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.

ADC3662: ADC3662 internal NCO

Part Number: ADC3662

Dear TI,

We are using a ADC3662 in our design with complex downconversion. One thing not clear from the datasheet is which component, I or Q, is obtained from the sine and which one from the cosine of the local oscillator.

Thanks and best regards

Antonio

  • Hi Antonio,

    By default, the I component is obtained from the sine of the local oscillator, and the Q component is obtained from the cosine. This would result in input frequencies greater than the NCO frequency to appear as a negative frequency (to the left of 0Hz) in the FFT.

    You can change this by setting the MIX PHASE bit to 1 in register 0x25. This swaps to the I component to cosine and Q component to sine, resulting in an FFT where input frequencies greater than the NCO frequency appear as positive frequencies (to the right of 0Hz).

    Best,

    Luke Allen

  • Thanks for the answer.
    Does it means that the resulting signals are:

    I = sin(2*pi*acc/2^32)*ADC_value
    Q = cos(2*pi*acc/2^32)*ADC_value

    Where "acc" is the NCO accumulator, and the I & Q signals then enter the down-conversion blocks.

    Another information that is not clear to me is the DC gain of the filter block. We are using a by 32 down-conversion with 20bit output from the filter. If the input is a sin signal at the exact carrier frequency and phase with amplitude +/-100pkpk, I would expect the I output to have a DC level of 50 and the image frequency. After the filter, considering the DC gain and the fact that the ADC output is 16bit and the filter output is 20bit, what DC level can I expect?

    Thanks a lot for the help.

  • Hi Antonio,

    The resulting equations are as follows:

    Complex decimation results in a 6dB reduction in signal amplitude as the input is split into I and Q components, which each carry half the power of the original input signal. You can correct for this by adding digitial gain by setting the MIX GAIN A and MIX GAIN B register fields. There is a 3dB and a 6dB option.

    Best,

    Luke Allen