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.

CCS/SM320F2808-EP: Unable to connect to the development board

Part Number: SM320F2808-EP

Tool/software: Code Composer Studio

Using an XP machine with a Spectrum Digital XDS510LC USB emulator to connect to the chip, but the connection seems very spotty

I can only get it to connect if I bring up the 1.8V core about 1/2 second slower than the main 3.3V rail... but I think that means there is something either wrong in the .gel file or something else for the part I'm using.  The datasheet doesn't make any mention about having to power up the rails at different timing.  Also, even when I power up the rails differently I can only get the board to connect about 10% of the time.

For the record the micro I am using is a SM320F2808, which there doesn't seem to be much documentation. I'm using the gel file for a f2808 (a TMS320F instead of a SM320F) but I think that should be ok. The .gel file I have is version 3.30.2

I am open to changing anything at this point. I have no idea why sometimes it will connect and other times it won't. If you have any ideas on drivers or config file versions to use I'm open to anything. Also if there is a better version of CCS that's fine too. I just need to get this working.

  • I got the programmer to connect again this morning, and I wanted to try and erase the chip since they have a program in them already.  When I tried to erase, this is the message I received.

    Flash API Error #65535:
    The device is in limp mode, operation failed.

    I'm wondering if these parts are locked and that's why I'm not able to connect to them?  Any thoughts?

  • It doesn’t appear your problem has to do anything with power sequencing. Besides, even with the “altered” sequence, your success rate is only 10%. 

    For the record, the micro I am using is a SM320F2808, which there doesn't seem to be much documentation.

    Please clarify this point. Is there some information that you are unable to find? The SM part is functionally equivalent to the TMS part. If you ignore the package/temperature rating differences, you should be able to use all the documentation for the TMS part. 

    I'm using the gel file for a f2808 (a TMS320F instead of a SM320F) but I think that should be ok.

    Yes, this is OK. 

    I am open to changing anything at this point.

    Have you considered moving to the most recent version of CCS? You can download the latest version at https://www.ti.com/tool/CCSTUDIO-C2000. You may have to move to Win7 (or later) though. I haven’t had connection issues with the recent versions of CCS 

    Could there be a problem with the XDS510LC JTAG debug probe? Have you tried using another debug probe? Also, you may want to check out our low-cost probes https://www.ti.com/tool/TMDSEMU110-U 

    Even if the parts are locked, that should not prevent you from connecting to CCS (Later devices have ECSL that would trip the JTAG connection, but this device does not). 

    In general, moving to the latest CCS version should help with connection issues, assuming there are no issues with your H/W or the XDS510LC. 

    Limp mode - Have you checked the i/p clock to the device? 

    Also, take a look at http://www.ti.com/lit/an/spracf0/spracf0.pdf 

    If it is OK with you, you could share the schematics with me privately (please send me a friendship request).

  • Do you see this behavior across multiple devices/boards? Do you have code in flash that would put the device in HALT or STANDBY?

    Please take a look at this post: https://e2e.ti.com/support/microcontrollers/c2000/f/171/t/899749

  • I have several boards, all with a firmware file already loaded into them.  They all experience the same connection characteristics (namely sporadic connections).  I hope to try and connect to a board with a fresh micro to see if that makes any difference.

    I had previously tried to connect with a newer version of CCS, but it didn't work any different.  I will try again and see if that changes anything.  I sent the debugger back to the manufacturer and they said it was fine and sent it back to me.

    I don't have access to the firmware files, so I am not sure what code may be in these devices.  If there are options I can try I will try them.

  • If you see this issue with multiple boards, it is unlikely it is a specific board/device issue (unless there is a problem with the circuit design). You mention a more recent version of CCS didn’t help. Which version did you try? 

    If possible, please try a different type of JTAG debug probe, like the XDS110.

  • So I am glad you mentioned about the clock for limp mode.  There was an enable line for the clock that I didn't think needed to be active, since I would assume it would still be able to connect with the debugger using an internal clock.  I was wrong.

    enabling the external oscillator makes it connect reliably now.

    Thank you so much for your time.