• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 47 subscribers
  • Views 601 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.

ADS7953: Data errors?

Will Jarrett
Expert 3825 points

Part Number: ADS7953

I have a customer who uses the ADS7953 in several designs that has been in production for a year or so.  They are seeing (2) specific issues with this part that happen very rarely, however, are causing us a fair bit of stress. 

We are not 100% sure if the issue is with the ADS7953 or somewhere else.

The first issue is – once in a great while, all the channels return double the ADC channel input voltage value.  It does not happen very often, but when it does, all the channels read double the input voltage over the SPI bus.  We, as default, always set the part up for 2*Vref, so we can read 0 to 5V.  Is there any reason for this?

 The second issue is – once in a great while, when we read say CH 0, we get back CH 1 values, read CH 1 get CH 2 values and so on.  What could be causing this?

 This ADC interfaces over SPI with a AM3352BZCZD60.  Initial review is the SPI bus looks good.

Please let me know what other information I can provide to help figure out what is going on here.  Thank you! 

  • 0
    TI__Mastermind 25070 points

    Hello,

    This is odd. Has any changes been done since this has gone into production? A change in manufacturer, equipment, or contractor?

    1. This is odd. If the input is sure to be known at the correct input voltage, this could be do to something digital. Reading double means that someone there is a shift left of the data. This could be due to timing issues, reading incorrectly, interpreting data incorrectly.  I would suggest looking at this through this angle. 

    2. This seems to also be related to timing issues. I predict the previous issue and this are both likely do to something going on the digital bus. 

    I would suggest diving into the digital bus. You mention that the SPI bus looks good. Have you taken a scope shot to confirm the data is as expected, no timing  errors, or glitches? This is where I would first start

    Regards

    Cynthia

  • 0 in reply to Cynthia
    TI__Expert 3825 points

    Hi Cynthia,

    Would you be able to support a call with my customer

    Thanks!

    Will Jarrett

  • 0 in reply to Cynthia
    TI__Expert 3825 points

    Also from customer:

    We've learned a little more about these 2 error modes.

    Using a SPI bus analyzer with the ADC in the error state, we can verify that we have selected the 2*Vref input range and that we are getting ADC readings that are double the expected value. It takes 100 or more cold starts to get the system into this state.

    We received some better data regarding the channel mismatch issue - the reason we're detecting "wrong channel returned" is because the ADC always returns 0xffff. Since none of the channels we were monitoring was channel 15 (0xf), software said we had a channel mismatch. Also very hard to reproduce this one.

  • 0 in reply to Will Jarrett
    TI__Mastermind 25070 points

    Hello Will,

    Yes please feel free to contact me directly to continue moving this forward

    Regards

    Cynthia