• Locked Locked
  • Replies 2 replies
  • Subscribers 54 subscribers
  • Views 936 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.

Working EVM now will not connect to the debugger

Andrew Giusti
Intellectual 370 points
Other Parts Discussed in Thread: CCSTUDIO

I have the TMDSEVM6678 eval board and had one of the example projects (PCIe_sample) running on the debugger.  I had added some code to do DMA transfer to the EMIF-16 interface and was able to see the write cycles on a scope.

Now, my board will no longer connect to the debugger.  I'm using CCS 5.1.0.09000.  Everything appeared to be working.  I then connected an adapter board that had nothing on it but a break out connector attached to the PHY-1 connector.  I forgot to bring the ground pin out to my new connector, so I grounded my scope to pin 3 of the COM1 connector (which documentation shows as a ground pin). 

From that point on, the debugger could not connect.  I have since unplugged everything and cycled power on both the EVM and my laptop.  When I power everything up and launch my configuration file, things look normal.  Then, I try to "connect target" on core 0 and I get the following error dialog:

Error connecting to the target:
(Error -2131 @ 0x0)
Unable to access device register. Reset the device, and retry the operation. If error persists, confirm configuration, power-cycle the board, and/or try more reliable JTAG settings (e.g. lower TCLK).
(Emulation package 5.0.641.0)

Again, the board was working at one point.  I can't imagine that I broke something just by grounding my scope to pin 3 of COM1, but that's the only thing that changed.  I've tried cycling power several times, and using the RST_FULL button. 

I know the USB drivers are installed (again, the board was working) and seem to be functioning properly.  I see the two entries in device manager I expect and the com port shows up.

Are there any self diagnostic checks I can run? 

Do I need to uninstall and reinstall the USB drivers?

Do I need to flash the board back to its factory defaults?

Any help would be greatly appreciated.

  • Intellectual 370 points

    I tried running through the reprogramming process described in Program_EVM folder on the TMDSEVM6678 installation cd.  Running the DSS script batch file using the program_evm.js script, I get the same error message -

    c:\ti\mcsdk_2_00_06_18\tools\program_evm>%DSS_SCRIPT_DIR%\dss.bat program_evm.js
     TMDSEVM6678L-le > err.txt
    SEVERE: IcePick_D: Error connecting to the target: (Error -2131 @ 0x0) Unable to
     access device register. Reset the device, and retry the operation. If error per
    sists, confirm configuration, power-cycle the board, and/or try more reliable JT
    AG settings (e.g. lower TCLK). (Emulation package 5.0.641.0)

    SEVERE: emulation failure occurred
    SEVERE: Error connecting to the target: emulation failure occurred
    org.mozilla.javascript.WrappedException: Wrapped com.ti.ccstudio.scripting.envir
    onment.ScriptingException: Error connecting to the target: emulation failure occ
    urred (program_evm.js#274)

  • TI__Guru**** 171329 points

    Hi,

    Sorry to hear about your system. Per your description my biggest suspicion is similar to yours: the hardware was somehow damaged.

    Andrew Giusti said:

    Again, the board was working at one point.  I can't imagine that I broke something just by grounding my scope to pin 3 of COM1, but that's the only thing that changed.

    Considering your ground pin did not touch any other pin by accident, I say it is perfectly possible... If the power cable of both the power supply of your board and the oscilloscope have three pins (live, neutral and ground), ground voltage differences between boards and measurement instruments like your oscilloscope are very common and can be fatal.

    I for one, had the motherboard of my PC destroyed because of this. When I connected a grounded power supply to a development board, then the board to the parallel port of my PC, the voltage difference in the GND caused a major spark. In this case, the board was stronger and destroyed the southbridge of my PC's motherboard. 

    Additional theoretical and practical details can be found at an interesting video blog here.

    Andrew Giusti said:

    Are there any self diagnostic checks I can run? 

    The screenshot and behaviour indicate the emulator circuit of your board is still intact, therefore the voltage difference may have affected only the processor (and any other subsystems). In this case, from CCS you can open the target configuration editor and run the "Test Connection" button. It will give additional details about the JTAG connection itself (but has very limited visibility on problems specific to the device).

    To check the device itself, I would first check if there is a test point for the PLL output of the device (this could help checking if the device is minimally "alive"), as well as test points for all the power rails of the board (maybe a power regulator is damaged instead of the main device). For additional details on how to check the device itself, I suggest asking specific details on the device forum.

    Hope this helps,

    Rafael