AD9683 - Two tone SFDR degrades as tone seperation gets closer in (10KHz). Thoughts?

We made an observation with regards to the two tone SFDR measurement (IMD3), where as the tone separation narrows from 3MHz down to 10KHz, the SFDR performance degrades. This is issue is more pronounce in the 2nd Nyquist zone.

Thoughts?

  • 0
    •  Analog Employees 
    on Jan 31, 2018 4:00 AM over 3 years ago

    Hello,

    Are you using our evaluation board or your own system board?   If you are using a balun, transformer...............its lower bandwidth may not extend down to these low frequencies and could be become a source of distortion.

  • The test tones are not at 10KHz, but the tone spacing is (e.g. 185MHz and 185.010MHz), so the resulting IMD3 are 10KHz apart (184.990MHz, 185.020MHz). Yes - we are using a driving the ADC with a differential BPF in the 2nd Nyquist where this issue is much more pronounce. In the 1st Nyquist, we have a low pass filter where the performance is greatly improved.

    We are in the process of repeating this test in the eval board, so I'll have results shortly. In the meantime any data you may have on this issue is greatly appreciate it.

  • 0
    •  Analog Employees 
    on Jan 31, 2018 10:54 PM over 3 years ago

    The AD9250 has an unbuffered input where the sampling capacitors are directly connected to its input.  This makes the ADC more sensitive to the source impedance presented to it.  Would it be possible to provide snapshot of the schematic that includes ADC and input driver+network.  Also, could you provide screen shot of FFT that you observe with 3 MHz and 10 KHz separation.  Thanks.

  • Hi

    Would it be possible on your end to duplicate this test? We are having some issues with the eval platform.

    Thanks

  • 0
    •  Analog Employees 
    on Feb 21, 2018 4:53 AM over 3 years ago

    Hello,

    I attached narrowband 2-tone data around IF=185 MHz with CLK=250 MSPS.  Two RF generators were combined via high isolation power combiner and then filtered with 200 MHz LPF which was then connected to the EVB.

    Data does show IMD does gets better as 2-tone separation increases.  My best guess is that it could be related to fact that a two-tone signal in the time domain exhibits AM modulation with modulation or beat frequency being equal to the delta difference between the tones.  Perhaps low frequency impedance on supply domain or internal ADC mode (i.e. reference buffers) could be issue.  

    Regards.

    attachFile.pptx