• State TI Thinks Resolved
  • Locked Locked
  • Replies 15 replies
  • Answers 1 answer
  • Subscribers 82 subscribers
  • Views 1353 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.

DSP emulation error during converter load change test.

Siyu He
Expert 1040 points

Other Parts Discussed in Thread: UNIFLASH

I am using a F2812 controlling a power converter. It happens to me that the CCS3.3 ran in an error while I was doing load changes test. The pop up window shows the information as below,

Trouble Halting Target CPU:
Error 0x00000020/-1156
Error during: Execution,
Lost processor clock. Device may be operating
in a low-power mode. Do you want to bring it
out of this mode?


Sequence ID: 15
Error Code: -1156

The command window shows:

Error: Error 0x20000024/-1135 Severe Error during: Register, Execution, Unrecoverable emulation error Sequence ID: 16 Error Code: -1135 Error Class: 0x20000024
Failed to remove the debug state from the target before disconnecting. There may still be breakpoint opcodes embedded in program memory. It is recommended that you reset the emulator before you connect and reload your program before you continue debugging.
Error: Error 0x20001020/-1135 Severe Error during: Execution, Target, Unrecoverable emulation error Sequence ID: 16 Error Code: -1135 Error Class: 0x20001020
Failed to remove the debug state from the target before disconnecting. There may still be breakpoint opcodes embedded in program memory. It is recommended that you reset the emulator before you connect and reload your program before you continue debugging.

I noticed that sometimes the CCS stopped at the ADC initial program while the error occurred. I guess would this because of the large current change making the ADC go wrong? 

  • 0
    Guru 192875 points

    Hi Siyu,

    This is happening everythime you debug or sometimes? Also, I would like you try loading the .out file after building the project using CCS uniflash once.

    Regards,

    Gautam

  • 0 in reply to Gautam Iyer
    Expert 1040 points

    Hi Gautam,

    The error happens once in every three or four load change tests,. The uniflash helps with this? I read the wiki, it seems like the software has a lot functions. 

    He, Siyu

  • 0 in reply to Gautam Iyer
    Expert 1040 points

    Hi Gautam,

    The uniflash shows "Error initializing emulator" in the new target configuration process. I am using a XDS510LC and ezdsp2812 which are included in the liabrary. I searched the FAQ and support of uniflash, there is no information about this error.

    He, Siyu

  • 0 in reply to Siyu He
    Guru 192875 points

    HI Siyu,

    The above 2 errors indicates that there seems to be some connectivity issue.

    The error happens once in every three or four load change tests

    Did you try with different USB ports? Also, is this emulator working fine with other boards without any issues multiple times?

    Regards,

    Gautam

  • 0 in reply to Gautam Iyer
    Expert 1040 points

    Hi Gautam,

    Yes, I tried it on different laptops and CCS shows the same error, but I only have one board. The emulation works fine with other tests except the load change test.

    He, Siyu

  • 0 in reply to Siyu He
    Guru 192875 points

    the load change test.

    Can you explain what's happening here? or what do you doin this test?

    What I feel is by changing the load esp. while increasing the load, the noise on board is effecting the communication between emulator and the board!

    Regards,

    Gautam

  • 0 in reply to Gautam Iyer
    TI__Guru 57255 points

    Siyu,

    This likely won't solve your issue completely, but can you try increasing the strength of R13 (make it about 2.2Kohm) on the ezDSP?  This will help make the JTAG's TRSTn line a bit more stable.

    I might also recommend checking your isolation strategy.  Is there isolation between the eZDSP and the power stage?  If not, I would recommend using an isolation adapter
    (something like the 14-pin adapter from blackhawk-dsp: http://www.blackhawk-dsp.com/products/Isolation.aspx). 
    This will protect help protect your computer from potential damage and give you a more stable JTAG connection.


    Thank you,
    Brett

  • 0 in reply to Gautam Iyer
    Expert 1040 points

    Hi Gautam,

    The load test was performed by step increasing and decreasing the load, the inductor current doubles in this test. I guess it might the noise or the current spikes.

    He, Siyu

  • 0 in reply to Siyu He
    Guru 192875 points

    The load test was performed by step increasing and decreasing the load, the inductor current doubles in this test. I guess it might the noise or the current spikes.

    Yup, that's what might be causing frequent disconnection. Do try what Brett has suggested and keep us informed.

    Regards,

    Gautam

  • 0 in reply to Brett Larimore
    Expert 1040 points

    Hi Brett,

    Thanks for your comment. I will try to replace the R13 by 2.2K resistor.

    The  ADC connecting to the current sense resistor is directly connected to the power stage. I will look at the isolation adapter. If I add a op amp formed voltage follower between the ADC and the current sense resistor, would this help the problem?

    He, Siyu

  • 0 in reply to Siyu He
    TI__Guru 57255 points

    Hi Siyu,

    The op-amp voltage follower might improve the accuracy of your current measurement, but I don't think it will resolve the root cause of CCS disconnecting.

    I still suspect that isolation will reduce your emulation connectivity issues.


    Thank you,
    Brett

  • 0 in reply to Brett Larimore
    Expert 1040 points

    Hi Bret,

    May I know  why the adapter would solve the problem? As you mentioned in the previous post, the power stage is connected to the eZdsp directly, why would a isolation adopter between the emulator and eZdsp would help with the problem? If download the program to the flash, would this happen again?

    Sorry for so many questions, since I have to convince my boss to spend the money.

    He, Siyu 

  • 0 in reply to Siyu He
    TI__Guru 57255 points

    Hello,

    How much power does your converter output? 

    If the power is low <50W then maybe this isn't an issue.  However, I am guessing that your converter is more powerful than this.

    The issue is that, without isolation, there will be a ground path (and a 5V/3.3V power path) between your converter PCB and your computer.   This path allows FET switching noise, etc to have a path through the USB cable to your computer (and whatever capacitance it has between 5V and GND) and potentially to the AC outlet. 

    Isolation mitigates these issues by keeping ground bounces and spikes on the converter PCB board.  This means that the signals in the USB cable will be cleaner (which is likely your root cause) and generally make your computer less vulnerable to damage.

    Another option is to completely remove the USB cable run from FLASH and then run your code with the minimal debug of a few LEDs.  This is probably better suited for final testing of your board;  not necessarily software development, experimentation, etc.

    The final option is to use a different isolated form of communication to do your debug on (like SPI/UART/etc).  At the end of the day, you should still isolate this interface somehow and in addition you'll also have to learn a new peripheral.  This of course assumes that you don't have to make this interface for some other reason.


    Thank you,
    Brett

  • 0 in reply to Brett Larimore
    Expert 1040 points

    Hi Brett,

    My converter is about 10W. We are trying to purchase a isolation adapter. I will let you know what is going on with the adapter.

    He, Siyu

  • 0 in reply to Siyu He
    TI__Guru 57255 points

    Hi,

    I am surprised that something like this is happening with a 10W converter.  However, it is possible.

    It may be worthwhile to also check your power rails and make sure that the rails are not  spiking out of the power rail specifications for your device (and perhaps resetting your device).

    Finally, you may also want to take a look at the XRSn line and see if some kind of software reset (or SVS reset) is occurring while your load change process is going on.  If a XRSn reset is occurring, you should then try to find why it is occurring.


    Thank you,
    Brett