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.

CC2340R5: CC2340R5RGE cannot be started.

Part Number: CC2340R5
Other Parts Discussed in Thread: , SYSCONFIG

Hi,

I have some problems when debugging my board, please help me. I am using the latest SDK(simplelink_lowpower_f3_sdk_7_40_00_64).

I use the project of basic_ble.The project uses CC2340R5RKP by default. Follow the link to change to CC2340R5RGE.

After compiling successfully, flash, but the program can not start (according to the debug serial port no data printing and the Bluetooth broadcasts cannot be scanned).

After the operation described in this post, it can run normally, but normal burning is unable to start. What could be the problem? 48MHz crystal oscillator problem or other?

  • Hi bing,

    Are you using the standard basic_ble project, or an OAD version?  If non-OAD then you should not need tips to debug OAD software, although some parts are useful for debugger insights.  I also recommend that you review the BLE5-Stack Debugging Guide.  If you have sufficient experience evaluating this example with a LP-EM-CC2340R5, and have followed all Migration Steps, then you should further investigate your custom hardware.  Consider submitting your design to SIMPLELINK-2-4GHZ-DESIGN-REVIEWS for further review.  If you have any GPIOs connected to LEDs then it may be worthwhile to attempt turning it on after device initialization.  What happens when you load and start the code inside of the CCS debugger (if SWD pins are connected)?  Please provide the call stack and device state when the debugger is paused.  

    Regards,
    Ryan

  • Hi Ryan,

    Our own project is OAD. Since direct writing can not start, so try this debugging method, found that it can start normally, it is suspected that Boot or Persistent problems. So I want to find a direct start project (basic_ble) to verify, found that after burning can not start, direct debugging did not have any reaction, the following is the basic_ble project direct debugging state. However, the debugging method of OAD can be started normally.

    In addition, we suspect that the 48MHz crystal oscillator is the problem, we also put the previous version (using CC2340R5RKP) can normally start the crystal oscillator replaced on this board, still can not start.

    There is no LED on our board, but I tried to initialize the high and low output of gpio inside after starting up, and there was no change in the multimeter test.

    Above is the state of basic_ble project debugging when paused.

  • Here are some of our schematics:

    In addition, while debugging the basic_ble project directly, the code has been running the following code in a loop.

  • Please make sure you are properly following all instructions from the Bluetooth Low Energy - Over the Air Download (OAD) Fundamentals SimpleLink Academy Lab for building and loading the MCUboot on-chip hex file, persistent binary image, and application binary image.  There are also further instructions given for running the demo.  Debugging the MCUboot application would be more helpful given that this is where device operation should begin after reset.

    The board layout will be more insightful than the schematic at this point, which is why a full hardware review request should be submitted.

    Regards,
    Ryan

  • Hi, Rayan

    The first version of our PCB has been successfully debugged and can operate normally with OAD. Now the chip is changed from CC2340R5RKP to CC2340R5RGE, and such a problem arises. The current problem is not related to OAD, I directly use the basic_ble project, modify it to CC2340R5RGE and then the debugging has been running in an endless loop, but this code can be run directly to the demo board (CC2340R5RKP), so the code should be fine. I suspect it might be a startup address issue, I wonder if you can give me some advice?

    Regards,

    billy

  • Are the startup addresses of CC2340R5RKP and CC2340R5RGE different? Or anything else that needs attention?

  • The startup address are not different between these variants.  Can you please confirm whether basic_ble (non-OAD) and empty examples run on your CC2340R5RGE after making the necessary modifications?  Be sure to erase all device memory before re-programming.  Also, provide call stacks while debugging if possible.

    Regards,
    Ryan

  • I tried the basic_ble and empty projects with the same results. Both of these(after making the necessary modifications) worked fine on the first version, but neither will work on the new board. 

    After I removed the 48MHz crystal oscillator from the first version, the phenomenon was the same as that of the new board. However, the first version of the 48MHz crystal oscillator was welded to the new board, and the new board still did not operate normally.

    PCB layout

    top view:

    bottom view:

     Antenna section:

    Regards,

    Billy

  • This certainly appears to be related to the custom hardware design.  I will loop in a HW colleague to advise as such.

    Regards,
    Ryan

  • Hi Ryan,

    Did your HW colleague reply?

    I found the following problem while debugging. Take a look at this code(ldr r0, [r0, #0x60]). The value of R0 is 0x40030100. Check 0x40030160 Memory value is 0xf. 0x40030160 should be the STA register inside the SPI register. 

    Why check the STA register in SPI when starting?

    Regards,

    Billy

  • I will ping HW to respond.  You findings depend on the image which has been loaded onto the device.  For instance, SPI would be referenced if debugging an off-chip MCUboot solution.  Loading project symbols and reviewing the call stack when the debugger is paused will give you further insight.

    Regards,
    Ryan

  • I'm using the basic_ble project. Does CC2340R5 have boot code built-in? Running from boot code?

    The above debugging I call step by step.

    Regards,

    Billy

  • "basic_ble" implies no OAD.  SPI is used to shut down external flash during start-up for LaunchPad boards.  If you do not have an external SPI connected then you should remove the "Generate Initialization Functions" checkbox from basic_ble.syscfg -> TI DRIVERS -> Board inside the SysConfig editor.  I would have expected this to already be accomplished or else your CC2340R5RKP board version would have experienced similar issues.  Otherwise, SPI is not used by the basic_ble project.

    Regards,
    Ryan

  • I found this when I did the above debugging, and removed it, but the problem is still the same. So I suspect an internal boot code problem?

    Regards,

    Billy

  • Have you programmed the CCFG to enable the bootloader, and are you enabling the backdoor bootload pin during device reset?

    Regards,
    Ryan

  • Hi Ryan,

    I use the default setting for the Bootloader Configuration. The CCFG Settings I used were the project default and did not change. 

    How to enable  the backdoor bootload pin?

    Regards,

    Billy

  • Hi Ryan,

    After I changed the setting value of Bootloader Configration to "Any bootloader forbidden", the program can run normally. 

    Is the default bootloader on the System Rom? Why does the same program with the same CCFG configuration work on the CC2340R5RKP but not on the CC2340R5RGE?

    Thank you very much.

    Regards,

    Billy

  • Hi Billy,

    I had momentarily forgotten that the default bootloader configuration is pre-determined on the CC2340R5 by the FCFG.  It was good of you to try disabling the bootloader entirely.  From the TRM:

     Thus if DIO 21 is held low during a reset.  You can avoid this by configuring a new bootloader pin setup in CCFG or disabling the bootloader altogether, as you've shown is effective.

    I did have some miscommunication with my hardware resources and do again recommend that you submit a request tSIMPLELINK-2-4GHZ-DESIGN-REVIEWS.  I advise that you edit your previous response so that your e-mail address is not visible on a public forum.

    Regards,
    Ryan

  • Thanks, Ryan. So I'm going to disable the bootloader. If I have any more problems, I will open a new post and ask questions. Thank you again for your help.

    Regards,

    Billy