I get confused about the effect of the "2xNRZ" mode of AD9162.
I'm checking the spectrum of AD9162 output using AD9162-FMCC-EBZ and ADS8.
The settings of the outline of AD9162 are follows:
- Variable IF : 1260.46MHz
- Carrier : 6 (1049.48, 1089.84, 1126.2, 1202.92, 1241.28, 1279.64)
- DAC Clock(FDAC) : 2770.56MHz
- Interpolation : 4
Figure-1 is an output spectrum using NRZ mode.
The center frequency is half of FDAC, 1388.28MHz.
The spectrum of the right side are alias of the left side 6 carriers.
Figure-1: NRZ Mode
Figure-2 is an output spectrum using "2x NRZ" mode(FIR85 enable).
I expected to "2x NRZ" mode the alias will be move to the right side of FDAC because the sample clock is twice of FDAC (Nyquist frequency moves to FDAC from half of FDAC).
But the alias does not move. It is only reduce the level. (About -23dBc reduced)
Figure-2: FIR85(2x NRZ) Mode
Q1. I understand the "2x NRZ" mode use DDR of DAC clock and the merit of this is to be able to use high Nyquist frequency. But figure-2 indicates the alias is still there.
Could you please explain why it remains the alias at the same frequency?
Q2. Comparing Figures 15 and 16 of the datasheet, the attenuation of aliases at 3 GHz is more than 40 dB.
But the alias in the above figure-2 attenuated only -23dB.
What could be the cause of this difference?
I corrected wording
[edited by: y_suzuki at 12:23 AM (GMT -5) on 24 Nov 2022]