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.

DAC39J84: Eye Scan Software MPSSE Configuration

Part Number: DAC39J84
Other Parts Discussed in Thread: TXB0106, TXS0108E, SN74AVC4T774

Hi

I am attempting to use the TI Eye Scan GUI on custom hardware that is running the DAC39J84. I am using the same FT2232H FTDI part that is on the DAC3XJ8XEVM board and have configured it in the same way. I am able to see all JTAG signals (TCLK/TMS/TDI/TRSTB) that is output from the FT2232H, however I cannot read back on the TDO line as the TDO pin on the FT2232H appears to be configured in software as an output (or perhaps bidirectional input/output).

Is it possible for you to provide us with the LabView source .vi file, or any settings that is used in the Eye Scan GUI for configuring the FTDI part so I can try to debug what is wrong.

Thanks

  • Jordan,

    The issue is the DAC output signal TDO cannot drive the FTDI device. You will need to add a buffer between the two parts.

    Regards,

     

  • Hi Jim,

    Yes I am buffering the TDO signal with a level translator (TXB0106 part, have tried a simple FET as well) to provide the necessary 1.8V to 3.3V translation. The TDO output of the buffer appears as clean square pulses as you would expect when the FTDI part is not connected, with a swing of approximately 0 to 3.3V. When the TDO output of the buffer is connected to the FTDI pin, the swing observed reduces to about 0V to 1V, which leads me to believe the FTDI chip is attempting to drive this line low (as an output).

    Interestingly, even without the DAC in the equation, when I attempt to short the TDI (output from FTDI) and TDO (input to FTDI), a similar drop from ~3.3V to ~1V occurs. I have attempted to change the 'Read Pin' in the Eye Scan GUI, although I'm not entirely clear on what that changes. The GUI appears to configure the TDI, TMS, TCLK, TRSTB signal outputs all correctly, it is just the TDO input that I can't get to work.

    Thanks for your help

    Regards

    Jordan

  • Jordan,

    The TI EVM has a TXS0108E part which is similar to the TXB0106. These do not have enough drive current for this. We had to bypass this translator and send the signal to an external buffer (SN74AVC4T774) to overcome this issue. Maybe you can try this.

    Regards,

    Jim 

  • Eyescan_source_code.zipJordan,

    Source code attached.

    Regards,

    Jim

  • Hi Jim,

    Appreciate your help, and for providing the source code.

    My assumption was wrong about how the FTDI part is configured - it appears to be using the synchronous bit-bang mode rather than MPSSE mode. From what I can tell, the default value for the 'Bit Mode Mask' input for the FTDI configuration is 0xFF which sets all pins as outputs, including TDO.

    It's still possible to read the state of an output pin and I am indeed able to see the TDO input triggering in software when I attempt to overdrive the TDO with a higher output-current buffer source. The amplitude is still reduced to only 0 to 1V and I am not convinced the software is successfully registering all bit transitions.

    I am missing many of the dependencies for the Labview project so I cannot recompile the Eye Scan GUI with a different value for the Bit Mode Mask input. 

    Regards

    Jordan

  • Hi Jim

    I am now able to recompile the Eye Scan project from scratch. I will try to modify the Bit Mode Mask variable from 0xFF to 0xFB (sets ADBUS2, or TDO pin, to an input) to see if that helps. It seems from initial testing to help but I will not be able to test full read capability until next week. If I find anything further I will add to this thread.

    Regards

    Jordan

  • Jordan,

    That is good news. If you have anymore issues, just open another post.

    Regards,

    Jim

  • Hi Jim,

    Yes I can confirm, recompiling the LabView project has worked. The TDO pin now swings the full 0-3.3V using the TXB0106 level translator to translate from 1.8V to 3.3V. The data coming out of the software looks correct now.

    Thanks for your help.

    For anyone else that faces the same issue, the missing dependencies from the source code attached earlier are

    • FT_Open_Device_By_Description.vi
    • FT_Reset_Device.vi
    • FT_Set_Baud_Rate.vi
    • FT_Set_Bit_Mode.vi
    • FT_Set_Timeouts.vi

    I have attached a zip folder with these files.

    Jordan

    3438.EyeScan Dependencies.zip