• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 82 subscribers
  • Views 623 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.

TMS320F28034: internal clock issue

Howard Zou
Genius 12365 points
Part Number: TMS320F28034

Hi,

my customer is using F28034, and the SYSCLK is set as 60MHz, XCLKOUT output is 15MHz to be measured.

when they do EFT test, they will observe XCLKOUT to be 7.2MHz and the PWM frequency change, which means the SYSCLK is changed.

When they remove the EFT pulses, the clock is still wrong.

They are using internal oscillator as the clock source, so they suspect that the internal oscillator is too easy to be influenced and caused the failure of EFT test.

I will check with the customer to see if the power(3.3V, 1.8V) is within our recommendation range during EFT test, but I don't consider power is the reason, because the clock is still wrong when they remove the EFT pulses.

Do you have any idea what else might be the reason for the wrong clock frequency?

Howard

  • 0
    TI__Guru* 95315 points
    Howard,
        I wonder if the oscillator trim registers are getting corrupted by the EFT test. Could you please check the COARSETRIM and FINETRIM register values before and after the test? Also, is the incorrect value of the clock the same every time you do the test? It is inconceivable that the register gets corrupted in precisely the same way every time.
  • 0 in reply to Hareesh Janakiraman
    TI__Genius 12365 points
    Thanks, I will check with the customer next week. According to them, the incorrect clock frequency is always 7.2MHz. But if it's caused by COARSETRIM and FINETRIM register values , the customer may blame that it's our parts fault that caused the failure of the EFT test.
  • 0 in reply to Howard Zou
    TI__Guru* 95315 points

    Howard,

                If the frequency is exactly 7.2 MHz everytime, it is unlikely that the trim registers are getting corrupted. In any case, I presume customer can check the contents of the trim registers to verify this.

     

    Is the change in clock the only effect of the EFT test? Does it impact code execution in any way?

     

    Would you be able to share with me offline the details of how the EFT tests are done?