• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 83 subscribers
  • Views 986 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.

TMS320F280049M: INTOSC Accuracy

Ricky Zhang
Genius 9520 points
Part Number: TMS320F280049M


Hi, Champs,

As I stated in below post: https://e2e.ti.com/support/microcontrollers/c2000/f/171/t/648540

My customer is now using F280049M at room temperature without trim, however, out of 3 parts there're 2pcs running the clock outside the spec of +/-3% accuracy.

It's confirmed that the Vdd is 1.177v, within range of 1.14~1.32v, but the clock is about 9.2MHz by measuring XCLKOUT on these 2 devices, no matter it's INTOSC1 or INTOSC2.

Please check attached waveforms and advise what needs to be checked, or if those samples are not validated for the accuracy item?

Best Regards,

Ricky Zhang

  • 0
    TI__Guru 60865 points
    Hi Ricky,

    This is likely do to these units being untrimmed TMX pre-production samples. The oscillator does require trim (but the user can't modify the trim on-the-fly to compensate for temperature shift). Are these individual units on a customer's board, or devices on a TI development kit? What do the device markings look like? What are the PLL settings and XCLKOUT divider settings?

    I believe that production F28004x devices should be available pretty soon (or production versions of F2807x, F2837xD, or F2837xS use the same oscillator design and factory test procedure, so oscillator performance on these devices should be a pretty good indication of expected performance on F28004x production devices).
  • 0 in reply to Devin Cottier
    TI__Genius 9520 points

    Devin,

    Thanks for quick response.

    Could you please confirm if those samples are untrimmed in factory based on below information?

    Devin Cottier said:

    Are these individual units on a customer's board, or devices on a TI development kit?

    On customer board rather than TI EVM.

    Devin Cottier said:

     What do the device markings look like? 

    See below picture:

                    980

       X
       F280049MPZS
       YFA - 69C00VW
               G4      

    Devin Cottier said:

     What are the PLL settings and XCLKOUT divider settings? 

    PLL is bypassed and INTOSC is selected to XCLKOUT directly:

    ClkCfgRegs.CLKSRCCTL3.bit.XCLKOUTSEL = 5;

    ClkCfgRegs.XCLKOUTDIVSEL.bit.XCLKOUTDIV = 0;

    Devin Cottier said:


    I believe that production F28004x devices should be available pretty soon (or production versions of F2807x, F2837xD, or F2837xS use the same oscillator design and factory test procedure, so oscillator performance on these devices should be a pretty good indication of expected performance on F28004x production devices).

    Customer is using 100pin package device so the fastest time can they get the TMS parts will be Jul. 2018 when it's RTM'ed.

    Do we have plan to get it's trimmed in coming 2 batches samples? As far as we know, we're to provide 100pin samples in Jan. and Mar. prior to its RTM, so customer wants to know if they can get it verified before mass production.

    Best Regards,

    Ricky Zhang

  • 0 in reply to Ricky Zhang
    TI__Mastermind 39620 points
    Hi Ricky,

    Thank you for sending the details of the device marking. We confirm that this unit is from a pre production batch that did not get the optimized trim values for the internal oscillator. The new customer orders arriving in Jan. and Mar. timeframe should have the production trim values that represent the expected oscillator performance. As a confirmation, the 4th line on the device marking for the parts that have the production trim values should start with YFB.

    Please let me know if you have other questions regarding this matter.

    Best regards,
    Joseph
  • 0 in reply to Joseph Casuga
    TI__Genius 9520 points

    Joseph,

    Thank you so much for clarification.

    I think customer did get Rev. B samples but unfortunately the quantity is not enough then we borrowed some from other customers which was Rev. A allocated earlier. Now we see we trim all silicon since Rev. B and customer will only get Rev. B samples from now on, so this will not be a problem any more.

    We can get this post closed and thank you and Devin for your support.

    Best Regards,

    Ricky Zhang

  • 0 in reply to Ricky Zhang
    TI__Mastermind 39620 points

    Hi Ricky,

    Thanks for that information.  For this case it so happened that all the Rev. B parts had the production trim flow for INTOSC so the die revision would be a good indicator of that.

    Best regards,

    Joseph

  • 0 in reply to Joseph Casuga
    TI__Genius 9520 points

    Joseph,

    Okay, thanks for your full support.

    Best Regards,

    Ricky Zhang