• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 97 subscribers
  • Views 370 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.

TMS320C6657: DSP hang-up

Koichiro Tashiro
Mastermind 22945 points

Part Number: TMS320C6657

Hi,

My customer reported DSP hang-up issue.
When the issue happens DVDD15 current is ~70mA smaller.
Please see attachment for details.
DSP hang-up.pdf
Thanks and regards,
Koichiro Tashiro

  • 0
    TI__Mastermind 42545 points

    Hi Koichiro,

    Are all the measurements shown in the attached document taken on the same board?  

    In the document, the customer defines the behavior of the BOOTCOMPLETE.  Is this the BOOTCOMPLETE signal from the C6657?  What is the boot mode used?  How did the customer determine what step in the boot loader process was active when BOOTCOMPLETE was went high?

    Has the customer checked each power supply rail for transient voltage events?  We have observed cases where the DSP would stop operating correctly when the voltages were not maintained within specified levels.  These transient events are often very short and will not be observed at the time scale captured in the attached oscilloscope traces. 

    Has this behavior been duplicated on the EVM?

    Regards, Bill 

  • 0 in reply to Bill Taboada
    TI__Mastermind 22945 points

    Hi Bill,

    Bill Taboada said:

    Are all the measurements shown in the attached document taken on the same board?  



    Yes, these are taken on the same board.

    Bill Taboada said:

    In the document, the customer defines the behavior of the BOOTCOMPLETE.  Is this the BOOTCOMPLETE signal from the C6657?  What is the boot mode used?  How did the customer determine what step in the boot loader process was active when BOOTCOMPLETE was went high?



    Yes, it is BOOTCOMPLETE signal from C6657.
    Customer sets BOOTCOMPLETE bit1 (BC1) from Core0 in user code just after RBL.

    Bill Taboada said:

    Has the customer checked each power supply rail for transient voltage events?  We have observed cases where the DSP would stop operating correctly when the voltages were not maintained within specified levels.  These transient events are often very short and will not be observed at the time scale captured in the attached oscilloscope traces. 

     

    Customer checked voltages with 2MS/s sampling with scope, but they could not see any suspicious voltage drops.

    Bill Taboada said:

    Has this behavior been duplicated on the EVM?



    No. It is not reproduced on EVM.

    Thanks and regards,
    Koichiro Tashiro

  • 0 in reply to Koichiro Tashiro
    TI__Mastermind 42545 points

    Hi Koichiro,

    The BOOTCOMPLETE signal reflects the end of the ROM bootloader operation.  What bootmode is the customer using?

    Can the customer determine what is occurring around the time that the DSP stops operating?  Can you provide more details on the behavior of the DSP which the customer defines as not operating?   Can the customer connect to the device using JTAG in that state?

    The customer should check for any transient voltage incidents that violate the specification with the scope set to capture very fast transients, in the order of 20-200nsecs.  This will not be seen at the time scale set in the captures provided.  Most often these are observed around the time that the PLLs are programmed.

    Regards, Bill