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.

DAC 3283 Power Up/Initialization

I have been using a TI DAC 3283 with a DACCLK of 250 MHz and a 2X interpolation to generate an IF output of 70MHz. I have a power up sequence and use the Frame signal once after all the SPI controls have been setup to synchronize the DACCLK and the DATACLK.   This design has been working well for over 2 years.

Now I am creating a new design that required a 500MHz DACCLK with 4X interpolation to create a 140MHz output.

I enable the second interpolating FIR in CONFIG1 (set it to 0x31).  This all works, but only 50% of the time.  The other 50% of the time, I have a bad output spectrum.   If I write 0x21 to CONFIG1 and then 0x31 (to disable and then re-enable the 4X interpolation), there is a 50/50 chance that the output will correct itself.  Re-issuing the frame signal to re-synch the system does not correct the problem.

Other than the bad output spectrum I don't have any other indications that there is anything wrong.  

Do you have any suggestions on how I might properly initialize the chip for this mode?  I am writing the registers starting with register 1 to register 1F (SPI is controlled from an FPGA - which incidentally also sets up the PLL chip that selects the DACCLK - DATACLK is derived from DACCLK to be 1/4 DACCLK).  I did notice that writing the SPI registers in different orders can make a difference.

Any help would be appreciated.

Urs

  • Urs,

    Typically, when the DAC is configured to run at a faster update rate (i.e. 250MHz to 500MHz) and then the output starts having problems, the power supply voltage, especially at DIGVDD, may have some problem. At faster update rate, the current consumption is higher. With some lossy power supply filtering network such as ferrite beads, the power supply drop may be below the recommended power supply voltage. Please double check the power supply voltage as close to the device pin as possible to ensure that the supply voltage to the DAC is still within the datasheet limit.

    I would also advise checking CONFIG7 for any alarms showing up when the problem occurs. Since the input data between the two designs remain at 125MSPS, I believe the input data timing should be good. If you can, please double check the data pattern test again to see if your new design still matches the input timing requirement.

    The following post may also be helpful to your issue:

    http://e2e.ti.com/support/data_converters/high_speed_data_converters/f/68/t/264424.aspx

    -Kang