• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 45 subscribers
  • Views 181 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.

ADS1245: Strange phenomenon in low temperature

Katsu Matsunaga
Expert 7730 points
Part Number: ADS1245

Hello,

My customer has observed a strange phenomenon in -20degC.

AD_OUT got stacking LOW if 3.3V power turn OFF and turn ON after 2 sec from turn OFF.  If period of turn OFF is longer than 3sec, no issue was observed. And no issue occurs at room temp or high temp.

customer report is attached here/cfs-file/__key/communityserver-discussions-components-files/73/ADS1245IDGST-issue-in-low-temp.pptx

Question:

1. What is root cause?

2. Is there any spec of the duration LOW of 3.3V power (ON->LOW->ON)?

Please check the report and respond these question.

Best regards,

Katsu

  • 0
    TI__Guru** 113765 points

    Hi Katsu,

    I'm almost certain that the issue is related to the POR circuitry of the ADS1245.  We have seen this type of issue in the past with some of these older devices.  Our newer devices use a revised POR circuit that more quickly drains the internal circuitry so that the proper POR of the device takes place for brownout or On-Off-On situations.  The problem is worse at cold temperatures as device parasitic leakage is reduced, but at higher temperatures the leakage is higher resulting in the internal charge in the POR circuit to deplete more quickly.

    You can kind of think of the POR cicuit as a series RC element where both the R and C values are quite large.  Even though externally you can see the charge depleted on the supply pins, internally there is still some residual charge stored in the cap that takes a significant time for  the charge to deplete to below the POR threshold for the ADS1245.  So the easiest thing is to make sure you wait a significant amount of time for the charge to deplete.  You ask how long this time should be?  And this is difficult to determine as the total time is dependent on not only the ADS1245, but also on the caps and voltages applied to the ADS1245 inputs.  This is not something we have characterized with the ADS1245 as the information is really dependent more on the external system circuitry than just the ADS1245 alone.  

    One other possible option would be to issue a second SLEEP mode period after the first DRDY by holding SCLK high for the appropriate amount of time.  At power-up the device will be running and although the period and data are incorrect placing the ADS1245 into SLEEP after the completion of the first conversion should allow the device to properly reset when taken out of SLEEP mode. At least this is true for the analog section of the device and based on some similar devices I believe this should work as I expect.  

    When the SCLK is taken high immediately after power-up we do not yet know the state of the operation of the device.  Placing the device into SLEEP after we know the device is fully operating will be a better point to work from to evaluate the ADS1245 operating state.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    TI__Expert 7730 points

    Hi Bob-san,

    Thank you for your explanation.

    Let me confirm one about below comment.

    "When the SCLK is taken high immediately after power-up we do not yet know the state of the operation of the device.  Placing the device into SLEEP after we know the device is fully operating will be a better point to work from to evaluate the ADS1245 operating state."

    How do we know the device is fully operating at second power up?

    Best regards,

    Katsu

  • 0 in reply to Katsu Matsunaga
    TI__Guru** 113765 points

    Hi Katsu-san,

    As the most likely scenario is related to POR, I suspect (but unfortunately have no way to confirm at this time) that by causing the ADS1245 to restart after power is applied should correct the issue being seen by the customer.  When SCLK goes high the ADS1245 will shut down the analog and place the ADS1245 into a SLEEP condition.

    The customer has already conducted tests showing both the normal and abnormal operation.  If running in the abnormal condition and then putting the ADS1245 to SLEEP by placing SCLK high, the ADS1245 should respond normally again following setting SCLK back low waking up the ADS1245.  After the wakeup, the ADS1245 should return to normal operating behavior as was seen previously.  As there is no RESET pin or RESET command for the ADS1245, the SLEEP condition, which powers down most of the device internally, should place the ADS1245 in the proper operating state when taken out of SLEEP.

    Best regards,

    Bob B