• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 82 subscribers
  • Views 324 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.

TMS320F28335: TMS320F28335 Overtemperature actions

Mert Sahin
Prodigy 150 points

Part Number: TMS320F28335

Dear Sir/Madam,

We currently use 28335 on our low voltage motor drive platform. We suspect an overtemperature case occuring in our system and wonder what the DSP actions are in case it reaches temperature above its defined zone?

Are there any protection against over temperature, like closing down interfaces or so on?

Thanks a lot.

Mert Sahin

  • 0
    TI__Guru* 85590 points

    Mert,

    The F28335 doesn't have any intrinsic method to determine if it is operating outside the limit defined in its datasheet.  There are a few safeguards like missing clock detect or the internal watchdog that would initiate a device reset if not serviced; but those may not fail at the same time other parameters go out of spec with a violation in operational temperature.

    The only way would be to use an external temperature sensor, and feed that information into the device for detection ahead of a possible violation.

    If you could give an idea of how much you think the temperature is over the datasheet, and what behavior you observe with the device I could give my opinion on the potential effect.  While the DS doesn't guarantee operation outside its limits; there are instances where logic will be out of spec, vs just outright non-functional.

    Best,

    Matthew

  • 0 in reply to MatthewPate
    Prodigy 150 points

    Hi Matthew,

    Thanks a lot for your speedy response.

    Actually we fit in a IP66 cabinet with no forced air cooling. Under overload conditions the inside temperature reaches to +128C and we observed a situation like this.

    There occurs a reset, then motor align current flows. After that the open loop current flows and there occurs another reset. This happened 10+ times, so we thought we reached to temperature limitations of the DSP. Could you please give your further comments on this issue?

    Best Regards,

    Mert

  • 0 in reply to Mert Sahin
    TI__Guru* 85590 points

    Mert,

    If you are seeing a reset occur regularly from the MCU, then likely the watchdog is overflowing due to not being serviced in time. 

    If your ambient temperature is 128C with no airflow, then we need to determine the Tjunction of the C2000.  I believe there are some calculations you can do to extrapolate this based on the current consumption of our device, based on the PsiJT listed in the datasheet(starting on page 38).  

    If we are above 150C Tj, then it is a thermal issue.  If we are well below that, then I'd suggest there could be another issue occurring at temp, perhaps a variable that is above an assumed max causing the code to do undefined things, etc.  The watchdog would catch those as well.

     Its hard to say absolutely what can go wrong outside of the allowed operational conditions, but in this case it could be a mis-decoded instruction that dis-rupts the normal program flow, or a flag check that never completes due some other timing condition.

    When temperatures are too high, we reduce the timing margin of the various logic in the MCU, and signals that used to arrive within a CPU cycle don't, etc.  In all these cases the watchdog is doing it job to reset the device to keep it from being in that unknown state for too long.  

    Let's see what the calculations tell us about the range to Tj and we can go from there with mitigation.

    Best,

    Matthew