• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 29 subscribers
  • Views 493 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.

Watchdog using PIOSC vs XOSC and watchdog locking up on XOSC failure.

John Osen
Genius 3585 points
Other Parts Discussed in Thread: TM4C123BE6PZ

I am using a TM4C123BE6PZ.  I am using an external oscillator for the system clock and WATCHDOG0 for my watchdog.  WATCHDOG0 also uses the system clock.

We are now testing noise susceptibility.  When we hit the device with 100MHz, it locks up.  A peripheral chip with its own clock keeps working.  A sister device that uses PIOSC as the system clock keeps working.  Power cycling the locked up unit returns it to normal operation.

We are assuming that the 100MHz noise is locking up the external clock and because the WDOG is based on the external clock, the WDOG locks up as well.

If I use WDOG1, using the PIOSC, will it still reset the processor even if the processor is locked up because XOSC is locked?

This change is hard to test as it requires testing at an offsite shielded facility.  I would like higher confidence this is the correct fix.

I am not an electrical guy, but are there XOSCs that are more noise immune?

  • 0
    TI__Expert 3555 points

    Hi John,

       What part are you using for the XOSC input?

    What are you doing when you say you are hitting the device with 100Mhz?

    Can you put a scope probe on the XOSC input to confirm that it has indeed stopped oscillating?

    Setting up the WDOG1 to use PIOSC will not necessarily help you if XOSC is locked.  You still need to have a way to get that started again.

    - Ken

     

  • 0 in reply to Stellaris Ken
    Genius 3585 points

    Stellaris Ken said:
    What part are you using for the XOSC input?

    Our HW engineer gave me these specs.  I hope they are sufficient to answer your question.  And I mis-spoke: it is the OSC input and not the XOSC.  We have both OSC (20MHZ) and XOSC (32.kHz)

    Operating Temp: -40C – 85C

    Freq Tol: 20ppm

    Stability:  30ppm

    Load Capactiance 18pF

    ESR: 40ohm

    Stellaris Ken said:

    What are you doing when you say you are hitting the device with 100Mhz?

    We are conducting noise immunity testing.  The testing is at a TUV facility that is remote for interference issues with planes and such.  An antenna is positioned 100meters from our machine and the machine is flooded with interference over a range of frequencies.  The unit fails in the 100MHz range.  The range goes beyond 2.4GHz.

    Stellaris Ken said:

    Can you put a scope probe on the XOSC input to confirm that it has indeed stopped oscillating?

    Unfortunately at this stage we are blind.  The device is sealed up and potted.  We have two inputs: CAN on the Tiva chip and a seperate third party wireless module.  The wireless module is still ack-ing messages.  The Tiva part is not responding to CAN messages.  The wireless chip being active indicates that the Tiva chip has not gone into hibernation, which would drop power to the wireless chip.

    Cycling power to the device restores operation.