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.

XDS110 debugger with PGA970

Other Parts Discussed in Thread: PGA970, PGA970EVM

Dear Scott,

   I managed to resolve the heating issue by correcting the error in the PCB. However, I am not able to program the device with its firmware through my XDS110 Debug probe as it is failing in "Test connection". I have already share my schematic that shows SWDCLK, SWDIO and Vcc(ref 5V through 10K resistor) and Vss connections. Could you pl share the connection diagram showing  the  XDS110 pins that need to be connected to the PCB for a successful connection. Kindly take this as urgent and reply.

  • Hi Anuj,

    The only connections that need to be made are the SWDIO and SWCLCK pins on the XDS110. You can see these pinouts in the user's guide for the XDS110:

    Also make sure that you use the correct target configuration settings as shown below. You will also need to make sure that the power selection is set to "Target supplied power"

    Regards,

  • Thanks Scott, just wanted to check if PGA 970 "Ground" needs to also be connected to the XDS110 connector ground pin?   

     

  • Hello,


    Yes, the ground pin should also be connected.

    Regards,

  • Dear Scott,

        As per your advise, I have tried by reinstalling the CCS and followed the 'Procedure to Build PGA970 Code' as per the SLDU025–November 2016 document but am noticing that the PGA970 starts getting  heated, the moment I connect the XDS-110 debugger and start 'Test Connections'. I have attached the screen shot of  my settings but do not understand what is causing the PGA970 to get shorted(even after correcting the AVdd and DVdd errors). As you can see in my schematic earlier that I have connected the gate pin to ground permanently with supply  of 5V DC to the SoC. Also, I had kept the voltage level to 'Default' in 'Target Configuration' settings. Would any of these settings make any difference or cause PGA970 to internally get damaged as the resistance is now showing 15ohms across 5V supply terminals after disconnecting it(and SoC heating) indicating that the there is an internal  short that has taken place. I am  having a very tough time using this SoC and have damaged 5 chips this way. Kindly reply urgently and help. Also attached is the error msg I am getting now.

  • Dear Scott,
    Just wanted to add that I realised that there also has been an error by me where I have also not disconnected the DVdd_Mem from the 5V supply. Could this be also the cause of failure/heating? Is DVdd_Mem pin not tolerable to 5V D.C.? Kindly let me know. Thanks
  • DVDD_MEM should be grounded through a capacitor. I recommend that you review the Typical Application section of the PGA970 datasheet, as well as the PGA970EVM schematic which is a good reference for how to wire up the device.

    Regards,
  • Dear Scott,

       As advised, I have corrected the errors made earlier but still am not able to have a successful connection in "Test Connection". I am sharing a photo of the board developed too this time showing the SWDCLK and SWDIO  and ground connected to the XDS-110 where pin no.1(of 14 pin connector of XDS-110 Debugger) is SWDIO, pin no.11 is SWDCLK and pin no.8 is connected to Ground as per your advise. Pls find settings and error messages received in screenshot below   and advise. I am using CCS 8.3.0.00009. I have also successfully updated the  bootloader and firmware of the debugger but in vain.

  • Hi Anuj,

    Can you please repost your photos? They are not showing up.

    Regards
  • Dear Scott,

      Pl find attached photos. Today I even tried connecting the VTRef pin to Pin 5 of the XDS-110( as shown in the schematic in PGA970EVM User's Guide-sldu017) but still getting the same errors as shown in screenshots below. Kindly let me know if I can directly reach you on your mail instead of this forum as it would speed up communications as we are already delayed.

  • Pl find attached photos.....

  • Dear Scott,

       Kindly ignore earlier recent  mails as today we have been successful connecting the debugger to the SoC with 'Test Connection' success. Also the firmware provided has also been downloaded to the target SoC. The issue was that the Target  voltage needs to be set to 'Default' .   Now we will be testing the rest of the device for functionality and accuracy. Thanks for your help.

    It would however be nice if you could have a document released that provides clear instructions on how to connect the PGA970 through XDS110 Debugger using  the latest Code Composer Studio version 8 or higher as the current document describes XDS200 along with CCS version 6 whose menus  are different in some places.

  • Dear Scott,

        I managed to run the code and get the PGA970 SoC started with wavefoems. But I am stuck with a common problem faced by many users of 'Controller held in Reset'  after executing the instruction "MICRO_INTERFACE_CONTROL = 0x03" as mentioned in the PGA datasheet(highlighted in datasheet below, which is misleading). But even after commenting the same, I am not able to connect to the controller since it stays in "RESET" immediately after power on. I have tried methods suggested by you on this forums by am not able to connect. Is there any way to erase the code or PGA970 chip without connecting to it. As I can see that the 'Test Connection' is still successful and operational. Kindly adivse and reply as soon as possible.

  • Dear Scott,

       Could you pl  reply to my query as I am waiting for your reply to take further action. The reset command/instruction was placed in the first line in main () function but doesn't resolve even after commenting it . Thanks

  • Hello,

    Can you please clarify how you set the MICRO_INTERFACE_CONTROL register to 0x03? Was this done in your firmware? The microcontroller should not be placed in reset by the firmware as a default because it will just prevent the program from running. It can be used as a result of some kind of diagnostic check or a failsafe mode, but having it as part of the normal operation will just result in what you have described.

    In order to get access once more through the debugger you will have to reprogram the device through another means. You can compile your firmware and produce a .out or .hex file that you can load using the PGA970 GUI.

  • Dear Scott,

        I managed to resolve the issue but had to replace the SoC as it had got locked and was not accessible by my XDS110 debugger. Just fyi, the instruction I used was "MICRO_INTERFACE_CONTROL = 0x03;  /* Reset Micro */"  that I found in fram.c file. But now I have another issue where I am not able to access the SoC again after I tried communicating with the SoC on SPI after enabling it using "DIGITAL_Interface_Config(SPI_ENABLE | OWI_DISABLE |OWI_DEGLITCH_FILTER_1MS);" and "  MICRO_INTERFACE_CONTROL = 0x01;" instructions. I see that the processor is again not not accessible by my XDS110 debugger. I tried connecting by resetting the SoC mutiple times with power cycling but in vain. I see a similar issue faced in this forum by  but you are yet to provide an answer(e2e.ti.com/.../737287

    Pl find attached screenshot showing issues. Kindly provide/guide me in detail as to when to enable and disable the "IF_SEL" bit in the MICRO_INTERFACE_CONTROL register in order for the master to get access/read the COM_MCU_TO_DIF_B1 register. I have added the following code to the sample code provided by TI for EVM. Kindly advise urgently. It would be nice if you could provide a sample code or protocol for SPI as the PGA970 datasheet is very confusing and not user-friendly.

    interrupt void COMBUF_Handler(void)
    {
       COMBUF_Rx_Buf[0] = COM_DIF_TO_MCU_B; /* Read combuf data */
       COMBUF_RX_STATUS_CLEAR();            /* Clear receive status */
       COMBUF_Flag = 1;                     /* Flag to process rx data */
       MICRO_INTERFACE_CONTROL = 0x00;
       COM_MCU_TO_DIF_B1 = DAC_Value;
       COM_MCU_TO_DIF_B2 = 0x00;
       MICRO_INTERFACE_CONTROL = 0x01;
    }

  • Dear Scott,

       Could you pl reply to my query as we are waiting for same as we are stuck too.

  • Just to clarify, do you have the same error appear when you load the unmodified firmware? Or is this only when you have your custom version?