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.

DAC3482: Ouput flatness

Part Number: DAC3482
Other Parts Discussed in Thread: TRF3722, , TRF3705

Hi

I am giving a constant sweep data at dac input. I have a reconstruction filter at the output. I probed the DAC output at 50ohm termination resistor and found that the flatness is 4dB.

I want to know, whether I can probe DAC output as it is a CMOS current output?

Whether the bad flatness can be due to the reconstruction filter in case it is not 50ohm matched properly. Or may be some programming mistake?

My problem is since i hve not given proper connector at the exact dac output, i am left with the option of probing it to find out the output so that i can even tune the filter if needed.

Is there any other option?

Can you suggest me a a reconstruction filter for 160-195MHz band? The filter I have in my layout is:

In this i have got 4dB flatness by changing 391nH to 250nH.

  • Sarika,

    Are you measuring the output in a single-ended mode? If so, are you using a transformer? Is this a custom board or a TI DAC EVM? What frequency range are you sweeping? What is your DAC sample rate? We may already have this data for you.

    Regards,

    Jim

  • Hello Jim,

    This is not TI EVM. We have designed this board with DAC3482 differencial output given to TRF3722 through the reconstruction filter I was talking about. Since I have not given any connector to check the DAC output or the output of the reconstruction filter, I was forced to use an RF probe. So, I am probing each output out of the four output lines separately. The probed DAC output flatness for IP, IN, QP, QN separately as bad.
    The output DAC frequency is 168-193MHz (seeping in this band)and I need a reconstruction filter design for the same, keepin in mind that the card in hand has filter prototype as in the pic I shared earlier.

    Fdata = 156Mbps
    fDAC=624Mbps
  • Sarika,

    I am not a filter expert and do not know how much the filter could be effecting your measurement. See if this attached document helps you.

    Regards,

    Jim

    DAC Interface to Modulators.pdf

  • Hello Jim

    Thanks for the pdf. I want to know if there is a possibility of a bad flatness at the output of DAC due to programming error? Or I am getting it since I am probing each output separately?
  • Sarika,

    I think it is because the way you are probing. You may get a better response if you could use a high frequency differential probe with very low capacitance. The issue you are seeing we think is due to your filter. We have noticed this in the past if we used discrete components with large tolerances. This will effect the filter response. Another test is to bypass your filter, sweep the frequency of interest, and monitor the output power of your modulator, ignoring the spurs the filter would cancel if possible. This should at least verify you programmed the DAC properly.

    Regards,

    Jim    

  • Hello Jim

    As suggested I bypassed the filter and checked DAC output which is pretty flat. I have one last doubt. The reconstruction filter was designed with 100 Ohm differential load at the input or output (as 50ohms pull up resistor are placed both at dac output and trf input). Is this correct?
  • Sarika,

    It should be 100 Ohm diff, with 50 Ohm pull downs on each leg, not pull ups, on both ends of the filter. See attached schematics for example. One has a DAC connected to a TRF3705 and the other a TRF3722 modulator. Both cases are using 50 Ohm pull down resistors.

    Regards,

    Jim

    4721.TSW38J84EVM-SCH_B.pdfTSW30H84EVM-SCH_C.pdf

  • Jim

    Thanks a ton. With this input I am able to solve the flatness problem.