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.

DAC8568: Half output voltage and bad VREF

Part Number: DAC8568

Dear Support Team,

We are using 2 DAC8568ICPW in our design.

One of them supplies the VREF to the other one and also to the ADC used in our system.

We noticed that sometimes one of the DACs output is only half of the output we set it to be.

If the DAC affected by this is the one that supplies the reference to the other circuitry we also see that our ADC measurements gets shifted because the VREF is not 2.5V anymore.

After a powercycle everything goes back to normal.

Unfortunately the boards are built into a system and we are not able to measure anything, we can only assume these things happen from our internal measurements.

We already have filtering on the AVDD and the VREF lines with ferrite beads.

AVVD is 5V.

Do you think this issue can be caused anything else than noise like how it happens in this blogpost:

https://e2e.ti.com/support/data-converters/f/data-converters-forum/930509/dac7568-half-output-voltage-after-applying-noise

 

Best Regards,

Pál

  • HI Pal,

    Are you able to monitor the VREF pin during this behavior? Can you share your schematic?

    Thanks,

    Paul

  • Hi Paul,

    Unfortunately, we cannot monitor any signals, because the boards are built deep into our system. We tried reproducing it in our lab without any success.

    Here is the schematic:

    Some more info about our situation:

    We have 32 boards in 1 building block of our system, and 6 of these blocks in total, so it means 6*32*2=384 DACs

    This issue only comes in the fully assembled system, where noise can really be an issue.

    This thing happens randomly, we could not find a pattern on which of the 384 DAC goes bad at what time. And most of the time they work good, but its annoying when it happens because we have to power cycle the whole system to get it fixed.

    Thank you,

    Pál

  • Hi,

    I don't think the issue is related to the one you mentioned in e2e. That was related to IEC high frequency test.

    Like Paul mentioned, any chance to monitor VREF during this event?Also in your schematics, I can see GNDA and GND nets, how are they connected in the layout?

    Regards,

    AK

  • Hi AK,

    The GND and GNDA is connected with a 0 Ohm resistor on the board. However it is a bit far from the DAC on the actual layout. It is surely a possible improvement for the layout to make the 2 grounds better connected.

    As I said it is very difficult for us to measure the VREF during operation, but we will see if we can do something to be able to measure it.

    Regards,

    Pál

  • Hi,

    I will wait for the REF results.

    Also as a debug, Can you power down the other DAC's in the board and check for this behavior?

    Regards,

    AK

  • Hi Pal,

    I believe that there is a separate email chain with some field engineers supporting this.  I suggest we use that thread for the time being.

    Thanks,

    Paul