• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 82 subscribers
  • Views 231 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.

Original question:

TMS320F280049: CMPSS issue

TMS320F28377D: CMPSS False Tripping

Noah Erickson
Prodigy 20 points
Part Number: TMS320F28377D

Hi E2E,

I have an inverter using a TMS320F28377D, configured to use the digital comparators for phase over-current detection as well as DC bus over-voltage. Running at a light load on this inverter and everything behaves fine. Additionally, previous designs using the same circuitry and configuration, everything behaves fine. Once I get to a certain operating point, at relatively "high" voltage for our application, I start to see all four of these comparators trip at the same time. I should note I'm running at about one third of my expected trip point for current and at about two thirds of the trip point for voltage.

I've done a few different things for debug so far. The first is turning off the trip zone functionality and datalogging how many times I get the digital comparators tripping, by immediately clearing them if they trip. The comparators trip proportional to my load, which sort of would indicate noise to me.

Is there some mechanism or means in the digital comparator block or the Xbar block that might cause these to all trip at the same time under some condition?

Regards,

Noah E

  • +1
    TI__Mastermind 33116 points

    Hi Noah,

    Noah Erickson said:
    Is there some mechanism or means in the digital comparator block or the Xbar block that might cause these to all trip at the same time under some condition?

    I think that if the input voltage is outside of the maximum specifications this may be possible, but I think it is more likely that there is some noise that is affecting all inputs simultaneously.

    Have you scoped the inputs to the comparators to see the voltage inputs when they trip? 

    Best Regards,

    Ben Collier 

  • 0 in reply to Benjamin Collier
    Prodigy 20 points

    Hi Ben,

    Thanks for the reply.

    For clarification on the input voltage, you are referring to the comparator input on the ADC input pins, correct?

    I haven't scoped the inputs yet, but I do have a circular buffer saving ADC sampling data at the comparator trip event for the last ~150 PWM cycles. Phase overcurrents and DC bus voltage look great, not showing even a hint of noise on any of the channels. Relatively unfiltered, but understood that the noise event could be higher frequency and scoping actual ADC inputs is our next step.

    On the theory of noise, I have my comparators configured to trip using the digital filter, which if my understanding (and setup is correct) should trip after ~27us, using a clock prescalar of 210, sample window of 30 and a threshold of 26. Meaning any noise would need to be in the ballpark of less than ~37kHz, not to mention nearly a volt of amplitude (trip lo is 40 counts, trip hi is ~3600 for phase currents) on the ADC input.

    Another interesting data point is I am also logging motor position (resolver) in the circular buffer and this event seems highly correlated to a specific angle. That is, when the event occurs it's within a few electrical degrees, but it doesn't happen every time it crosses that position.

    Regards,

  • 0 in reply to Noah Erickson
    TI__Mastermind 33116 points

    Hi Noah,

    You make a good point about the noise needing to be fairly low frequency if you are using the digital filter. Still, I am curious about what you will see if you scope the comparator input pins.

    Best Regards,

    Ben Collier

  • 0 in reply to Benjamin Collier
    Prodigy 20 points

    Hi Ben,

    I wanted to close the loop on this. Our hardware engineeers played a joke on me. Under certain conditions we were generating a noise spike that was getting onto our phase voltage feedback wires, overvolting the comparators and causing this to occur.

    Thanks for pointing out this condition could occur if we overvolt the comparators, that was extremely helpful. Issue resolved.

    Thanks,

    Noah