• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 46 subscribers
  • Views 293 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.

LMP90080: LMP90080 Not responding / Question about thermal pad connection

Tim McCartney
Prodigy 145 points
Part Number: LMP90080

I'm using the LMP90080 in a high power DCDC application.  At higher power / noise the ADC stops responding.  It is likely I am getting too much noise on an ADC channel and/or on my power rails.  But upon reviewing the data sheet one thing I see is that the Pin Description table says to leave the thermal pad floating, however, later in the document the recommended layout shows it connected to GND.  I don't know if that pad is internally tied to GND or if I now have a floating connection picking up noise.  Certainly thermal performance would be better if I had tied the pad to my large GND plane.  So should that pad be tied to GND.  Is it tied internally? 

  • 0
    TI__Mastermind 49831 points

    Hi Tim,

    I found this excerpt from another thread: https://e2e.ti.com/support/data-converters/f/73/t/449623

    "We were told by the designers that the pad should be floating. They may be concerned about a parasitic gate being formed in the silicon when it is grounded."

  • 0 in reply to Alexander Smith
    Prodigy 145 points

    Ok, thanks.  Good to know I have the thermal pad connected properly.  I am still seeing the chip stop responding during high power operation.  I have a second LMP90080 on the same board in a different location that will continue working and both units talk to a SPI Isolator chip.  I am trying to figure out if my 5V rail is going to low or high during noise events or whether I am latching the chip up by driving the ADC inputs too high.  I can send a reset command and the chip will begin responding again.  I am going to try to reduce my sample rate.  I have it at 13.42 SPS and will change to 1.6 SPS.  I currently have the buffer enabled even though my gain is 1.  I would think this would be better for high noise environment, but I will try turning it off to see what happens.  Other than that I can maybe increase my bypass capacitance or add a tantalum (currently have 2 MLCCs), maybe that will snub the noise on the rail a bit.  Or I will try increase my filter capacitance on my ADC channels since I don't expect their values to change quickly since they are measuring temperature.  Anything else that might help?

  • 0 in reply to Tim McCartney
    TI__Mastermind 49831 points

    Hi Tim,

    I would observe the power rails and inputs as you've said to see if there is anything going on that would stress the part. Depending on that, is what would decide my next decision. Sometimes with these issues it ends up being that a lot of noise is being injected into the ground and extra capacitance helps to maintain the voltages at the device. I'm not sure if the sample rate reduction will have an impact, but it's worth a try. 

    If you end up taking scope shots I'm happy to take a look. Schematic and layout snapshot would help as well.