• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 46 subscribers
  • Views 425 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.

ADS7828-Q1: ADC IC not working until power cycle

Kiran S
Prodigy 50 points
Part Number: ADS7828-Q1
Other Parts Discussed in Thread: ADS7828

Hi,
I am facing issues in I2C Communication with ADC IC ADS 7828.
On my custom board, the microcontroller is connected to the ADC IC via an I2C isolator.
Most of the time the I2C Communication is working fine (400 KHz).
Sometimes (once in a week), we observe I2C communication failure (randomly) while the board is in use.
The failure does not go away until we power cycle the board.

We don't know what causes the problem, so we are not able to reproduce it on a test bench.
Note: My board does not have power control for the ADC IC alone.

Is there a way to recover from such a failure by just using the I2C bus?

I've attached images of I2C bus sections from the schematic.

Any inputs on this issue will be a lot of help.

Regards,
Kiran

  • 0
    TI__Guru**** 175081 points

    Hi Kiran,

    The behavior you are describing with the ADS7828 sounds like a latch-up condition to me.  I don't see anything on the controller side of your schematic that might be an issue, but if there is anything on the input side that momentarily exceed the datasheet Vin limits, that could be the root cause of the problem you are facing.  

  • 0 in reply to Tom Hendrick
    Prodigy 50 points

    Hi, 

    Apologies for the delayed reply and I have got an update on this issue. I am attaching the images of SDA(green ) , SCL(yellow) waveforms captured on oscilloscope with bus pullup resistor 1k ohm. It can be observed that there are undershoots and the numbers are -1.675V  for 59.6ns. Can you confirm whether this is what causing the latch up  on ADC?. 

  • 0 in reply to Kiran S
    TI__Guru**** 175081 points

    Hi Kiran,

    I suspect that could be possible.  Do you have any droop of the 5V supply or spikes on the analog inputs?

  • 0 in reply to Tom Hendrick
    Prodigy 50 points

    Hi Tom, 

    I don't see droop in the Vdd/Vref rail connected to the Ext ADC, attaching the snap of Vdd measured using oscilloscope. And on the analog inputs I didn't find any spikes, I have zener diode for protection too in case as a part of safety. 

    Can you Kindly check  how long the negative voltage at the  I2C bus the Ext ADC can sustain, in my case its -1.675V  for 59.6ns. I have added damping resistors but no improvements on the undershoots (I don't have option to add schottky diodes).  

    Regards, 

    Kiran

  • 0 in reply to Kiran S
    TI__Guru**** 175081 points

    Hi Kiran,

    I don't have any good numbers for you regarding 'how long' a negative going pulse would need to be.  Since you see this pretty regularly in normal operation, perhaps it gets worse (once a week) at times.  What about your layout?  How are the grounds between the isolator and the ADS7828 connected?

  • 0 in reply to Tom Hendrick
    Prodigy 50 points

    Hi Tom,

    attaching a snap of the layout between isolator and ADC, Layer 1 RED, Layer 2 YELLOW (other layers are hidden) . I2C bus lines are impedance matched to 60ohm, layer 2 shown below is connected to the ground which is also reference for Isolator and the ADC.

    Regards, 

    Kiran

  • 0 in reply to Kiran S
    TI__Guru**** 175081 points

    You have me scratching my head...

    Is the Isolator absolutely needed for safety reasons?  Can you pull that out and short SDA, SCL and GND from input to output and try to see if a.) the negative spikes are diminished or gone and b.) if the random lock up condition stops?