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.

MSPM0G3507: Regarding SYSPLL_ERR_01

Expert 2975 points
Part Number: MSPM0G3507
Other Parts Discussed in Thread: SYSCONFIG

Hi All,

Confirmation regarding the SysPLL errata:

Does this issue occur on all G-series devices, but the probability of occurrence is random?
I have a question about the following statement in the SysPLL errata:

>>The count will vary depending on the combined clock accuracies, so it is recommended to add a >>+-5% to allowed range.Estimated time for FCC is 30us.

Does this mean that SysPLL originally has a ±5% error?
Or is it because of this issue that the error has become ±5%?
For example, does the ±1% error of SYSOSC described in the datasheet need to be considered as ±5% after passing through SysPLL? Was this already determined before the errata was issued?

If the FCC value is incorrect, disable and reenable the SYSPLL by setting SYSPLLEN to 0 then 1. Rerun the FCC check.

Regarding this, should I loop the process of setting SYSPLLEN from 0 to 1 and rerunning the FCC check until SysPLL is locked?

Best Regards,

Ito

  • Hi Ito,

    Does this issue occur on all G-series devices, but the probability of occurrence is random?

    Yes. The probability is small and do see difference in different device. (some device never failed in the my bench test).

    Does this mean that SysPLL originally has a ±5% error?

    No, that is due to LFCLK has 3% and SYSOSC has 1% accuracy,

    We are planning to update the WA to give a more accurate method for users to count the SYSPLL.

    1. Use FCC to count SYSPLL REF clock

    2. Use FCC to count SYSPLL OUT clock

    Then do comparison of these two counting with expectation, this remove the clock accuracy impact between SYSPLL REF and LFCLK.

    I plan to release a FAQ with an example and then update to this thread.

    B.R.

    Sal

  • Unfortunately I do not find the way to generate FAQ in the new system.

    See below example for your reference. I put the necessary instructions in the code comment.

    LP_MSPM0G3519_PLL_FCC_TEST

    B.R.

    Sal

  • Hi Sal,

    Thank you for your reply.

    In any case, since it is unrealistic to sort out all defective products before shipment,
    it is essential to apply WA to every unit when using PLL.

    This errata has significant implications and could act as a barrier for customers considering the G series, so we expect it to be corrected.

    Best Regards,

    Ito

  • Hi Ito,

    Thanks for your feedback. We definitely want to avoid inconvenience for user developing or firmware maintaining. Errata will be fixed in next release (add a new approach in WA to improve the check accuracy).

    In the meantime, we are planning to add some reference code into SDK/SysConfig tools to help our user quickly apply the WA into their code, without additional code modifcation. We are targeting to release in the very next version. Hopefully we can release it this month.

    B.R.

    Sal