we use ADS1198 we are able to plot the ECG correctly by using the conversion formula according to ADS1198 datasheet. But with ADS1192 we are observing some kind of noise. And we are suspecting this may be because of conversion factor. Is it because of that or something else ?

    We are using this formula:

    ADC=(Vref  * gain / (2^15  -1) ) for ADS1198 and ADS 1192.

    What else might be wrong?

  • 0 in reply to Bharath G
    TI__Guru**** 168605 points

    Hi Bharath,

    How much additional noise do you see?  If you compare Table 1 in the ADS1192 versus the ADS1198 datasheets, you will notice that the input referred noise for the ADS1192 is about 2 times that of the ADS1198 at all gain settings through 1 ksps.

  • 0 in reply to Bharath G
    Expert 1330 points

    Here is an update.

    We took the data received from ADS1192 and plotted in MatLab. We found the graph is same as that on our app. This proves that data from ADS1192 itself is wrong. The graph is correct if simulator signal voltage is at 0.5mv. But if signal is at 1mv the graph contains a negative sample where its supposed to be a QRS peak. I mean to say one or two samples read are wrong. And that happens exactly on the QRS peak.

    Here: drive.google.com/open

    What might be possibly wrong?

    We are using 3.3V , Internal Clock and Continuous Read mode with gain set at 6 and 250 sample rate.

    Please help us solve this problem.

  • 0 in reply to Bharath G
    TI__Guru**** 168605 points
    Hi Bharath,

    If I read your screen shot correctly, your 'peak' value is essentially being turned upside down. I don't think this is a conversion factor or noise issue because all samples would be impacted, not just a QRS peak. It looks to me to be more like an issue where a binary 2's complement value is being wrapped around in your display. What does the raw data look like?
  • 0 in reply to Tom Hendrick
    Expert 1330 points
    Thanks for the support Tom,

    We found that we were not accounting the higher byte of the data in our plot.
    Unfortunately we did the same in Matlab. After a deep check we were able to find the problem and solve it.

    Bharath
  • 0 in reply to Bharath G
    TI__Guru**** 168605 points

    Great news!  I'm happy to hear you resolved the issue!