• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 47 subscribers
  • Views 426 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.

AMC1100: Latch-up?

Richard Glover
Prodigy 40 points
Part Number: AMC1100


We have an application with thermocouple and AD8495 and buffer on isolated side. With no probe, AMC1100 sees 1.5V differential with pin 3 on floating ground.

We expect fully saturated positive output, but occasionally see fully saturated negative output. Both supplies look good. Nothing appears to be outside range on data sheet. This appears to persist for some time (minutes typically).

Is there a required sequence for the 2 supplies?

Is there anything else known about spurious phase inversions when significantly over driven?

  • 0
    TI__Guru**** 181341 points

    Hi Richard,

    There is no particular supply sequence for the AMC1100.  Can you expand a little on the events leading up to the saturated outputs?  If you are removing VDD1 or just powering up VDD2, there is a possibility that you could have either full-scale positive or full-scale negative on the outputs.

  • 0 in reply to Tom Hendrick
    Prodigy 40 points

    We see this in powering up with large (+0.7V) input. Sometimes output is large negative rather than large positive. Setting input to zero momentarily does restore correct operation.

    Clamping input to 0V prevents this happening.

    It is not related to removal of either supply. Input supply after power up is stable at 4.9V.
    Hope this gives more of a clue.
    Thanks for the response.
    Richard Glover 
  • 0 in reply to Richard Glover
    TI__Guru**** 181341 points

    Hi Richard,

    Thank you for the added detail.  When you say you see this with 700mV on the input, that would be the first item to address.  If you take a look at page 3 of the AMC1100 datasheet, VINP/N should not exceed VDD1 by more than 500mV.  Do you actually have to remove VIN and power cycle VDD1 before getting back to normal operation?

  • 0 in reply to Tom Hendrick
    Prodigy 40 points

    Hi Tom,

    Thanks again for he response.

    Our 700mV is potted from supply to VDD1. It never (ie cannot) exceeds VDD1 or drops below GND.

    Removing VIN alone is sufficient to bring it back to proper operation.

  • 0 in reply to Richard Glover
    TI__Guru**** 181341 points

    Interesting, I've never come across a situation like this so the more detail you can give me the better!  If you hit this 700mV transiently and go back to zero or somewhere within the normal operating range, does it work normally or still remain 'stuck' until you physically remove Vin?  If you can share a schematic or waveforms, that would be great.