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.

ADC34J23EVM: Unable to configure ADC & LMK04828 using ADC3000 GUI

Part Number: ADC34J23EVM
Other Parts Discussed in Thread: LMK04828, , ADC34J23, TPS62080

Tool/software:

Hello TI support,

I am using the ADC3000 GUI in order to program the ADC34J23 and the LMK04828 on the ADC34J23EVM. However, changing any setting in the GUI or loading any config file using "Low Level View -> Load Config" has no effect on the hardware: neither the LMK04828 chip nor the ADC. The strange thing is that the GUI window shows "Operation successful" when a config file is loaded, but the LED D4 does not switch on, nor does the change in setting take effect (verified by measuring the clock output using an oscilloscope).

I have verified the hardware jumper settings:

The current consumption of the EVM stays constant throughout with 468 mA at 5V DC.

Could you help me debug what could be holding the chips back from being programmed?

I look forward to hearing from you.

Sukhdeep Oberoi

  • Hi Sukhdeep,

    Are you using the on board LMK setup? Or are you applying an external sampling clock?

    What data capture board are you using? TSW14J56?

    Regards,

    Rob

  • Hi Rob,

    • I intend to use the on-board LMK04828 to generate the ADC sampling clock (LMK DCLKOUT0), FPGA GBT reference clock (DCLKOUT2) & the FPGA JESD core clock (DCLKOUT8), and have set the jumpers as instructed in "3.5. Test Set-up Connection (Onboard LMK04828 Clock)" of the ADC3xJxxEVM User's Guide (SLAU579D).
    • I do not have the TSW14J56 data capture board, but I am using a Xilinx Zynq XC7Z030SBG485 SoC with the TI-JESD204 IP as the Rx device.

    Kindly let me know if you require any further details about the setup.

    Regards,

    Sukhdeep

  • Hi Sukhdeep,

    Thank you for the details.

    Can you please probe the clock inputs with an oscope to make sure you are getting the sampling frequency that you expect?

    Thanks,

    Rob

  • Hi Rob,

    I have probed the LMK clock outputs, and I do not get the desired sampling frequency, which is the issue: no matter how the chips are configured in the ADC3000 GUI, the settings do not take effect on hardware. I have tried loading the preset configurations from the 'Configuration Files' folder as well, but all to no avail.

    Regards,

    Sukhdeep

  • OK thank you for confirming.

    Might be good to troubleshoot thru the schematics.

    Is the 100MHz oscillator on board and powered up?

    If so, is there a clock signal on the LMK at pin 44?

    Also, so make sure the GUI is actually talking to the ADC, can you verify the ADC can be powered down?

    The current should change on your lab bench power supply.

    Please let me know.

    Thanks,

    Rob

  • Hi Rob,

    Thank you for the detailed troubleshooting steps.

    1. The 100 MHz oscillator is on board and powered up. I have measured on pin 44 of the LMK (Oscin_n) and confirm that the clock input is present:

    2. Using the GUI, I performed a global power down of the board, and this does not change the current consumption at all (still at 468 mA), although the checkbox gets selected and the ADC register 0x15 read operation gives value 0xFF.

    3. Also, as a simple test, I perform a 'Read All' operation in the low-level view after restarting the GUI and power cycling the ADC board, and the GUI returns 0xFF values for all registers of the ADC and the LMK. After this, I am not able to change any setting or write directly to any register address in the GUI.

    Since it is evident that the GUI is unable to communicate with the ADC and LMK, could you suggest any additional debugging steps? I am using the ADC3000 GUI version 1.2 (build date: 02/23/2022). Is there a later version of the GUI available?

    Regards,

    Sukhdeep

  • Hi Sukhdeep,

    If you have downloaded the GUI from the product webpage, then this is the latest GUI version. I am assuming you have done that, correct?

    Next, I would directly connect the USB cable from the PC or laptop to the EVM to make sure it is talking to the EVM. Please do that and not use a USB hub.

    I would also make sure there is enough current to the EVM from the power supply. I would set the EVM current to 3A at 5V.

    Lastly, check all the power supply to make sure you have the right voltages on each power domain.

    Let me know what you find.

    Regards,

    Rob

  • Hi Rob,

    1. Yes I have downloaded the GUI from the product webpage.
    2. I have already tried connecting the USB cable directly from my laptop to the EVM, unfortunately that does not change the result.
    3. As mentioned, no matter what the current limit is set in the power supply, the current consumption of the board remains 468 mA, even when a global power down of the board is performed using the GUI.
    4. Lastly, I checked the voltages on inputs and outputs of the LDO regulators, and did not find any critical deviation (measured voltage marked in orange):

    Regards,

    Sukhdeep

  • Hi Rob,

    I would like to add that the PG pins of the two TPS62080 Buck converters (U8 & U11) are at logic level 0. Does this point to power integrity issues on the board?
    I look forward to hearing from you.

    Regards,

    Sukhdeep

  • Hi Sukhdeep,

    Thank you for all the details and for trying a few things out.

    Let me set this EVM up in our lab and do some testing to see what the issue maybe.

    Please give me a few days to respond back.

    Regards,

    Rob

  • Hi Sukhdeep,

    I assume you are going to the Common tab on the GUI and checking the global power down checkbox?

    Please advise.

    I had some time to set this up quickly and just installed the GUI and the USB connection works fine for me.

    When you hit the global power down pin, you should see the current on the EVM go down by 100mA or so.

    Thanks,

    Rob

  • Hi Rob,

    Yes, as mentioned in earlier exchanges, I go on the Common Tab in the GUI and check the Global Power Down checkbox.

    However, the current consumption does not go down even a bit, pointing out that the GUI is not able to correctly talk to the hardware. What’s strange is, after every write operation in the GUI, “Operation Successful’ is displayed on lower left side of the GUI window. 
    I probed the SCLK and SDA outputs of the FT245R on board and it appears to be working fine (I have not verified the actual serial data being sent on SDA using a LA). 
    The next piece of hardware in the way of the signal is the CPLD. Is there any quick method to test if the CPLD is working correctly?

    Regards,

    Sukhdeep

  • Hi Sukhdeep,

    Can you send me a picture of your board? I want to see if any jumpers are out to place that might cause this.

    Thanks,

    Rob

  • Hi Rob,

    here’s a picture of the board:

    Regards,

    Sukhdeep

  • Hi Sukhdeep,

    From what I can tell this looks fine.

    I can send you another EVM to try to see if this solves the issue.

    If you are open to that, please let me know and I will contact you offline.

    Regards,

    Rob

  • Hi Rob,

    Thanks for your response. Yes, we would really like to try with another EVM to see if it solves the issue. 

    Regards,

    Sukhdeep

  • Hi Sukhdeep,

    Okay, sounds good. I will close this post and please be on the lookout for my email.

    Regards,

    Rob