• State Resolved
  • Locked Locked
  • Replies 12 replies
  • Answers 3 answers
  • Subscribers 79 subscribers
  • Views 761 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.

Compiler/MSP430F2272: MSP430F2272 DCO Issue

Teach Me
Expert 1940 points
Part Number: MSP430F2272


Tool/software: TI C/C++ Compiler

Hi TI Experts,

For the MSP430F2272, inside the flash memory has got the DCO (Digital Controlled Oscillator) constant data at location from 0X10F8 to 0X10FF (Information Memory Segments)

We would like to confirm is it the DCO Constant Data is pre-programmed at Texas Instrument?

Can we manually change it?

Thanks,

Br,

Teach Me

  • 0
    TI__Genius 12315 points

    Hello,

    Yes, these values are pre-programmed at Texas Instrument. These are calibration values for the DCO at room temperature and DVCC = 3 V to get more accurate DCO frequency. You can re-calibrate DCO by yourself and change them. Or use the flash memory for your own data storage.

  • 0 in reply to Winter Yu
    Expert 1940 points

    Hi Winter Yu,

    Can you please further explain what you mean by “can re-calibrate DCO by yourself and change them. Or use the flash memory for your own data storage”.

     

    We have read the programming file on few components (raw ICs) and found that the hex values for DCO constant are different, can you please advise why the DCO values different from each other?

     

    IC #1

    cid:image005.png@01D54C5D.A5917320

    IC #2

    cid:image006.png@01D54C5D.A5917320

    IC #3

    cid:image007.png@01D54C5D.FE6D40A0

     

    Br,

    Teach Me

  • 0 in reply to Teach Me
    TI__Genius 12315 points

    Hello,

    The DCO is an integrated digitally controlled oscillator. The DCO frequency can be adjusted by software using the DCOx, MODx, and RSELx bits.

    Because the DCO frequency of each devices with same DCOx, MODx, and RSELx setting is different and has error. If you want to get more accuracy frequency, you need calibrate the DCO or use TI pre-programmed DCO calibration value.  

    Each MSP430F2xx device (and most MSP430G2xx devices; see device-specific data sheets) has calibrated DCOCTL and BCSCTL1 register settings for specific frequencies stored in information memory segment A. To use the calibrated settings, the information is copied into the DCOCTL and BCSCTL1 registers. The calibrated settings affect the DCOx, MODx, and RSELx bits, and clear all other bits, except XT2OFF which remains set.

    Please use below code example and setting to calibrate the DCO.

    http://dev.ti.com/tirex/explore/node?node=ADduqBLuT1.NNcnnNwFtig__IOGqZri__LATEST

  • 0 in reply to Winter Yu
    Expert 1940 points

    Hi Winter Yu,

    Understand that the DCO constant data is pre-programmed at Texas Instrument. We had read the DCO data from the flash memory for the failed components and found that the Information Memory Segment showing the DCO constant data are blank (FF) compared with the good component (0X10F8 to 0X10FF), please see below for the screenshot and attached for the .txt file read from flash memory.

     

    Sample# 1

     

    Sample# 2

     

    According to the document FlashPro430 USB-MSP430 Flash Programmer User’s manual under DCO constant verification section, it mentions that the address from 0X10F8 to 0X10FF should contain DCO data but not 00 or FF. Else, the application cannot work due to unpredicted DCO frequency.

     

    We have also compared the Main Memory Segment with our source code and no discrepancies found, only the DCO constant data are missing (FF).

    Please advice is this a problem caused by the IC itself?

    In this case, can we send the part to TI for failure analysis?

    Thanks,

    Br,

    Teach Me

  • 0 in reply to Teach Me
    TI__Genius 12315 points

    Hello,

    Since these DCO calibration data can be modify or erase by user code or programmer tool, this problem mostly is a application issue.

    You should check your code and programming sequence firstly.

    A recommend solution is to re-calibrate the DCO by yourself.

    And see if these data will be 0xFF again without any write or erase operation by user.

    If still 0xFF, then it may be a IC failure, you can send the part to TI for FA.

    Besides, what's the failure rate and device revision? 

  • 0 in reply to Teach Me
    TI__Genius 12315 points

    Hello,

    I checked the MSP430F2272 Device Erratasheet (Rev. M). There is a bug about your issue. The FLASH21 bug is only present in revision D.

  • 0 in reply to Winter Yu
    Expert 1940 points

    Hi Winter Yu,

    Currently we have few returned ICs, we had read the date and found the DCO data showing lost (0xFF). 

    As mentioned by you that there is a bug about the issue. The FLASH21 bug is only present in revision D.

    So can I send the IC to TI for Failure Analysis in this case?

    Thanks,

    Br,

    Teach Me

  • 0 in reply to Teach Me
    TI__Genius 12315 points

    Hello,

    What's your revision? Not revision D? If so, you can apply for FA.

  • 0 in reply to Winter Yu
    Expert 1940 points

    Hi Winter Yu,

    May I know how to verify that the MCU at my side is a revision D or not?

    Is it from the body marking?

    Please advice.

    Thanks,

    Br,

    Teach Me

  • 0 in reply to Teach Me
    TI__Genius 12315 points

    Yes, the body marking.

  • 0 in reply to Winter Yu
    Expert 1940 points

    Hi Winter,

    Is the below body markings fall under revision D?

    Please advice,

    Thanks,

    Teach Me

  • 0 in reply to Teach Me
    TI__Genius 12315 points

    Please see the MSP430F2272 Device Erratasheet (Rev. M). The revision is H and J in above marking.

**Attention** This is a public forum