Ad9082 interleaving calibration

There’s fs/4 spurious artifacts in the frequency domain when I use ad9082, which probably caused by the offset mismatches of four sub adcs. And I found the same thing when I use ad9082-fmca-Ebz. I’ve tried to use background calibration and to set appropriate nyquist zone that is mentioned in the user guide, but it doesn’t work. So I wanna know if there’s any other methods to calibrate.

  • 0
    •  Analog Employees 
    on Jul 6, 2021 2:42 PM

    Hello,

    The residual fs/4 spurious artifacts are actually due to coupling of digital clock noise operating at Fs/4  as opposed to the AD9082 interleaved ADC core.  The ADC core operates in a interleaved random mode which breaks up any residual offset/gain/timing mismatch between sub-ADC's hence is not the source of the fs/4 "image" spurs.  To minimize this spurious, follow the recommended decoupling strategy in the UG as well as what is done on the EVB.

  • Hello, Thanks for your reply. I’d like to follow the recommended decoupling strategy in the UG to minimize the spurious on the EVB first, and noticed that the decoupling notes are all about the way the decoupling capacitors install(DNI). So does it mean that I have to adjust the position of these decoupling capacitors if I want to reduce the spurious? Or to make matters more convenient, any other option instead of adjusting the placement?