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.

TM4C123BH6PGE: Device locked

Part Number: TM4C123BH6PGE
Other Parts Discussed in Thread: UNIFLASH,

Hi,

 one of my customers is complaining about the above device, because they detected several devices, that after programming and boot loader installed, stop working.

They do not know how to verify the failing condition because they cannot access anything inside and JTAG is not working.

The devices seem to loose the program , but I am not sure they are completely failed, because for example LDO is working.

Is there a way to reset the device or to debug it?

Thanks

Umberto

  • Hi Umberto,
    Prior to the bootloader installation everything was working? If this is the case what was in the bootloader? What communication interface does the the bootloader use? What debug probe do they have on the board? They can refer to the section 5.3 in this app note www.ti.com/.../spma075.pdf to unlock the device. If they have XDS100 or XDS200 they can use dbgjtag.exe command to unlock the device. If this is a LaunchPad with the onboard ICDI debug probe then they can use the LM flash Programmer to unlock. 

    Can they also use either the LM flash programmer or Uniflash to erase the flash?

  • Hi Charles,

    pls see customer answers:

    Prior to the bootloader installation everything was working?

    Yes

    If this is the case what was in the bootloader?

    Texas Instruments IAP standard using CAN interface

    What communication interface does the the bootloader use?

    CAN interface

    What debug probe do they have on the board?

    XDS100v2 (for Debugging), TMC1294XL LaunchPad (for Programming)

     

    I tried to use LM Flash Programmer to unlock the processor but the procedure  did not given any result

     

    Part number: TM4C123BH6PZI

    Thanks

    Umberto

  • If I may - part "TM4C123BH6PGE" (listed poster's opening post) - to my knowledge - does not appear upon any common LPad.     Thus - is it true that client's "locked device" resides upon a board of his "own design & creation?"

    If that proves true - upon how many of such boards - does this problem reveal?     (dreaded Single Board Anomaly!)     Newly designed/assembled boards have "many areas & components suspect" - thus the odds against "total success - from the start - appear high!"

    It is noted that (even) JTAG suffers from lockout.       Are external pull-up resistors present upon the (assumed) custom boards?       Firm/I have long noted the "desire" by some to "save" the size/cost of "4, 06-03 smt pull-up Rs" (by using MCU's internal Rs) - which while working "sometimes" -never proves as robust as the "far lower value" external Rs.   (good 10x lower in value - vastly reducing signal reflections & "rounded signal edges" - both yielded by the "saving!")     (some saving!)

    I would suggest the addition of external pull-ups - so that JTAG has far higher chance of establishing connection...     And again - the number of and percent of "failed boards" IS important - yet silent...

  • Hi,
    the customer produced 3000 boards and 14 are not working; yes , the device is mounted on their own design.
    Actually there is only 10K pullup on TMS pin and I told to put also on the others
    Customer feeling is that for some reason the internal memory is lost.
    Thanks
    Umberto
  • Thank you, my friend - and what you've provided - with the addition of "one more" question - can prove extremely helpful.

    So 3000 (custom boards - Not LPads) were produced - 14 are reported as "Not working" - does this mean that the "2986 remaining" have "Passed Testing" - and "DO WORK?" That is the issue - is it not?

    The resulting "Method of Attack" must address the "Percent of Failure" and that clarity is (again) sought.       To be completely clear - do > 2500 of such boards Work?   (i.e. Pass Testing?)

    (and GREAT that you directed use of the external pull-up (10K fine) and avoided the "uncertainty" which any "misguided savings" (use of high valued MCU internal Rs) insures...