• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 1867 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.

UCD9090: Fault Response - Test simulation to Validate the options

Felix Arul Rajesh Francis
Intellectual 470 points
Part Number: UCD9090

Hi,

We implemented the fault response - Glitches, Restart both for OV(5 times), UV(1 time) & Max Fault Turn On(5 times)

We would like to validate these implementation.  

Could you please suggest a way or test case simulation to test and validate these fault responses?

Kindly refer the attached project file.UCD9090 2.3.5.0 Address 101 Project File_02282017.zip

Regards,

Felix.

  • 0
    TI__Guru 65750 points
    Hi Felix
    You can use TI EVM to hoop various power supply to test. or you can manually adust the voltage threhsold including the OFFSET to generate fake fault to test.
    BTW, Do you have a local FAE to support your commany?
    Regards
    Yihe
  • 0 in reply to yihe
    Intellectual 470 points

    UCD9090_Resequence_Restart_Testing.zipHi Yihe,

    We are continuing this fault simulation test and trying to validate the Re-sequencing and  Restarting features.

    We also getting help from Local FAE support too.

    Also, thanks for the FAQ published. That is really helpful for our testing.

    We would like to share our observations and queries and get some more clarity with our results. Kindly bare with me on more details in this.

    1. We have a system requirement that even if one rail sees a fault, all the rails should be shut down.

    For this we are selecting all other rails as fault slave shutdown.

    At any situation, when the faulty rail shuts down, we don’t want to have any other rails alive at the board.

    If we use the restart feature, Will that satisfy this requirement?

    What we understood from the datasheet and FAQ, restart feature is a rail specific.

    Only after the number of restarts exhausted, when we select the fault slave shutdown option, UCD will shut down remaining rails.

    We are seeing that the fault slave shutdown rails are alive at the board, while the faulty rail performs its restart.  If this is the case,  at any situation, we should not select the restart feature for our application.

    For your information, we didn’t enable the re-sequencing option while testing this.

    2. At the same time, when we have the re-sequencing option alone, during the Over voltage fault, we could see all the rails are going off and again re-sequencing as per the On-sequence. It repeated re-sequencing as per the set number and after exhausting the number of re-sequences, if still fault exits, it shuts down all the rails.

    This suits to our requirement.

    3. But, when we have the re-sequencing option alone, during the Under-Voltage fault, we could see only one re-sequencing and after that the rails are not shutting down. All rails are existing, including the faulty line. The preset number of re-sequencing is not happening in the UV case.

    Kindly refer the attached word document with wave forms attached to support this findings.  Also, we attached the UCD Image file for your testing.

    Kindly give your inputs on these findings.

    Regards,

    Felix.

  • 0 in reply to Felix Arul Rajesh Francis
    TI__Guru 65750 points

    Hi Felix

    Based what described in the word file, the issue was present with UV case. 

      In the word file, you said that the failed rail did not cross POWER_GOOD threshold after either resequencer or re-start. Here my understanding is that when UV fault was detected, UCD shutdowned the fault rails successfully(the faulted rail should be down to 0V or with some very small volatage), but when UCD tried to restart it or resequencer it, the rail failed to power up corrected(below POWER_GOOD threshold) after EN is asserted. is this correct? it seems that rails has problem to restart somehow. I did not see any wrong from UCD point view.

    If you expected a rail should be power-up after EN signal is asserted, a TON_MAX timer setting should be set. If the rail does not reach power_good within TON_MAX, a TON_MAX fault could be triggerer, a proper response could be set to turn off all rails.

    Please let me know if what i understood is wrong.

    Thanks

    Yihe

  • 0 in reply to yihe
    Intellectual 470 points

    Hi Yihe,

    Thanks for your input and the consequent conference call along with the TI local FAE, we could resolve the issues that we were facing.

    Specially we understood that the Restart feature as fault response wont be work out for our requirement, as we would like to have all the board rails need to shut down and restart whenever there is a fault in any of the rails. So, we end up using Re sequencing for this and it helped us to bring out our requirement

    We used the TON_MAX_FAULT delay we set to over come the UV fault re-sequencing scenario.

    Once again thanks a lot your support.

    Regards,

    Felix.