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.

TMS320LF2406A: TMS320LF2406A: Flash error with sprc074

Part Number: TMS320LF2406A
Other Parts Discussed in Thread: SPRC074

Hi.

TMS320LF2406A: Flash error in factory(https://e2e.ti.com/support/microcontrollers/c2000/f/171/t/805113) has not been resolved. 

Error occured again in sprc074(SCI)  like below. 

<Error log>

BAUD rate detect .........
OK
Initializing device ...
Waiting.
OK
Clearing
Waiting..
Wrong ACK received. program terminated.

I checked program memory with CCS3.3, 0000h - 4000h is 0xFFFF and 4001h - FF00h is 0x0000.

So, password location,0040h - 0043h is 0xFFFF.

Is this locked flash?

Once, DSP in one of this error boards is removed and new one is attached, error dose not occur.

How can I solve this?

Regards.

  • Yamamoto-san,

          0xFFFF in 0x40-0x43 indicates unlocked flash. However, you should see 0xFFFF from 0x4000-0x7FFF as well. The fact that you are seeing 0x0000 from 0x4000-0x7FFF makes me wonder of something else is wrong with the flash array. (If you check the memory map in the datasheet, you can see that flash memory is from 0x0000-0x7FFF). 

    Did you attempt to perform a CLEAR/ERASE operation with the built-in flash programmer of CCS3.3?

  • Hi, Hareesh.

    Thank you for your reply.
    Of course, flash memory is 0x0000 - 0x7FFF.
    I checked the program memory map in CCS.(containing RAM)
    I tried "clear" flash memory sector1-4 with CCS on-chip programmer.
    But I could not finish "clear".
    I attach the log below.
    Please tell me how I solve this error.
    **** Begin Sector Clear Operation ****
    OK: Flash Algorithm loaded into DSP memory
    OK: Write sector value 0x000f to symbol PRG_options
    OK: Set Breakpoint at PRG_stop
    OK: Running flash algorithm
    OK: DSP Halted
    OK: Read status value 0x0000 from symbol PRG_status
    OK: PC set to PRG_erase
    OK: Running flash algorithm
    OK: DSP Halted
    Error: Read status value 0x0001 from symbol PRG_status
    Flash algorithm failed during clear operation
    regards.
    k.yamamoto
  • The fact that another (fresh) device on the same board solves the problem indicates that this device is defective. Did you try clearing just one sector at a time?

  • Hi, Hareesh.

    Thank you for your reply.

    I think device is defective too, but baud rate matching is success with these device.

    I have 6 devices that same error occurred with.

    In all devices sector0,1 is 0xFFFF and sector 2,3 is 0x0000.

    Are all 6 devices is defective? 

    I tried clearing sector 1(1000-3FFF), but clearing is failed. 

    I attached the log below.

    **** Begin Verify Operation ***
    - Verifying Sector 1
    Error: Verify Failed at program address 0x1000, data is 0xffff, expected 0x0000
    Error: Verify Failed at program address 0x1001, data is 0xffff, expected 0x0000
    Error: Verify Failed at program address 0x1002, data is 0xffff, expected 0x0000
    Error: Verify Failed at program address 0x1003, data is 0xffff, expected 0x0000
    Error: Verify Failed at program address 0x1004, data is 0xffff, expected 0x0000
    Error: Verify Failed at program address 0x1005, data is 0xffff, expected 0x0000
    Error: Verify Failed at program address 0x1006, data is 0xffff, expected 0x0000
    Error: Verify Failed at program address 0x1007, data is 0xffff, expected 0x0000
    Error: Verify Failed at program address 0x1008, data is 0xffff, expected 0x0000
    Error: Verify Failed at program address 0x1009, data is 0xffff, expected 0x0000
    Error: Sector 1 failed verification, first 10 errors reported out of 12288 found
    **** End Verify Operation ****

    regards.

    K.Yamamoto

  • Baud rate matching is largely dependent on the proper operation of the SCI port and has nothing to do with the flash. 

    Are all 6 devices is defective? 

    As improbable as it sounds, it does appear to be the case. Especially if you are able to program other devices in the same setup (H/W & S/W). I presume these are devices that are being programmed for the first time. 

    Do you have access to CCS 3.3? If so, could you try the same operation with the built-in flash programmer of CCS?

  • Hi, Hareesh.

    Thank you for your reply.

    6 devices are 2 devices from factory and 4 devices from our laboratory.

    4 devices programming have been for factory error validation, and error occured in 4 devices is not first time.

    It's about several tens times.

    Unfortunately I don't know 2 devices.

    But I don't think that they don't try to program same devices after success in factory. 

    I think that 2 devices error is for first time or after the other nonfatal communication error.

    I will try the same operation with fresh device and CCS up to a hundred times.

    regards.

    K.Yamamoto

     

  • Yamamoto-san,

                It is important to distinguish between the following:

    (1) Whether the device failed the Clear/Erase/Program (CEP) operation for the first (and subsequent) time(s). That is the device was never put through a successful CEP operation

    Or

    (2) Whether the device was put through a successful CEP operation at least one (or more) time(s), but is now failing the CEP operation.

  • Hi, Hareesh.

    Thank you for your reply, and sorry for my replying late.

    Two devices from factory correspond to (1).

    Four devices from laboratory correspond to (2).

    Do you think they have different causes each other?

    Once this error occurs, should we change device and try again?

    How do we prevent error corresponding to (1) ?

    regards

    K.Yamamoto

  • Yamamoto-san, 

    Do you think they have different causes each other?

    I presume the end-equipment has been in production for a long time and you have started seeing this issue only recently. If that is the case, it is likely that 1 & 2 have the same root cause. 

    Once this error occurs, should we change device and try again?

    I think so. I presume serial port method is your normal way of programming and you first noticed problems there. If you face a problem with that method, you can examine the flash memory contents in CCS3.3 and check if you are able to program the device using the built-in flash programmer of CCS3.3. If there is no indication that the device is secure and yet you are unable to program, you indeed need to replace the device. It does surprise me that you have multiple devices failing. If you are willing to share more information about your programming setup offline (that is, directly with me and not through the e2e forum), I can send you a friendship request. 

    How do we prevent error corresponding to (1) ?

    Two things to ensure are (1) Do not interrupt the C/E/P process (2) Have a good power-supply that provides stable power throughout the C/E/P process.