• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 82 subscribers
  • Views 279 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.

TMS320F28374S: TZ working or not after code crash

Jordan Zhou
Genius 10420 points

Part Number: TMS320F28374S

Hi,

Customer faced one issue from the code, in a special case, the code crash. We don't know where the program point is. 

But, the clock is still there. WD is not enabled. 

We'd like to know, in this case, does TZ and CMPSS still work or not?

I surpose Yes.

Thanks a lot.

Br, Jordan

  • 0
    TI__Guru* 83050 points

    Jordan,

    Based on my understanding of the CMPSS/TZ modules this path is purely in HW so they should still be working in this condition. 

    A concern would be if the code is in an unknown area what potential it has to disable these paths, etc.  However, with the EALLOW protection this should prevent a rogue write from code if it were to occur.  Furthermore, depending on how the customer has configured the logic, there may not be a need for a clock for this to work.

    As you said the WD module is there to help contain these kind of events and limit how long they will propagate.  I would definitely suggest that customer uses WD to limit their exposure to a code crash(whatever the cause).

    Best,
    Matthew

  • 0 in reply to MatthewPate
    TI__Genius 10420 points

    Matthew,

    Some other doubts about TZ:

    1) TZ5 is used to switch off PWM after system clock missing. How long does it take, from clock missing detection to PWM off?

    2) If system clock missed, internal 10MHz is used instead. In this case, other TZ like TZ1, can it still work? Can it detect the signal on the TZ1 pin and switch off PWM?

    Thanks a lot.

    Br, Jordan

  • 0 in reply to Jordan Zhou
    TI__Guru* 83050 points

    Jordan,

    1) TZ5 is used to switch off PWM after system clock missing. How long does it take, from clock missing detection to PWM off?

    A1)In this case the biggest time factor is the missing clock detection itself.  Assuming this occurs at the beginning of the counter cycle it would take ~82us to detect the missing clock(1/10MHz * 8192).  The time for the TZ to trip after it triggered is likely in the 20-30ns domain.

    2) If system clock missed, internal 10MHz is used instead. In this case, other TZ like TZ1, can it still work? Can it detect the signal on the TZ1 pin and switch off PWM?

    A2)This is correct, the comparators and the TZ logic can be configured in async mode(I beleive this is the default), in which case they would trip in the same amount of time(50-60ns total time) using external, internal, or even with no clock.

    Best,

    Matthew